Wikipedia talk:Verifiability

I've noticed that many right-wing sources on this site are considered unreliable; my question is: why? I'm politically neutral and, considering that the politics of Italy and the United States are different, I have asked myself this question. Thanks in advance. JacktheBrown (talk) 22:48, 28 September 2024 (UTC)Reply

That's not a question with a simple answer, but whether a source is perceived as right wing or left wing shouldn't be a consideration in questions of reliability. If you look into discussions about sources the issue are rarely ideological. Instead the cause is that media organisation have diverged since the days of print media, and that divergence has impacted media on different parts of the political spectrum differently. That change in the real world has then had an impact on Wikipedia. -- LCU ActivelyDisinterested «@» °∆t° 10:47, 29 September 2024 (UTC)Reply

@ActivelyDisinterested: the thing that seems strange to me is that extreme left-wing sources aren't banned (maybe I'm wrong), while extreme right-wing sources are.
I repeat: I don't know the political situation in the United States in detail. In Italy we have a right-wing government; it's a government that many Italians support, because Italy is quite conservative, which for the Italian context is a very good thing; for the US context, however, it seems not, since the American right has, for example, denied climate change, which unfortunately exists.

Returning to the main topic: the thing that seems strange to me is that extreme left-wing sources aren't banned (maybe I'm wrong), while extreme right-wing sources are. JacktheBrown (talk) 14:11, 29 September 2024 (UTC)Reply

extreme left-wing sources aren't banned (maybe I'm wrong) yes you age +are+ wrong. Take for example in British news media The Canary and Skwarkbox are both considered unreliable, and are on the far left of politics. Also to be clear sources aren't banned so much as actively discouraged if the consensus is that they are unreliable.

Whether governments or voters are of a particular part of the political spectrum is also not a consideration in judging the reliability of sources. -- LCU ActivelyDisinterested «@» °∆t° 14:21, 29 September 2024 (UTC)Reply

@ActivelyDisinterested: thank you very much for this detailed and very useful explanation. JacktheBrown (talk) 14:23, 29 September 2024 (UTC)Reply

The primary reason media sources get Deprecated (which is not quite a “ban”), is that they have repeatedly been shown to not fact check (or, in some cases, shown to completely invent “facts” which they report).

As to why right leaning sources seem to be more likely to be deprecated than left leaning ones… this is due to the fact that right wing sources tend to get more scrutiny. It’s no secret that Wikipedia attracts academics, who tend to be a somewhat left leaning group. They notice (and complain) when right-wing sources don’t fact check… and they tend to be a bit more forgiving when left-wing sources do the same.

That being said, we have deprecated a few left-wing sources when enough evidence has been presented to show they are not properly fact checking. It’s difficult, but possible. Blueboar (talk) 15:04, 29 September 2024 (UTC)Reply

@Blueboar: Recognizing and admitting this problem is certainly a big step forward, but actions are more important than words; I would like neutrality not to be compromised in the encyclopedia. JacktheBrown (talk) 20:12, 12 October 2024 (UTC)Reply

User:JacktheBrown - Outside of politics the banning also happens. I think it is just the mechanics or effect of WP:RSP. For the UK it seems more an elitism thing than a political leaning -- for example it seems most British press by volume is excluded as low class. RSP seems to interpret 'reliable' to mean 'respected' or 'truth' to that WP editor, excluding consideration of 'available' or 'accurate' from 'verifiability'. (You can elsewhere see discussions on a paywalled source or remote paper being preferred despite readers generally not having access to such or WP:VNT) The terms RSP uses are "Blacklisted" meaning mechanically edits including that site are blocked, or "Deprecated" meaning the guidance includes "generally prohibited" by automatic warning of such an editor and removal of such edits by third parties, or "Generally unreliable" meaning outside "exceptional circumstances" not to be used with removal by third parties and pings in TALK. If you want to see what discussions on what to ban are like, they are mixed in at WP:RSN. Cheers Markbassett (talk) 04:22, 13 October 2024 (UTC)Reply

@Markbassett: I'm not convinced that this isn't political. What's the percentage between left-wing and right-wing sources that can be used? Perhaps an 80/20? JacktheBrown (talk) 18:43, 12 November 2024 (UTC)Reply

I'm saying that banning sources via blacklisting and deprecated designations seems basically an elitism act, and that it exists in all venues or topics - so it's in effect at articles for sports or music or basically anything. I'm not talking politics, or whether some parties might exploit it for financial or personal reasons -- I'm saying it simply is always a direct issue to the policies for V and WEIGHT. You cannot have WEIGHT properly measured if you exclude considering any significant circulation because editors think those are labelled as lesser venues, and if you are selecting paywalled or tiny elite items as the one to cite, then it's preferring the not accessible ones which is contrary to it being V verifiable. Cheers Markbassett (talk) 22:25, 12 November 2024 (UTC)Reply

@Markbassett: very interesting discussion. So is Wikipedia making a mistake in rejecting certain sources or is it doing so in good faith? In my opinion the second option is correct, but I would like to read yours. JacktheBrown (talk) 22:40, 12 November 2024 (UTC)Reply

Well I tend to view RSP usage here as problematic, but my wider view is Wikipedia content is from editors that are just people - good people and bad people, informed one way among many, obsessive or ditzy, with good days and bad days, days they have time to be careful and days that are rushed. You can look over RS debates and RSP debates or just filter recent changes to the ones possibly vandalism and judge for yourself.

For RSP in particular, I was not a fan of the idea back when it started and events have not improved my views of it and the way mechanics of it work. It supposedly was meant to capture the RS conclusions from earlier "Perennial" RS questions, so it could help the RS consideration. But folks just propose and argue from whatever stance for banning - there is not necessarily prior RS considerations or policy criteria in play. And instead of informing a RS consideration it seems mainly a blanket forever judgement - or at least I have not seen any reversals up or down. My revision of your 80/20 question back to you would be more along the lines of if WP bans 80% of the UK press, then is it a 20% valid portrayal of views ? Cheers Markbassett (talk) 23:25, 12 November 2024 (UTC)Reply

@Markbassett: is it possible to have an accurate statistic of how many left-wing and right-wing sources are deprecated? JacktheBrown (talk) 00:17, 13 November 2024 (UTC)Reply

I think one can only get a statistic of the simple sort and count by their status like 20 banned, 44 deprecated, 133 generally unreliable, 100 no consensus, 119 generally reliable. Anything else would need a chosen filtering and categorizing. But I think RSP counts is less a concern than percentage of external, or individual articles being totally dependent on limited source.

The first concern is more how large a percentage of the total WP:WEIGHT of coverage for a POV is being excluded by RSP, because it seems the largest or most known sources are the ones to get excluded, so a simple count such as '5' banned needs the context of is that 5 of only 6 or is it 5 of 100, and does the 5 constitute 95% of the WEIGHT or what ? For example again, almost all of major circulation UK newspapers are banned, and the banning for reason for banning for being 'sensationalist' or 'just sports' then means not having the best available data or best known portrayals on topics that were UK scandals or UK sports events.

The second concern is that at some articles the cites are too limited by selections which perhaps excludes what the most common view is outside of WP, or at least will not show all the common views so be a failure of NPOV. Again, this is all categories -- for example one cannot have a music article that only shows the academic views and think that reflects the world opinions and complete story. Cheers Markbassett (talk) 21:53, 13 November 2024 (UTC)Reply

@Markbassett: Wikipedia:Reliable sources/Perennial sources shouldn't have been created, perhaps Wikipedia is unnecessarily self-complicating. JacktheBrown (talk) 22:17, 13 November 2024 (UTC)Reply

I don't think it's possible to get such a count, because you'd have to first agree on where to draw the line between left and right. For example, CNN is most commonly described as centrist, but some editors think it's leftist. The New York Times seems to be considered center-left, except that there are complaints that it's right-wing on trans issues. WhatamIdoing (talk) 00:33, 14 November 2024 (UTC)Reply

We've discussed this in the past and I'm wondering what the current thinking is on ONUS as a reason to remove something without any other reason, even long-standing, multi-years long content that once had a consensus to remain presumably. Is it enough for a few editors to remove something then claim ONUS? What about if an RFC is started? Usually, an RFC means that the status quo remains for the duration, and WP:NOCON means the status quo remains. Does ONUS still have the ability though to just be a blank check to remove anything at all that a few editors don't like? Andre🚐 19:33, 28 October 2024 (UTC)Reply

I've always supported ONUS but I don't believe, or have ever believed, that the wording of ONUS can be used as a reason for removing content. If content is removed for a valid reason and is restored, then ONUS is a reason it shouldn't have been restored until there is consensus for it. It's shouldn't be the reason that the content was removed in the first place. -- LCU ActivelyDisinterested «@» °∆t° 19:55, 28 October 2024 (UTC)Reply

Right, but what about a situation where content has a consensus at a past time, or was added and stood for years at a reasonbly well-watched article, then a few editors come along and create a no-consensus situation due to either inactivity of the editors who originally added/supported that or simply attrition/changing perspectives and userbases, and use ONUS as a justification for "nocon -> remove or change" as opposed to "nocon -> status quo"? If you catch my drift. Andre🚐 19:57, 28 October 2024 (UTC)Reply

I guess my feeling is that if someone is going to claim ONUS as a reason for removing content, they'd better have an underlying explanation ready, and it would probably be best to provide some insight into that reason in the edit summary rather than simply claiming ONUS. I also think unless it's linked, newcomers may very well have no idea what "ONUS" as an edit summary is even referencing. "Better before" isn't much better than "ONUS", but at least it gives a bit more of an inkling as to the reverting editor's mindset.

I'm not readily aware of any situations like the one you've described, though I can't rule out the possibility...however, consensus can change. If editors who came to Consensus A have all gone inactive, and a new number of editors come along six months later with no knowledge of the prior consensus and achieve Consensus B without the prior set of users engaging with them, I'd argue that that's an indication that consensus may indeed have changed. I'm not quite sure I'm addressing the scenario you're describing, though. DonIago (talk) 20:03, 28 October 2024 (UTC)Reply

Yes, but we're talking about a scenario where there's no consensus whether consensus has changed. Therefore, usually, WP:NOCON means status quo. Andre🚐 20:05, 28 October 2024 (UTC)Reply

Isn't that why dispute resolution processes exist? DonIago (talk) 20:07, 28 October 2024 (UTC)Reply

I mean, sure. Although that's a very different discussion. I'd say my recent experience with dispute resolution has left me with some ideas for improvement, as a former mediator with the defunct MEDCOM. Which wasn't always the best either. But you still haven't really answered my question. Which is OK, it's OK to answer my question with another question. I will then answer that with another scenario: What if we went through dispute resolution and the outcome was to have an RFC (which is an outcome), and we're getting NOCON again? Equally strong camps, equally strong arguments (as in the WAID style thought experiment) Andre🚐 20:10, 28 October 2024 (UTC)Reply

There was an RFC on that exact thought experiment, it was never closed as there was no consensus. -- LCU ActivelyDisinterested «@» °∆t° 20:16, 28 October 2024 (UTC)Reply

Yep. Exactly. Which means further discussion and here we are. Andre🚐 20:19, 28 October 2024 (UTC)Reply

Yes and all this has been said before. "It depends on the specifics" is the best there is. Editors have been blocked for running around quoting ONUS removing content and being disruptive, and of course the same is true on NOCON and stokewalling. Being open to others ideas and viewpoints while working to find a solution is the best you can do. If editors are displaying behavioural issues and being disruptive there are other means to resolve the issue. -- LCU ActivelyDisinterested «@» °∆t° 20:24, 28 October 2024 (UTC)Reply

At some point if editors who prefer a difference consensus aren't getting their desired result even after going through an RfC, maybe they need to accept that there simply isn't a consensus for their preferred version, and perhaps try again in six months or such? Anyone who's spent significant time on Wikipedia likely has experience in losing an argument and hopefully trying to be graceful in defeat. DonIago (talk) 20:26, 28 October 2024 (UTC)Reply

I have always thought that being long standing on its own barely counts as a consensus, it's certainly very weak. Consensus shouldn't stand forever, the project will become moribund.

The real issue is that both ways have issues, some editors misuse ONUS to remove content just because they dislike it and at the same time other editors misuse NOCON as a way to stonewall. -- LCU ActivelyDisinterested «@» °∆t° 20:05, 28 October 2024 (UTC)Reply

I agree that NOCON is a form of stonewalling if abused, but what about a legitimately stipulated situation where something was added on a well-watched article, had a consensus at one point, and then a NOCON exists today? Does ONUS therefore allow removing anything that can be fought to a NOCON? As opposed to the usual situation as I understand it that a NOCON means status quo, except perhaps in extreme situations such as a BLP UNSOURCED where policy is clearly on one side. I'm talking about an even content dispute NOCON which leads to removal through ONUS. Is that legit? Andre🚐 20:07, 28 October 2024 (UTC)Reply

Maybe? That's the best answer there is. It's not something that can be given a satisfactory answer in a general sense, only in a specific case by case basis. Discuss it, discuss it some more, discuss it at a venue with more with a bigger audience, have a formal discussion, all the normal dispute resolution processes. -- LCU ActivelyDisinterested «@» °∆t° 20:11, 28 October 2024 (UTC)Reply

“Consensus can change”… so a prior consensus can be overturned by a newer consensus. If there is currently a consensus to omit or remove material from an article, then that material should be omitted or removed… even if there used to be a consensus to include that material.

Of course, whether consensus has in fact changed (or not) can sometimes be a matter of dispute. If so, use the normal dispute resolution steps (seek 3rd party opinions, file an RFC, etc) to resolve the issue. Blueboar (talk) 20:11, 28 October 2024 (UTC)Reply

and when/if the RFC ends with no consensus being clear to omit or remove the material, does ONUS default to removing it? Or does NOCON default to the status quo? Andre🚐 20:15, 28 October 2024 (UTC)Reply

It's starting to feel as though you're asking for a global remedy to a problem that should be handled locally if and when it arises. DonIago (talk) 20:58, 28 October 2024 (UTC)Reply

I'm not asking for any specific remedy. If the answer remains unclear then it's unclear. Andre🚐 21:41, 28 October 2024 (UTC)Reply

Technically, NOCON doesn't "default" to anything, as it's not a rule. It is a claim that "the common result is" to retain the STATUSQUO. Perhaps we should make that clearer, e.g., When discussions of proposals to add, modify, or remove material in articles end without consensus, the common result – but not a result that is required by any policy, including this one – is to retain the version of the article as it was prior to the proposal or bold edit. WhatamIdoing (talk) 23:07, 29 October 2024 (UTC)Reply

That would resolve the conflict by making ONUS supercede NOCON. Not my preference but a valid resolution of the tension. However, I would argue that this is an ambiguity in the current wording. I would read the current wording as defining the "common result" as carrying the force of policy/guideline (like all guidelines, not always observed strictly) and then there are 3 bullet points defining the bounded, but not exhaustive, exceptions to the "common" result. Certainly, while this may not be an accurate interpretation of the policy, it's a common one. Andre🚐 23:20, 29 October 2024 (UTC)Reply

Well, I didn't write that sentence originally, so I can't tell you what the original intention was for the wording choice, but based on what I know of the now-blocked editor, he would be very happy to have you interpret it as requiring inclusion, or at least STATUSQUO.

When I write that something is the "common" choice, I never mean that it is "carrying the force of policy/guideline"; I mean what it says, i.e., it is a statement of fact about the frequency of an event. For example, when I wrote the paragraph in MOS:FNNR about which section heading to use for ref lists, it was based on a survey of a random set of articles. It strictly says which section headings are most common, as a point of objective fact, without telling editors that they should prefer the most common. We did expect that editors would use this information to voluntarily choose the most common section heading (and they have), but it is not "a rule" and it is not a matter of "force". WhatamIdoing (talk) 23:36, 29 October 2024 (UTC)Reply

My guess is that given how old this part is, it was a case of someone writing something down that was observed before its incantation was transcribed. In keeping with the section in policy wherever it is - I know you know the one - that says that all of the policies and guidelines are merely description of a consensus and not the thing in itself. AFAIK, it's always been the case in RFCs and VFDs/AFDs that nocon meant keep. Whereas the ONUS/BURDEN supporting remove thing seems much more recent. Without having exhumed the remains myself to confirm that. There also might be an alternate older wording of this concept. Maybe in the deletion or RFC policy if not here. Andre🚐 23:43, 29 October 2024 (UTC)Reply

The NOCON expansion to prefer QUO was discussed repeatedly before the addition was made, and the editor knew that there were concerns (from me and others) about it.

The specific sentence in ONUS was added two years later, but the edit summary when he updated the shortcut (originally created in 2008 and pointing to the same section as BURDEN) suggests that this was not intended to be "new" content. WhatamIdoing (talk) 00:04, 30 October 2024 (UTC)Reply

That version from 2012 is clearer and more decisive than the current language: In discussions of textual additions or editorial alterations, a lack of consensus results in no change in the article. No "common result" there. It may have been watered down later. But I'm sure that nocon meant keep/status quo even in probably 2005. Here's an essay from 2007. Wikipedia:What "no consensus" means Andre🚐 00:10, 30 October 2024 (UTC)Reply

That essay doesn't address article changes, and on the day it was written, the main point appears to have been: Often, people feel that no consensus should mean that the current status quo prevails. That is not, however, always the case. WhatamIdoing (talk) 00:48, 30 October 2024 (UTC)Reply

I agree, but, that supports the idea that this was the generally practiced procedure at the time. Andre🚐 00:59, 30 October 2024 (UTC)Reply

Maybe. Or maybe that back then we had problems with oversimplification of rules. Or maybe that back then we had problems with wikilawyers claiming that whatever asserted rule supports my preferred outcome is the One True™ Rule. WhatamIdoing (talk) 04:29, 30 October 2024 (UTC)Reply

I see. Well, that's not how I remember it. Andre🚐 05:11, 30 October 2024 (UTC)Reply

I found you, WAID, in 2016 removing some status quo from an old version of BRD. Andre🚐 01:17, 30 October 2024 (UTC)Reply

I have made many edits to BRD over the years. In general, my earlier ones took BRD away from its original concept, which was less about WP:EPTALK and more about what to do when achieving consensus has already proven difficult. WhatamIdoing (talk) 04:27, 30 October 2024 (UTC)Reply

(Hopefully useful links) Wikipedia talk:Verifiability/Archive 74 and Wikipedia talk:Verifiability/Archive 75 host the 2022 debates about ONUS. Wikipedia talk:Verifiability/Archive 77 is the long discussion on BURDEN. ONUS comes up again on Wikipedia talk:Verifiability/Archive 80. All very lengthy discussions, probably helpful to read before reviving the arguments. Schazjmd (talk) 20:35, 28 October 2024 (UTC)Reply

yes, and my name appears in Archive 77 and Archive 80 so I participated. Those archives still don't provide a resolution to the question. Nor does Wikipedia:Requests for comment/When there is no consensus either way Andre🚐 20:43, 28 October 2024 (UTC)Reply

The links are for everyone. Other people are participating here and might not have read those discussions. Schazjmd (talk) 20:45, 28 October 2024 (UTC)Reply

That's true. Thanks. Andre🚐 20:48, 28 October 2024 (UTC)Reply

ONUS has many problems including conflicting with other policies and processes, being out of place (what the heck is a ham-handed arbitrary finger on the scale towards exclusion doing in wp:verifiability?). Also the commonly quoted word "ONUS" is not even in the policy. The original intent was to prevent somebody from using meeting wp:verifiability as an argument for or a way to coerce inclusion. We should fix the whole mess and turn it into a big plus by removing it and substituting "WP:Verifiability is a requirement for inclusion, not a reason for inclusion"' which was the original intent of the wording linked by ONUS. That change would also be neutral regarding inclusion/exclusion because it makes slight balancing changes in either direction. Sincerely,North8000 (talk) 21:41, 28 October 2024 (UTC)Reply

How many times we going to have this discussion ) Selfstudier (talk) 21:47, 28 October 2024 (UTC)Reply

Apparently we need to have it since editors still cite ONUS to remove things. I agree with North8000 that this would solve the problem in a fair way. Andre🚐 21:50, 28 October 2024 (UTC)Reply

We have problems with NOCON, let's remove them both and let the talk page decide through discussion. -- LCU ActivelyDisinterested «@» °∆t° 22:24, 28 October 2024 (UTC)Reply

Something to consider… if there WAS a consensus to include, but is now at NO CONSENSUS, then the consensus has changed… the pendulum is swinging towards not including. So, my “tie breaker” is to omit… and ask again in a year to see whether the pendulum has continued to shift towards omit… or has swung back towards include. Blueboar (talk) 21:53, 28 October 2024 (UTC)Reply

I don't think I agree with that, Blueboar. If there is no consensus to remove material, WP:NOCON would retain the status quo. At least, that's how it's normally worked that I've observed Andre🚐 21:58, 28 October 2024 (UTC)Reply

In principle but that doesn't necessarily apply in all cases and I notice you have studiously avoided being specific. Selfstudier (talk) 22:57, 28 October 2024 (UTC)Reply

Andrevan… Is there “no consensus” to remove the material, or is there “no consensus” to retain the material? Answer: it depends. Ultimately you have to look at how the RFC question was worded, read the comments people made in reply, and figure that out on a case by case basis. Yet another good example of why RFCs are not simply !votes. Blueboar (talk) 12:45, 29 October 2024 (UTC)Reply

That's one of the problematic conflicts of wp:ONUS with the wp:consensus policy. If there is no consensus to remove or reaffirm some long standing material, wp:consensus says that it stays in and wp:Onus says that it comes out. North8000 (talk) 18:54, 29 October 2024 (UTC)Reply

Again… it depends. If the question asked was “Should Xxx be retained?” and there is no consensus to retain, then I would take that lack of consensus as a sign we should remove. Alternatively, if the question asked was “Should Xxx be removed?” and there is no consensus to remove, then I would take that lack of consensus as an indication that we should retain. Blueboar (talk) 19:17, 29 October 2024 (UTC)Reply

Those are the same question, though. Should it be retained or removed - the only differing factor is whether it's there now and is stable and long-standing, or whether it was recently added and its addition is contested. Andre🚐 19:31, 29 October 2024 (UTC)Reply

It's not only that, there is the question of WP:CONLEVEL to consider. Selfstudier (talk) 19:41, 29 October 2024 (UTC)Reply

Yup… which is yet another reason why we can’t have a simple one-size-fits-all “default”. We have to look at the actual discussion. How many editors commented? Did they focus on reasons to include (which failed to gain consensus), or did they focus on reasons to omit (which failed to gain consensus)? How we proceed will depend on all of these nuances. There is no single correct answer. Blueboar (talk) 19:59, 29 October 2024 (UTC)Reply

I don't see the relevance of CONLEVEL. Whether to include or exclude a given bit of material in a specific article has nothing to do with a small group of editors trying to "override community consensus on a wider scale". CONLEVEL does not refer to small discussions. CONLEVEL is about small groups of editors declaring that "their" articles are exempt from site-wide policies. WhatamIdoing (talk) 20:40, 29 October 2024 (UTC)Reply

In my post that started this subthread, (That's one of the problematic conflicts of wp:ONUS with the wp:consensus policy. If there is no consensus to remove or reaffirm some long standing material, wp:consensus says that it stays in and wp:Onus says that it comes out.) I was intending to imply that both questions (remove and keep) got asked and neither attained a consensus. Then ONUS dictates the opposite of wp:nocon. North8000 (talk) 21:25, 29 October 2024 (UTC)Reply

Four years ago, you and I talked about moving ONUS to NOCON, so the contradiction could be side-by-side in the same page.

Recently, there's been talk about moving NOCON out of CON. That seems to have stalled, and not just because a couple of editors were worried about my suggestion that it be moved to a page that is more relevant but doesn't currently say "policy" at the top. WhatamIdoing (talk) 21:33, 29 October 2024 (UTC)Reply

It's been many months since I did this but I went back and analyzed the roots. IMO there was a tacit local agreement with the inclusion (including an edit summary the gist of which is what I said above....to prevent using using verifiably as a way to coerce inclusion) but no big discussion. And certainly NO agreement with the way that it is being used today as I described above. IMO it's a change that would fix many problems and conflicts, aligns with the intent of why it was included in the first place, does not go against any previous large discussion decision, and would be pretty neutral in the inclusion/exclusion equation. Sincerely, North8000 (talk) 23:21, 28 October 2024 (UTC)Reply

While I lean toward North8000, what I'm gathering so far from the discussion is that there still isn't a consensus and it depends. Which is fine. That's how things were last time I checked in. NOCON/PRESERVE is a policy and so is BURDEN/ONUS. However, I do think that last time we discussed it, I think it was ActivelyDisinterested who proposed adding something saying that ONUS can't stand alone as a rationale or when paired with other otherwise invalid rationales, it doesn't add anything. Or clarifying that ONUS specifically doesn't mean that you can remove some long-standing content, claim ONUS if there isn't immediately produced a new consensus to retain it, and have policy justifying that this content has now lost its consensus even if it's been previously considered to have one. I don't think it's theoretical that this is or can be disruptive. ONUS was never intended to be a blank check, that much is clear from the discussion. If we're not going to resolve the conflict with NOCON, at the very least, could we consider an option to limit the scope of potential misuse of ONUS? I just don't think it's a good idea or accurate to the intent of ONUS to allow a situation where long-standing content, with no consensus to change, can be removed by default. This effectively means that 2 editors can have policy endorse removing material that previously, let's say 10 editors supported including, if only 2 of those editors show up for the re-discussion at such time as it is called. Maybe they show up but not until 60 days later, too late. Now you need a new consensus to include that content. I don't see that as a good thing for the integrity of the material on Wikipedia. Andre🚐 20:54, 29 October 2024 (UTC)Reply

Huh… I thought one of the selling points of Wikipedia was that its material is dynamic… stuff gets added, stuff gets changed, stuff gets removed. Blueboar (talk) 21:06, 29 October 2024 (UTC)Reply

Dynamic, yes, but there's a point where it's not a net benefit to keep changing something, particularly a stable and controversial high-quality article. Surely you can come up with an example of this? There's a real example that caused me to come here, but I think better to keep it more abstract for a variety of reasons. There's a problem of defending article quality. I wrote that essay in 2005 and the original title was defending the status quo. I think now that the original title was more along the lines of the point needed. There are other solutions to the problem or maybe you don't agree it's a problem, but I think we're oversimplifying things if you think that Wiki just hums along with people adding or removing stuff, especially in controversial areas, and that every time some long-standing content is challenged that it's a legitimate and helpful challenge. Andre🚐 21:13, 29 October 2024 (UTC)Reply

I'm not so much worried about the 2 editors now but 10 editors give years ago stuff, things change and even if 2 editors removed it 2 can add it back 6 months later. Content changes and I fear the idea it must stand because a group of editors said so at some point in the past just tends towards stonewalling. Content can be removed for many reasons, even just to improve the article. Bloated articles are not better by default.

I don't think ONUS, or VNOT if editors dislike the word, should be used as a initial reason to remove content, as I've said, but my ideas to reformulate the statement didn't get much of any support and I'm in no hurry to try and revive it. -- LCU ActivelyDisinterested «@» °∆t° 22:52, 29 October 2024 (UTC)Reply

• My impression for previous discussions is that the dispute between WP:ONUS and WP:NOCON mostly amounts to editors disagreeing about when text gains a degree of implicit consensus by being longstanding and having people edit it. This is a complex question that we've avoided nailing down (for good reasons; having it too rigid would lead to people arguing over it rather than focusing on actual content.) But as a general rule-of-thumb I feel policy could be a bit more clear on the main points - content that has been seen by many eyes eventually gains a degree of consensus, since each of those people is presumed to have approved of it to some degree, and after that at least some consensus is needed to remove it. While people often use time as a rough handwave (often approximately three months of stability), and the "status quo" or "longstanding" are sometimes used to refer to it, it's really about the eyes; something on an obscure article that few people see might never accrue that sort of consensus, while higher-traffic articles can get it more quickly if it's clear a diverse group of people are editing it and seeing it. And any sort of indication that it's controversial (any objections on talk, say) prevent it from happening. --Aquillion (talk) 14:37, 31 October 2024 (UTC)Reply

If there is no longer consensus to include something, it can be removed. Also a "finger on the scale towards exclusion" is the right way to have it. Almost all content policy is about restricting what can be put into articles and hardly any of it is about restricting what can be taken out. There's no reason for this example to be different. Zero^talk 03:10, 29 October 2024 (UTC)Reply

That's not how policy or practice works; if we have an RFC over whether to remove an established section of an article (one that previously had consensus of some form), and it reaches "no consensus", the result will be that it is retained. I disagree with your proposal to add a finger on the scale towards exclusion, too; current policy places a priority on stability instead. Adding a finger on the scale towards exclusion encourages WP:STONEWALLing and discourages people from engaging. Generally speaking I don't think that proposals for these sorts of "strong" policies with default outcomes that could drastically change articles are ever workable - heavy-handed policies seem good in people's heads when they picture it giving the "right" answer, but in practice they reduce incentives to engage and compromise, encouraging people to use policy as a bludgeon instead. It's better to have a grey area with. (As an aside, this is, to me, the most frustrating part of discussions about ONUS - people who want to push for an expansive interpretation of it, which would allow for the removal of longstanding text without a consensus to do so, frequently make baffling assertions that this is already how things work, which anyone who edits Wikipedia in controversial areas knows is not the case. Outside of a few situations, like WP:BLP, no-consensus outcomes result in the status quo being retained, not in removal; the standard for disputes is WP:BRD, putting the article back in the status quo until a consensus is reached, not Bold-Revert-Remove.) --Aquillion (talk) 14:37, 31 October 2024 (UTC)Reply

I agree with this and your previous post. WP:ONUS is a finger on the scale towards exclusion and I opined that such a ham-handed arbitrary thing is a bad idea. Zero0000 opined that it is a good idea. So it wasn't a proposal to add something. Sincerely, North8000 (talk) 18:29, 31 October 2024 (UTC)Reply

Agree with North and Aquillion in my humble opinion. While it may not gain consensus to be changed, I encourage a constructive proposal to change text on these lines to make it clearer. Andre🚐 19:12, 31 October 2024 (UTC)Reply

There are differing levels of consensus though. If something has an RFC support for, then maybe it makes sense to maintain inclusion until there is a consensus against. But if something just has an implied consensus through dint of being in the article over time, then once that implied consensus is gone by having been challenged then it should not be retained absent a consensus for it. By making it so default to status quo you are empowering filibustering to retain challenged material that never had any real firm consensus to begin with. Something that say 4 users discussed but is now challenged by 20 shouldn't be retained either. The circumstances matter, and filibustering shouldn't be rewarded. When there was a larger discussion among 20 users, then yes I agree that this consensus holds until a new one forms, but the other way around? Not so much. nableezy - 19:27, 31 October 2024 (UTC)Reply

Well...

If I write an article and include some bit of content, and it 'sticks', then that's evidence of a presumption of consensus.

If we have a discussion that does not prove that presumed consensus to actually exist, should we keep assuming that there is a consensus? WhatamIdoing (talk) 21:05, 31 October 2024 (UTC)Reply

At best, that's silent consensus, conlevel 0, WP:SILENT. Selfstudier (talk) 21:37, 31 October 2024 (UTC)Reply

Do you mean the pre-discussion situation is a silent consensus? Once a discussion has happened, I wouldn't say that we're still silent. WhatamIdoing (talk) 02:07, 1 November 2024 (UTC)Reply

Hard to say without an understanding of the "discussion" (conlevel again). Selfstudier (talk) 10:53, 1 November 2024 (UTC)Reply

@Selfstudier, you surprise me. Do you mean that if I add a sentence to an article, and nobody edits that section of the article for the next year, nobody posts anything on the talk page, and nobody mentions it or refers to my edit in any way, then it's difficult to for you tell if that's "silent"?

Or that if someone does react to it, e.g., by posting a message on the talk page, then it's difficult for you to tell whether "saying something" is "silent"? WhatamIdoing (talk) 22:45, 1 November 2024 (UTC)Reply

First case, that's silent, the second, which was what I was referring to, hard to say with knowing what it is that was said. You want to be general, I want to be specific. Can't have that so we are left with the unknowable. Selfstudier (talk) 23:16, 1 November 2024 (UTC)Reply

Okay, I think that the word silence means the opposite of someone communicating with you, especially in the context of Wikipedia:Silence and consensus, which says "Consensus can be presumed until disagreement becomes evident. That is typically through reverting, editing, or stating disagreement on a relevant talk page." In that model, "if someone does react to it, e.g., by posting a message on the talk page", then it's not silence. WhatamIdoing (talk) 23:43, 1 November 2024 (UTC)Reply

I agree with what is said there in general but I think silence means have not specifically said whether something is agreed. If the edit was originally made, say on a page with low traffic, it might well sit there for a while but I don't think that means it necessarily has consensus, even if someone had commented about it but not said something yea or nay. Selfstudier (talk) 00:21, 2 November 2024 (UTC)Reply

Even on high traffic pages, an edit might stay under the radar for quite a while, can't assume it has consensus just because no one said anything. In practice it has, because it sticks in the article until someone actively does something to unstick it. Anyway, I find all these discussions to be the same, we end up going around in circles because the issues are all connected, I don't think there is supposed to be one correct answer to any problem and the PAG reflect that, perhaps not in the most efficient or clear way, but they do all the same. Selfstudier (talk) 00:26, 2 November 2024 (UTC)Reply

I agree with you that things can be overlooked, but you said above that if someone actually, directly has a discussion about this – a discussion that results in no consensus, which can only happen in there are editors saying that they support and other editors saying that they oppose – then that whole discussion could still be "silence". That does not sound reasonable to me. WhatamIdoing (talk) 00:54, 2 November 2024 (UTC)Reply

I don't think I said that. Where? Selfstudier (talk) 10:47, 2 November 2024 (UTC)Reply

I asked if it's difficult for you to tell whether "saying something" is "silent"?. You replied that it's hard to say whether "saying something" about an edit is "silent".

And all of this in the context of a scenario in which:

• There is a discussion
• involving more than one person
• that ends up with a no-consensus result.

I cannot imagine a scenario in which a discussion(!) that ends in "no consensus"(!) constitutes silence. If nobody said anything, how would there be a discussion? If nobody said anything about supporting or opposing the change, how could the discussion end in no consensus? WhatamIdoing (talk) 17:11, 2 November 2024 (UTC)Reply

Depends on the discussion (that's what I actually said). Selfstudier (talk) 17:16, 2 November 2024 (UTC)Reply

Please explain to me how you can have a discussion, about the desirability of an edit, that results in no consensus, and still not have anybody talking about the desirability of that edit. WhatamIdoing (talk) 17:28, 2 November 2024 (UTC)Reply

You are assuming that the discussion (of an edit) is about the desirability of that edit. I said it is not possible to generalize, specifics can be discussed but that's no use for PAG, right? Selfstudier (talk) 17:31, 2 November 2024 (UTC)Reply

Can you give me an example of a discussion that

1. is about a particular edit, and
2. ends in "no consensus" for or against that edit, but
3. is somehow not about the desirability of that edit?

WhatamIdoing (talk) 20:15, 2 November 2024 (UTC)Reply

2. is ur condition, not mine. Selfstudier (talk) 23:06, 2 November 2024 (UTC)Reply

I agree: "2" is the condition I have been asking you about, and whose answer you have been steadily evading.

@Aquillion said "if we have an RFC over whether to remove an established section of an article (one that previously had consensus of some form), and it reaches "no consensus"..." I said "If we have a discussion that does not prove that presumed consensus to actually exist..."

And you basically said ¯\_(ツ)_/¯, as if a whole RFC on whether to remove an edit could actually represent silence on the question of whether there is consensus for having that edit. WhatamIdoing (talk) 23:43, 2 November 2024 (UTC)Reply

In my opinion, it seems best to conceptualize WP:IMPLICITCONSENSUS only in terms of identifiable individual editors signalling identifiable (if implicit) assents. If an obscure article remains in a given state for seven years and only one editor has ever contributed to it, there's no evidence that another editor is aware of its contents, never mind that they've consented to them. The same would generally hold regarding editors that previously engaged with an article but are no longer active—their consensus can no longer change, so it can't really be considered. Remsense ‥ 论 21:51, 31 October 2024 (UTC)Reply

The above are discussions about the complexities of wp:consensus. This overall discussion is about what has operatively become arbitrary interference/conflict with the wp:consensus process....wp:onus. Which has jumped the track from it's original intent of keeping meeting wp:verifiability from being used to coerce inclusion. Again suggest fixing the whole mess and going back to the original intent by substituting onus with "wp:verifiability is a requirement for inclusion not a reason for inclusion". Sincerely, North8000 (talk) 02:23, 1 November 2024 (UTC)Reply

Apologies for getting sidetracked. If I've read the room correctly: would it be fair to characterize WP:ONUS as a blunt tool to initiate the consensus-making process, rather than as a blunt tool to circumvent it? While I don't think you're wrong in observing it can function as a finger on the scale towards exclusion, I think in most cases it is better described as tending towards stability, because in a majority of applicable situations stability and exclusion happen to be the same outcome. I don't really think it is a natural policy point to apply if one is trying to remove material, unless I'm missing soemthing? Remsense ‥ 论 03:51, 1 November 2024 (UTC)Reply

Not in my experience. ONUS can be cited when a few editors want to remove something, or prevent the addition of something. Verifiability doesn't guarantee inclusion, so this should be removed unless there is a consensus to retain it. Andre🚐 03:58, 1 November 2024 (UTC)Reply

Someone yank me back off-stage if I'm veering off-topic again, but I do think this is directly relevant? It's tough, because I will diverge with opinions stated above (and likely majority opinion in the community, which is fine) in that I don't think presence should be privileged over absence when it comes to discussions on the fringes or outside of WP:V and WP:NPOV. Very often, if we want to entertain cohesion or parsimony as valid editorial goals, we have to fall back on a hamfisted-if-not-false argument that a given content presentation (or any given aspect that doesn't boil down to the claims made by prose) "doesn't reflect sources". WP:PRESERVE says what it says, but it remains indefensible to me that additions should categorically require lower editorial scrutiny than removals do: it is equally important to the quality of articles to discern what they shouldn't say or present, compared to what they should. Remsense ‥ 论 04:15, 1 November 2024 (UTC)Reply

I think I agree with the general sentiment. What I think shouldn't be a permissible practice is to remove something that is long-standing or had consensus in the past citing ONUS. If that BOLD removal is reverted, it should remain unless there is consensus to remove it, not just no consensus to keep it. The BURDEN should be on the BOLD change, not on keeping the status quo. Andre🚐 04:18, 1 November 2024 (UTC)Reply

Depends (again). ONUS should not be an initial reason for removal and then it depends on the "discussion" (and editing) that follows. Selfstudier (talk) 10:56, 1 November 2024 (UTC)Reply

A lot depends on the circumstances of a removal. If I remove something with the explanation "this does not appear in the source", then it should stay out until someone shows that it is in the source or provides another source. Nobody should be saying "it stays in until you get consensus to remove it". Conversely, people who remove sourced relevant content without adequate explanation shouldn't cry "ONUS" if it is reinserted. In general, removals that are accompanied by cogent policy-based reasoning should be treated differently from removals that aren't. Zero^talk 10:55, 1 November 2024 (UTC)Reply

Yes, I agree with that. Andre🚐 23:03, 1 November 2024 (UTC)Reply

Regarding the "finger on the scale" issue, I stand by my comment. Articles are our face to millions of readers and we should revere them as precious places that contain only our best work. Treating stuff in an article as equal to stuff not in an article is fundamentally opposed to that principle. In my opinion, a policy that makes removals as difficult as insertions can only lead to deterioration of quality. Zero^talk 10:55, 1 November 2024 (UTC)Reply

I don't think I share the same values.

Firstly, I don't think readers want us to revere articles as precious objects; I think they want us to get some information to them. In the case of a recent event, I definitely think that they prioritize quantity over quality. If treating the article like a precious thing means that people don't get the information they want and need, then we're failing them.

Secondly, the Wikipedia:Editing policy says that Wikipedia is, in principle, best off when it contains more information instead of less. We are also supposed to Wikipedia:Be bold, which sometimes looks closer to Move fast and break things than like perfectionism. WhatamIdoing (talk) 23:00, 1 November 2024 (UTC)Reply

I'm an agile guy, which tries to get at a process that is fundamentally move fast don't break things, and I tend to think that articles, like software, should be shipped warts and all if having an article is the alternative to not having one, so I tend to agree with WAID that perfectionism can be harmful in the goal of continuously shipping articles. I don't think articles are precious any more than chickens are precious. Sure, in some sense, all life is precious, and there's a preciousness to your pet chicken, especially your prize chicken that has special longevity or traits. On the other hand, there are likely at least 20,000 less impressive but more accessible dead chickens within a 15 mile radius from you. But, I do agree with Zero0000 that we should have a set of quality heuristics to ensure that successive changes are improving quality. So let's try to get as many articles being organic free range chickens as we practically can while recognizing that some are going to be factory farmed, might have salmonella chickens. We want to limit salmonella, while providing healthy enough non-organic chickens to people who are hungry. The food shouldn't make them sick (have false information or other obvious, 5-alarm fire, showstopping problems) but it may not always be a heritage farm free range chicken with the top organic certifications. So in my mind shipping a stub or a rough article is fine, because as WAID says everything has mistakes and we shouldn't let the perfect be the enemy of the good. ONUS being a fairly blunt instrument in its present formulation, I wonder if there's a version of ONUS/BURDEN that looks something like the last few comments, i.e., changes to text that is verifiable, neutral, and encyclopedic that is long-standing with no affirmative consensus to remove need some rationale, while new bold changes that are reverted should be discussed and need a consensus to add if challenged. Because I do think ONUS is useful if it's keeping out bad new additions and giving some proportional weight to older and less controversial stuff. Andre🚐 23:15, 1 November 2024 (UTC)Reply

If you're Agile, then Zero seems to be proposing Waterfall. But more importantly, he seems to prioritize "remove existing stuff" over "add missing stuff". Your precious chickens can get sick because they have too much salmonella, but they can also get sick because they have too few nutrients in their diet. WhatamIdoing (talk) 23:50, 1 November 2024 (UTC)Reply

You wrote "Wikipedia is, in principle, best off when it contains more information instead of less". As a dedicated inclusionist, I agree with that sentiment. However, "more information" does not mean "more information of any sort". The right interpretation of the principle is "the more good stuff, the better". We should add stuff to articles if it improves them, and the way we decide whether stuff will improve an article is by consensus. That's what ONUS means to me. If something already in an article loses consensus that it improves an article, then its inclusion becomes subject to recall (to choose a trendy word) and consensus has to be reacquired. That is not balanced between inclusion and exclusion, and I approve of that. Basically, something good missing from an article is undesirable, but it's a smaller sin than having something bad included in an article. Zero^talk 03:58, 2 November 2024 (UTC)Reply

WP:Consensus should be our guide. WP:Onus was originally intended to prevent interference with that (to prevent people using merely meeting wp:verifiability to coerce inclusion) but has turned out to be instead an arbitrary interference (and conflict with) with wp:consensus. Let's fix both problems by substituting "Verfiability is a requirement for inclusion, not a reason for inclusion". Sincerely, North8000 (talk) 15:14, 2 November 2024 (UTC)Reply

You proposed this before, I think, what happened? Selfstudier (talk) 17:43, 2 November 2024 (UTC)Reply

• I don't think ONUS should be cited as the reason for removing something, but if there is a reasonable claim that the content violates our other content policies or guidelines, then it seems like that's a case where ONUS would apply? That isn't too different from WP:LOCALCONSENSUS in terms of where and when I see it applying. Alpha3031 (t • c) 04:18, 2 November 2024 (UTC)Reply

  So... first of all, LOCALCONSENSUS is not about "a couple of editors made a decision on the talk page", even if I think those editors made the wrong choice. It's about "a couple of us decided that 'our' article is exempt from some generally applicable policy or guideline". Imagine, e.g., some editors saying that the Suicide article should be completely replaced with a big disclaimer. That violates our ordinary policies and guidelines, especially Wikipedia:No disclaimers, and no matter how strongly they feel about it, it's a LOCALCON and doesn't count (unless, you know, the rest of the community agrees with them and decides to change the rules).

  Or don't imagine a hypothetical scenario, if you prefer, because we have real ones: the reason we have LOCALCON in its current form is because WP:WikiProject Composers declared that 'their' articles are exempt from MOS:INFOBOXUSE. It took a couple of years of disputes and a painful RFC to convince them to change their tune (slightly: they went from "we officially ban infoboxes from all our articles" to "we strongly recommend against infoboxes, though technically we can't ban them").

  Secondly, I think ONUS gets invoked more often in disputes in which most editors agree that the addition is a bad one, but its lone proponent disagrees. After several rounds of "Yes, we know that you want to say that you disproved Einstein's theory of relativity, but it's not okay to put this in the article because it violates all the core content policies", followed by "But I did disprove Einstein, and it's compliant with every single policy in the world!", then editors tend to shrug and say: "The ONUS is on you to get consensus before you can re-add it. Good luck with that." WhatamIdoing (talk) 17:27, 2 November 2024 (UTC)Reply

The sentence "The onus is on those seeking to include disputed content, to achieve consensus for its inclusion." was added to the "Verifiability does not guarantee inclusion" section by JzG in August 2014 with no dispute at the time (this is Archive 62 time). I don't find any argument about it until 2020 (Archive 70) but I could have easily missed it. I don't agree that it conflicts with the core message of WP:Consensus, since the latter says "An edit has presumed consensus until it is disputed or reverted." If we accept that things in articles should have consensus to be there, it follows that disputed material without consensus should not be there. It could be that this principle belongs in a different section, or even in a different policy, but I strongly believe that it belongs somewhere. It is not contradicted by PRESERVE ("fixing something is better than removing it") since that envisages keeping something different from what is disputed. Zero^talk 07:26, 4 November 2024 (UTC)Reply

A consensus is basically a supermajority (yes, I know it's not a vote so lets say of arguments). Wp:Consensus covers the bases pretty well for the various scenarios. Why have arbitrary interference saying that a supermajority is needed for retention, even of long standing material? As you noted, this was added under "Verifiability does not guarantee inclusion" which I think is a strong indicator of why it was put in and allowed to stay....to prevent meeting wp:verifiability from being used to coerce inclusion. Simcerely, North8000 (talk) 14:15, 4 November 2024 (UTC)Reply

I was going to start a new discussion, but this was right in my face, so I decided I am going to add onto it rather than have two threads going here. Basically, editors are not on the same page on the interpretation of ONUS. Please have a look and comment there or discuss here. Graywalls (talk) 18:43, 13 November 2024 (UTC)Reply

A discussion on RSN has broken out which revolves around what exactly a self-published source is. From my perspective, WP:USESPS provides the only comprehensive guidance for what is or is not self-published, and it is pretty unambiguous.

However, many of the responses have come down to claiming that USESPS is wrong and should be disregarded or deleted.

Leaving aside the surrounding argument about bias and reliability for this specific source, does anyone uninvolved have any opinions on this? If editors are supposed to disregard the guidance on USESPS, I would like to know. Void if removed (talk) 14:01, 31 October 2024 (UTC)Reply

There was something way back that I had brought up the idea that self published means there is no editorial oversight from a person publishing their work (ignoring copy editing or the act of publishing it on a website).
So someone putting info up on Medium clearly is SPS, while aember of an advocacy org like GLAAD or SPLC will need editorial approve from their in house editors before sonething goes online, and thus is not an SPS. — Masem (t) 14:15, 31 October 2024 (UTC)Reply

Ok, but that isn't at all what USESPS says, and by that measure absolutely every corporate website isn't self-published. So - is USESPS just completely wrong?

My understanding is there is supposed to be some level of separation between the publisher and author. Void if removed (talk) 14:23, 31 October 2024 (UTC)Reply

No, that bears no relation to how WP:SPS has been interpreted until now. Most sources employ their own writers, after all; if the fact that the writers work for the publishers made something a SPS, virtually everything would be a SPS. What the part of USESPS that you're misinterpreting means is that something is a SPS if the writer and publisher are literally the same person, ie. there's no editorial oversight at all; or if the entire thing is presented as a single monolith with no indication of who writes what and therefore no way of determining that there's a distinction between publisher and author. GLAAD lists individual writers for the things it publishes in the acknowledgement section, and those writers are not themselves the publishers, so they don't fall under the part of USESPS that you're citing. Otherwise, by your argument, I could say "well, the New York Times' reporters and writers work for the New York Times, therefore the editorial controls are not independent and it's a SPS." Now, whether the editorial controls are good enough is another story, but you can't just say "the writers work for the publisher, therefore this is a SPS." --Aquillion (talk) 14:50, 31 October 2024 (UTC)Reply

I'm hoping for uninvolved input here, but again: that's not what USESPS says, which is:

If the author works for a company, and the publisher is the employer, and the author's job is to produce the work (e.g., sales materials or a corporate website), then the author and publisher are the same.

What I'm trying to get at is why USESPS says things that are completely different to how you and some other editors are interpreting SPS. If Business, charitable, and personal websites are given as a specific example of SPSs, why is this charity's website not self-published? If USESPS says a SPS can have a professional structure in place for deciding whether to publish something, why are people arguing that having some sort of notional approval process means it isn't self published?

I want to know why what the guidance says, and what half a dozen editors are currently saying over on RSN, are so at odds, and which is correct? Void if removed (talk) 16:02, 31 October 2024 (UTC)Reply

Masem, you give an example of Medium.com vs SPLC, but the two collide at https://medium.com/@splcenter I'm sure their PR department does in-house review for both the content on their official Medium.com posts and the content on their https://www.splcenter.org/ website. The level of editorial oversight is the same in both cases, and I suggest that both of these are self-published, because in both cases, the org is posting whatever it wants, whenever it wants. WhatamIdoing (talk) 23:03, 31 October 2024 (UTC)Reply

Counterexample, apparently NYTimes authors are also writing on medium with the NYTimes logo.

https://medium.com/@timesopen

And the WaPo folks just have their own account on there:

https://medium.com/@washingtonpost

Similar counterexamples, all the news media sites out there operate Twitter accounts, and WP:twitter is usually always considered WP:SPS.

Operating on a SPS platform to promote itself does not imply SPS of the organization. Bluethricecreamman (talk) 23:21, 31 October 2024 (UTC)Reply

The definition of self-publishing exempts traditional publishing houses, so news articles are non-self-published no matter where they post.

Of course, any organization can self-publish something. The pages on their websites that tell you how much a subscription costs, or that try to talk you into running ads in the newspaper are all self-published. WhatamIdoing (talk) 02:02, 1 November 2024 (UTC)Reply

I'm not sure that USESPS is the issue so much as this material from WP:V, Self-published material is characterized by the lack of independent reviewers (those without a conflict of interest) validating the reliability of the content. Further examples of self-published sources include press releases, the material contained within company websites, advertising campaigns, material published in media by the owner(s)/publisher(s) of the media group, self-released music albums, and electoral manifestos: Springee (talk) 16:27, 31 October 2024 (UTC)Reply

I agree that there are some difficulties in the line "Self-published material is characterized by the lack of independent reviewers (those without a conflict of interest) validating the reliability of the content."

I think that speaks to our hopes and aspirations for reliable sources, but this sentence tacks on some irrelevant things. Self-published material is characterized by the lack of editorial control. Hopefully, usually, that editorial control will involve independent reviewers and even professional standards; hopefully it will involve something validating whether the content is accurate. But a source can be validated for reliability by independent reviewers and still be self-published. WhatamIdoing (talk) 22:36, 31 October 2024 (UTC)Reply

USESPS seems a mess, and doesn't seem to match commonity pratice. If we're relying on an essay that says a self-published source have certain features, but then has to specifically exclude news organisation because they have all the features mentioned, then it's obviously going to cause confusion.

Sources such as SLPC and SBM are accepted as not being self-published, and there are even RFC with that result. Interpreting such sources as self-published doesn't seem to match up with the wording of WP:SPS, and it doesn't appear to line up with the ideas expressed in Self-publishing. Particularly that self-publishing involves the author of the work publishing the work themselves at their own cost. -- LCU ActivelyDisinterested «@» °∆t° 23:49, 31 October 2024 (UTC)Reply

So when a dictionary says that self-publishing means to "publish (a book etc.) oneself rather than through a publishing house", do you find that their exclusion of publishing houses (a group that includes newspapers) to be confusing? WhatamIdoing (talk) 02:05, 1 November 2024 (UTC)Reply

What makes no sense is you dividing publishing. News is not the only thing that is published. And you seem to have lost 'oneself', a writer reporting for a non-governmental organization is not just one self and is not publishing oneself, the organization is the publisher. Alanscottwalker (talk) 09:42, 1 November 2024 (UTC)Reply

And Donald Trump doesn't write the content on his campaign website, so he's not "onself", either, and it's mostly not being paid for with his own money. Are you interested in declaring https://www.donaldjtrump.com/ to be a non-self-published source? Or do you think that it might be possible for a self-published website to involve more than one author? WhatamIdoing (talk) 20:52, 1 November 2024 (UTC)Reply

Your question is too provocative in this day and time, but the concern should be whether in the given content situation, it satisfies source requirement, particularly for accuracy and fact checking. This will be informed in part by what others have published about it. -- Alanscottwalker (talk) 14:50, 4 November 2024 (UTC)Reply

So ...just not care about whether it's self-published? We could, I suppose, remove those rules entirely, and tell editors that they should only be concerned about whether in the given content situation, it satisfies source requirement, particularly for accuracy and fact checking. This will be informed in part by what others have published about it. While I think this could actually be a desirable way to handle this policy (and BLPSPS, which depends upon it), I'm extremely doubtful that we could get the rest of the community to accept it. WhatamIdoing (talk) 18:48, 4 November 2024 (UTC)Reply

We always care about publishing in the Source analysis but it does not turn your distorted concepts of self-publishing. It's also pretty easy to make a principled argument that, as a rule, political campaign literature is not considered sufficiently reliable for opponents and others for a whole host of good reasons, none of which turn on your essay views. Alanscottwalker (talk) 19:53, 5 November 2024 (UTC)Reply

SBM I have no idea how that was arrived as non-self published, it seems to be a group blog, but from reading the conversations there is some credence given to the fact it has editorial oversight? And yet recent controversies suggest that isn't actually the case for some of the editors, so again it is a mixed picture.

SPLC, I see no RFC to that effect, and lots of similar debates in the past with a similar split of people arguing it both is and isn't self-published. IMO, any material a lobby group writes and posts itself on its own website is self-published, whether that's SPLC or Heritage or anyone in between. Maybe there's a grey area around externally authored reports that they endorse and publish, but website content written by employees or an internal team? Its self-published. There's no independent editorial process, and a simple process of internal authorisation before publishing something is not sufficient.

While I accept there might be grey areas and hard cases, I really don't think the specific example that kicked off this discussion - material straightforwardly written and posted by GLAAD to its own website - is one such case. Void if removed (talk) 09:50, 1 November 2024 (UTC)Reply

GLADD write? It's not a writer, its an organization, it can't write (at least before AI). Alanscottwalker (talk) 10:13, 1 November 2024 (UTC)Reply

This is an unfeasibly narrow definition of self-published.

WP:USESPS says: If the author works for a company, and the publisher is the employer, and the author's job is to produce the work (e.g., sales materials or a corporate website), then the author and publisher are the same.. That seems sensible to me and in line with the view that virtually everything on any corporate website is self-published.

What you're saying is completely contrary with the essay provided as guidance for WP:SPS. Your answer is that WP:USESPS is wrong or that it isn't policy and we should ignore it. Fine, but I disagree, and what you're saying is at odds with what it says in the footnote on WP:V quoted above: the material contained within company websites. And that is policy.

What I want is non-involved input to help explain which is correct and why this seeming vast gulf in editor opinion exists. Void if removed (talk) 10:24, 1 November 2024 (UTC)Reply

The essay, you rely on is not policy and needs no consensus. The essay is written broadly, lacks nuance, and is overinterpreted by you. That it leads in your mind to absurdity is the problem you face. When you come up with sales material that you want to use in a BLP, we can talk about then (but a more fruitful discussion would be COI) but your wanting to suggest in effect everything is sales-material is silly. Alanscottwalker (talk) 10:49, 1 November 2024 (UTC)Reply

Can you please stop calling me silly. Reading guidance attached to a policy is not silly. You have made your personal opinion clear, many times over, so I suggest you move on.

What I want is non-involved input to help explain which is correct and why this seeming vast gulf in editor opinion exists. Void if removed (talk) 11:18, 1 November 2024 (UTC)Reply

I did not call you silly, I explicitly referred to your suggestion. The problem is your reading of that essay not a gulf. Whether you consider me involved also makes no sense.Alanscottwalker (talk) 13:37, 1 November 2024 (UTC)Reply

That essay needs to be rewritten from the ground up. It doesn't match community practice and implements a definition that is way broader than that found in policy. -- LCU ActivelyDisinterested «@» °∆t° 16:35, 1 November 2024 (UTC)Reply

Repeating a comment I made at WP:RSN:

The footnote at WP:SPS that Springee quotes is badly written in that it conflates two different things. By comparison, the sources quoted in the same footnote do a good job distinguishing those two things: The University of California, Berkeley, library states: "Most pages found in general search engines for the web are self-published or published by businesses small and large with motives to get you to buy something or believe a point of view." (my emphasis); The Chicago Manual of Style, 16th Edition states, "Any site that does not have a specific publisher or sponsoring body should be treated as unpublished or self-published work." The problem with coca-cola.com is that the only content on there is about the product sold by the publisher, and everyone involved in publishing it has an interest in that product selling well; this is a good reason to not use coca-cola.com as a source, but it has nothing to do with the concept of self-publishing, and is not analogous in any way to material published by GLAAD.

(Personally I do not consider it shocking that Wikipedia policies are not always carefully or thoughtfully written: they're produced by the same process that Wikipedia is, but without recourse to reliable sources, and with much less scrutiny than any high-profile article.) 100.36.106.199 (talk) 13:38, 1 November 2024 (UTC)Reply

• Perhaps what is needed is a NEW guideline (or at least an essay) that specifically focuses on Advocacy organizations and think tanks. This could cover the nuances of reliability when it comes to such organizations and also discuss the need for attribution due to bias, and opinion. Blueboar (talk) 11:44, 1 November 2024 (UTC)Reply

  Possibly, but IMO this is quite a fundamental policy to have such divided opinion. We have a policy that by my reading - and by the explanatory essay - would consider the vast majority of company websites to be self-published, and yet when raised a roughly equal number of editors insist that the opposite is true, and that the vast majority of company websites are not self-published. The consequences of the latter being true are significant for contentious BLP claims about third parties, and since in this specific case the entire purpose of this website is for GLAAD to make BLP claims about third parties who it is politically opposed to, with no independent oversight or accountability, the matter of whether this is a SPS or not is something that needs settling. If this is fair game for inclusion on BLPs with attribution, that is a concerning state of affairs, and discussing this specific case reveals (to me) a divide in the way editors are interpreting SPS narrowly and completely dismissing USESPS, and more than anything this needs to be clarified. Void if removed (talk) 12:28, 1 November 2024 (UTC)Reply

  I don’t think company websites are the issue here. The vast majority of company websites do not contain material about third party people - ie those who are not directly connected to the company. (material about people who are directly connected to the company would be covered by WP:ABOUTSELF).

  The disputes seem to all be centered on advocacy groups… which do comment on third parties. Blueboar (talk) 12:46, 1 November 2024 (UTC)Reply

  Ah right I see. I don't think the for-profit nature of the organisation changes whether something is self-published or not, but yes advocacy groups are simply more likely to self-publish partisan opinions of the sort that probably shouldn't be in BLPs. Void if removed (talk) 14:31, 1 November 2024 (UTC)Reply

  It's definitely not all centered on advocacy groups. For example, with respect to BLPSPS, I've previously had to ask whether the Mueller Report was self-published or could be used as a source for a statement about one of the people it discusses, whether a professional society's newsletter was SPS or could be used a source for an NPROF who received a notable award from them, whether material written by an unknown person on a university website could be used, etc. FactOrOpinion (talk) 18:37, 1 November 2024 (UTC)Reply

  I would treat the Mueller Report as something like a SPS. In general we shouldn't cite without a RS telling us what we should be looking at first. Note that we of course would cite it as a supporting reference. Springee (talk) 19:37, 1 November 2024 (UTC)Reply

  I'd take the opposite approach. Governments – that's little-g government, meaning the organization that is in charge of providing roads and armies for a given place, not the name-brand government that refers to the reign of a particular political party (not, e.g., the Sunak government, the Putin regime, the Clinton Administration) – are, in some respect, traditional publishers (e.g., of laws and reports). WhatamIdoing (talk) 19:52, 1 November 2024 (UTC)Reply

  By this logic, all newspapers are SPS and missing a key distinction. Non-SPS are sources that at least have some kind of editorial oversight for accuracy, which newspapers obviously have (outside of opinion sections).

  What is being confounded here it seems is WP:INDY which is not the same as SPS. If a company website claims something, the question isn't whether it is an SPS or not, but rather that it generally wouldn't be included without third-party mention from a WP:DUE perspective due to lack of independence. That is why WP:SPS policy specifically says in relation to BLPs Never use self-published sources as third-party sources about living people, even if the author is an expert, well-known professional researcher, or writer. KoA (talk) 13:38, 1 November 2024 (UTC)Reply

  What is being confounded here it seems is WP:INDY which is not the same as SPS. Yes exactly this. 100.36.106.199 (talk) 13:52, 1 November 2024 (UTC)Reply

  I don't see your point - WP:INDY doesn't have the same BLP restrictions as SPS, so doesn't really cover the case where a source has a vested political interest in negative portrayal of third parties. Without independent editorial control, advocacy websites are a ready source of hyperbolic descriptions of BLPs. Void if removed (talk) 17:32, 1 November 2024 (UTC)Reply

  Again, please read SPS policy I mentioned above. It is explicit not to use self-published sources as third-party sources when it comes to BLPs. There's a ton of guidance linked right there. INDY helps give supplemental guidance on some of the topics you are trying to navigate here.

  If something is self-published, you can't use it that way. If something isn't independent of the subject, but is not an SPS, then that's where WP:DUE comes with respect to what WP:INDY gives guidance on as well, especially whether or not something is going to be included or not even with attribution. The main issue here is that you're missing existing guidance related to SPS and barreling past some of that. As others mentioned at the beginning of this thread, most advocacy groups are not just a single person setting up a blog, but instead have some degree of editorial oversight between the writer and the group actually publishing something. What you're trying to address is primarily not an SPS issue.

  I've seen and dealt with a lot of problematic advocacy group sources in the past, but generally SPS isn't the way to handle it. You either have to work through the lens of WP:RS or WP:DUE, and in general you're going to avoid advocacy group publications and reach for higher quality sources anyways. KoA (talk) 18:11, 1 November 2024 (UTC)Reply

• And universities, and historical societies, and . . . . The drive to see everything as for profit companies is not only not reality, it is generally useless for our purposes, we have a narrow area of concern, statements about living persons, and a narrow prohibition, statements that are self published about blps, because we doubt they have the safety in WP:SOURCE reliability that having a writer and a publisher has (among other things there are more/real defendants to sue), and that the self-published more likely lead to NPOV and OR problems. Alanscottwalker (talk) 12:50, 1 November 2024 (UTC)Reply

  I'm primarily concerned about the drive to conflate INDY and SPS. If we want to accept INDY SPS's, then we should just say so, instead of saying, "well, technically the author and the publisher are the same, and they're not actually traditional publishers, but we'll just close our eyes and pretend that they're not SPS, because WP:IAR is dead and otherwise I don't like the result". WhatamIdoing (talk) 19:55, 1 November 2024 (UTC)Reply

• I agree with Blueboar. To that effect, I propose the insertion into WP:RS, either in WP:SOURCETYPES or the section after, following a sufficiently widely advertised RFC, depending on whether the community considers grey literature in general to be self published, one of the following subsections:

Governmental bodies, as well as non-governmental organisations like think tanks and advocacy groups, often produce a variety of material in support of their missions. These are usually published and distributed outside of traditional commercial or academic channels, and may vary in reliability and degree of oversight. In general, any such material that has undergone some form of formal review process independent of the original author should not be considered self-published, whether or not the author is employed by the sponsoring organisation. This material may still be unreliable in other ways, and only groups with a well established reputation should be considered generally useful. If an organisation produces multiple forms of content, the level of review may vary, and sources subject to minimal or no review, or with a poor reputation for fact checking, should be considered self-published and only useful in the same situations as other self-published sources.

Or:

Governmental bodies, as well as non-governmental organisations like think tanks and advocacy groups, often produce a variety of material in support of their missions. These are usually published and distributed outside of traditional commercial or academic channels, and may vary in reliability and degree of oversight. In most cases, the internal review of such groups are not considered on par with traditional publishing, and not independent of the sponsoring organisation, thus usage should be constrained to that typical of other self published sources. Content authored by subject matter experts may be reliable if previous work has been published by reliable traditional publishers. If there is a formal review process, such content is usually still reliable for the opinions of the sponsoring organisation, and use with attribution may be appropriate for organisations well established in a topic area.

Please feel free to edit or do whatever to the proposed sections, I do not consider them part of my comment for TPG purposes. Alpha3031 (t • c) 13:12, 1 November 2024 (UTC)Reply

I think that second one is a pretty fair assessment of how I'm reading USESPS, and the sort of use I would personally think is appropriate. Such sources can be reliable, indeed are sometimes the best possible sources, irrespective of any bias. But unless there is compelling evidence otherwise in any specific case I would treat them as self-published and I would still consider them unsuitable for third party BLP claims, given the low threshold for publishing and lack of accountability. Void if removed (talk) 14:45, 1 November 2024 (UTC)Reply

I'd argue that if this remains contentious, we do an RFC to see what folks come to consensus on. Bluethricecreamman (talk) 15:32, 1 November 2024 (UTC)Reply

I don't think its contentious, I think some people just don't see where it leads: 'celebrity column, just fine!' . . . serious organization, absolutely not!' Alanscottwalker (talk) 15:42, 1 November 2024 (UTC)Reply

though i agree that i think the answer is straightforward, the significant convos here and WP:RS/N suggests otherwise.

If it doesn't quiet down, an RFC is useful for reorienting all of us to whatever consensus is nowadays. Bluethricecreamman (talk) 15:54, 1 November 2024 (UTC)Reply

I agree we're heading quickly towards RFC territory. Loki (talk) 16:24, 1 November 2024 (UTC)Reply

Not always. But really to create such a broad rule for such narrow issue, can 'this publication be considered for in BLP's, is likely even more contentious. Alanscottwalker (talk) 16:41, 1 November 2024 (UTC)Reply

I would oppose this outright as it goes in the wrong direction, it is instead USESPS that needs rewriting. -- LCU ActivelyDisinterested «@» °∆t° 16:32, 1 November 2024 (UTC)Reply

To be clear I don't think this is needed in V, SPS is a simple explanation and covers everything well. The broader definition used in USESPS appears to be the problem that needs solving. -- LCU ActivelyDisinterested «@» °∆t° 16:53, 1 November 2024 (UTC)Reply

I disagree that nothing is needed at WP:V. The footnote on WP:SPS says Self-published material is characterized by the lack of independent reviewers (those without a conflict of interest) validating the reliability of the content. Further examples of self-published sources include press releases, the material contained within company websites, advertising campaigns, material published in media by the owner(s)/publisher(s) of the media group, self-released music albums, and electoral manifestos, where "further examples" is a reference to self-published material such as books, patents, newsletters, personal websites, open wikis, personal or group blogs (as distinguished from newsblogs, above), content farms, podcasts, Internet forum postings, and social media postings in the main WP:SPS section.

For me, that phrase implies that material contained in other organizations' websites (e.g., non-profit websites, university websites, government websites) is also SPS. But I've seen multiple WP editors saying that company websites and organizational websites are not SPS, as they expect that there are editors involved in crafting the final version of the text on those websites. If that view is the one that people want reflected in policy, then this text about company websites needs to change, and other text (e.g., about newsletters) may need to change as well. Either way, I think the text needs to be modified to make clear that “material” means the specific work that’s being used to substantiate a WP claim (a specific webpage, book, article, etc.) rather than an entire website, all of an organization's publications, etc., as it’s possible that the latter includes a mix of content that’s self-published and content that isn’t self-published. For example, I'd say that:

• a transcript of a congressional hearing is SPS, but a Congressional Research Service report is not;
• a Youtube video posted by an individual is SPS, but a Youtube video posted by a news organization is not;
• a non-profit's edited report is not SPS, but their weekly podcast or a live-blog at an event is;
• an article on a site like Forbes.com might be SPS, while another article on that site originated in Forbes magazine and is not.

FactOrOpinion (talk) 18:15, 1 November 2024 (UTC)Reply

Sure, that is likely an example of the more one goes on about something, the more trouble on is likely to stir up. The truth is, we have a clear consensus on what is self published and it has nothing to do with just being on a website (perhaps that made more sense in the past). Now, we may have less or even no consensus on going out from the core of self published (what one publishes on their own), but perhaps we should consider less is more, here. Our purpose should not be self-publishing our thoughts on the world, or the state of publishing, it is just to address a narrow issue . Alanscottwalker (talk) 18:40, 1 November 2024 (UTC)Reply

(I made an editing mistake. When I said "For me, that phrase implies ...," the phrase I was referring to is "the material contained within company websites.") When you say we have a clear consensus on what is self published, who is "we," and where is that consensus stated? As a not-that-experienced editor, I'm trying to abide by WP's policies and I have to rely on what's written in a policy to help me figure that out, and I'm saying that what's written for WP:SPS says that company websites are SPS, when you and others here are saying that they aren't. FactOrOpinion (talk) 18:54, 1 November 2024 (UTC)Reply

I think I said it 'one publishing', things like social media, buying publication, etc. Alanscottwalker (talk) 21:02, 1 November 2024 (UTC)Reply

Honestly, I feel like this problem could be mostly solved simply by deleting this line from WP:USESPS: If the author works for a company, and the publisher is the employer, and the author's job is to produce the work (e.g., sales materials or a corporate website), then the author and publisher are the same.

This is obviously not true, since it would mean the New York Times is self-published. The fact that WP:USESPS says the NYT is not self-published doesn't change this, it just means that WP:USESPS contradicts itself. Loki (talk) 16:57, 1 November 2024 (UTC)Reply

Sure, go delete it. It's just an essay. Alanscottwalker (talk) 17:06, 1 November 2024 (UTC)Reply

That line comes from this diff [1].

its been a bit longstanding, but i'll go ahead and WP:BOLDLY delete it. Bluethricecreamman (talk) 17:10, 1 November 2024 (UTC)Reply

Thanks; I think your edit is an improvement. 100.36.106.199 (talk) 17:29, 1 November 2024 (UTC)Reply

@WhatamIdoing Just notifying you of this discussion about that line, I think you added that line.

Feel free to revert bold change, I see you are part of the current discussion too, and am curious about what community consensus ends up being. Bluethricecreamman (talk) 17:49, 1 November 2024 (UTC)Reply

I wrote Wikipedia:Identifying and using self-published works to reflect the results of the previous major blowup over this question. I think the underlying problem is still the same, namely: The source is technically self-published, BLPSPS prohibits using self-published sources for certain content, and I really, really, really want to use it anyway.

As long as the community doesn't twist the definition of this dictionary-definition word into some sort of special wikijargon, I'll likely be satisfied with the outcome.

This specific sentence: If the author works for a company, and the publisher is the employer, and the author's job is to produce the work (e.g., sales materials or a corporate website), then the author and publisher are the same does not appear to be clear to people, so it might benefit from including a concrete example (e.g., "Ima Investo–Relations produces the annual financial report, which Cal Communications posts on the corporate website").

It is also possible that it would help to show a color-coded table explaining the application of the definition to some of the typical questions. WhatamIdoing (talk) 18:53, 1 November 2024 (UTC)Reply

I would agree with this, it's far to broad a definition. -- LCU ActivelyDisinterested «@» °∆t° 17:07, 1 November 2024 (UTC)Reply

I mean, when an essay directly contradicts the controlling line of policy, I'm inclined to just ignore it until whoever wrote it brought it into alignment. Nobody approves essays, they don't have any force. I'm suggesting something inserted into WP:RS (not WP:V) because it covers NEWSORG as well, it's an OK place to discuss common types of sources. Alpha3031 (t • c) 01:21, 2 November 2024 (UTC)Reply

I think it's most useful to start at the beginning. When you are evaluating whether a given source is reliable for a given claim, it is useful to know:

• Is this WP:SELFPUBLISHED or not?
• Is this WP:INDEPENDENT or not?
• Is this WP:PRIMARY or not?

All three of these WP:UPPERCASE are ultimately meant to refer to their ordinary, real-world meanings. None of these three are meant to be special wikijargon (contrast with WP:NOTABLE), and even when we have had to settle a few details on our own (e.g., WP:V's definition of WP:Published), we have tried to stick to the real-world meanings.

For the question of self-published sources, we can look to proper reliable sources, like dictionaries, to find out what they mean. These sources consistently give definitions that amount to this:

1. The author and the publisher are the same.
2. The publisher is not an established or traditional publisher (e.g., like a daily newspaper).

For example, if we look at coca-cola.com, we find these facts:

• Author = the employees of The Coca-Cola Company
• Publisher = the employees of The Coca-Cola Company
• The Coca-Cola Company is a multinational beverage company, which is not a traditional publisher.

The evaluation looks like this: Author = publisher; publisher ≠ traditional publisher; ∴ coca-cola.com matches the definition of self-published.

We can check our interpretation of the definition against what sources say, e.g., in the library website that says "Most pages found in general search engines for the web are self-published". A page from coca-cola.com is the kind of page you'd expect to find via a general web search engine, and it is not the kind of page you'd expect to find in a more scholarly-oriented search engine (e.g., via PubMed or EBSCO). This therefore confirms our definition and its application to a website like coca-cola.com.

The problem here does not actually seem to be about what the word self-published means. The problem here seems to be a straight-up Argument from consequences, which is "based on an appeal to emotion and is a type of informal fallacy", per that article's lead. In other words, no editors have any factual reason to believe that coca-cola.com isn't actually self-published; however, some of them just don't like the consequences of that factual determination. For those editors, I suggest the following two paths forward:

1. Determine whether the entity in question should be considered a traditional or established publisher. For example, it is not unreasonable to decide that the United States Census Bureau is a "traditional or established publisher", at least with respect to the United States census reports, and therefore automatically exempt from the definition of self-publishing. (This is what @Alpha3031 was getting at with the "grey literature" notes.)
2. Change the policy to accept that, yes, some sources are technically self-published, but we're going to use them anyway. WP:BLPSPS is our own homegrown rule, and we can change it. There's no inherent reason why we couldn't say that self-published sources are generally banned, but in the case of certain reputable self-publishing organizations, such as widely respected advocacy groups, a well-advertised discussion could determine a consensus for a limited exception to the normal policy. For example, we could accept that GLAAD's GAP database is self-published and still choose to allow its use.

WhatamIdoing (talk) 18:42, 1 November 2024 (UTC)Reply

Merriam Wesbster [2] (per that definition, no, the Coca Cola site is not self published, certainly not by the employees), so it is quite a stretch for you to argue you did not just write a whole bunch of Wikijargon. And it is entirely silly in this day and age to say, it is on a website, thus it is self published. Alanscottwalker (talk) 19:04, 1 November 2024 (UTC)Reply

I would agree with this, the definition that was used by USESPS was wikijargon that didn't match real world sources. That's not to say that such sources are reliable or due for inclusion, but those issues don't need to be covered by this single definition. -- LCU ActivelyDisinterested «@» °∆t° 19:35, 1 November 2024 (UTC)Reply

@ActivelyDisinterested, I'd love to see a list of real-world sources that actually disagree with USESPS. WhatamIdoing (talk) 19:44, 1 November 2024 (UTC)Reply

Well Wikipedia editors disagree with the definition in USESPS, and have done in mutiple discussions. If it can't be brought inline with actual practice it should be discarded. -- LCU ActivelyDisinterested «@» °∆t° 19:48, 1 November 2024 (UTC)Reply

@ActivelyDisinterested, you said the definition that was used by USESPS was wikijargon that didn't match real world sources.

The definition at the top of USESPS is: "Self-published works are those in which the author and publisher are the same".

I see no wikijargon in that sentence. Do you?

The definition in that sentence identifies the core point of every dictionary definition I've seen, including the one Alan links to above. Can you provide a counter-example? Any definition, from anywhere, that says "If you write it, and someone else decides whether to make it available to the public, then it's still self-published"? Or one that says "If you write it, and you publish it, then it's probably not self-published"? WhatamIdoing (talk) 20:03, 1 November 2024 (UTC)Reply

Stop playing games, of course I accept "Self-published works are those in which the author and publisher are the same".

The question is in regard to the definition you added to USESPS of "If the author works for a company, and the publisher is the employer, and the author's job is to produce the work ... , then the author and publisher are the same". This is very obviously not how self-publishing is being understood in practice and doesn't match the result of discussion and RFCs about sources.

If you want that definition to be used, which would overrule those discussions, then you need consensus to overrule those discussions. Otherwise those discussions and RFC show a consensus that your definition doesn't have consensus. -- LCU ActivelyDisinterested «@» °∆t° 20:16, 1 November 2024 (UTC)Reply

I agree that this sentence is suboptimal. I do not agree that it is a definition, and I do not agree that it contains any wikijargon.

I also don't think that this example is "not how self-publishing is being understood in practice and doesn't match the result of discussion and RFCs about sources". This is consistent with the discussions in Wikipedia:Reliable sources/Perennial sources#Peerage websites, WP:RSPYT, WP:RSPTWITTER, The Skeptic's Dictionary, etc.

I think what editors want is a way to say:

• Science-Based Medicine is non-self-published, but Quackwatch is at least partially self-published and ScienceBlogs are entirely self-published: I think the existing advice lets people do that.
• Southern Poverty Law Center and GLAAD – even if they technically meet the real-world definitions – are not subject to BLPSPS, but CounterPunch magazine should be, and Consortium News at least partially should be, and all political campaign websites definitely should be.

WhatamIdoing (talk) 20:48, 1 November 2024 (UTC)Reply

No what editors want to say is that you definition in USESPS doesn't match their expectations of a self-published source. I have no idea how you think Twitter and YouTube back up your definition, it would still be self-published by anyone's standard. -- LCU ActivelyDisinterested «@» °∆t° 20:53, 1 November 2024 (UTC)Reply

Also as it stands the wording at USESPS doesn't have consensus. The RFC on SBM and many such discussions on SLPC have rejected that definition. This should reflect practice, not be a definition thought up and imposed. -- LCU ActivelyDisinterested «@» °∆t° 19:37, 1 November 2024 (UTC)Reply

I don't think the editors in the SBM discussion rejected the consensus; I think they rejected the consequences. WhatamIdoing (talk) 19:45, 1 November 2024 (UTC)Reply

Oh come on, that is just playing word games. In practice your definition stands rejected. -- LCU ActivelyDisinterested «@» °∆t° 19:49, 1 November 2024 (UTC)Reply

I don't think so. I think what editors are rejecting is the constraint that BLPSPS imposes on them. They'd rather change the definition than change the rule, but I don't see anyone saying things like "I looked in my dictionary, and it totally says that corporate websites, published with no external controls, are non-self-published". I only see editors saying "I hate this definition, because if I follow this definition, then the BLPSPS rule applies, and I don't want the BLPSPS rule to apply." WhatamIdoing (talk) 20:05, 1 November 2024 (UTC)Reply

No. You are playing word games to no useful end. Alanscottwalker (talk) 20:34, 1 November 2024 (UTC)Reply

It is just word games. The RFC doesn't match how you want self-published to be defined, so you denigrate the RFC so it's results can be dismissed.

The result of the RFC is that it is not self-published, and as it's not self-published BLPSPS doesn't apply.

"I dislike the RFC result and believe BLPSPS should still apply" is the quote you are looking for. -- LCU ActivelyDisinterested «@» °∆t° 20:46, 1 November 2024 (UTC)Reply

The RFC on SBM at Wikipedia:Reliable sources/Noticeboard/Archive 256#RfC on sciencebasedmedicine.org ended with this consensus summary: "There is consensus that Science-Based Medicine is not a self-published source. Editors acknowledge that the website describes itself as a "blog". However, most editors believe that the website's published editorial practices are sufficiently robust to prevent it from being classified as "self-published". Some editors note that submissions from both staff writers and contributors are subject to fact-checking."

In other words, SBM is non-self-published because it has a "sufficiently robust" version of the "editorial practices" that we associate with traditional publishing houses. WhatamIdoing (talk) 21:00, 1 November 2024 (UTC)Reply

I would agree that the result agrees that an author cannot being able to publish their work without editorial oversight means it's not self-published. -- LCU ActivelyDisinterested «@» °∆t° 21:07, 1 November 2024 (UTC)Reply

And I also agree with this, and so does the definition in USESPS. WhatamIdoing (talk) 21:34, 1 November 2024 (UTC)Reply

If editors won't to impose a definition that would overrule multiple community discussions, then they need to find consensus for it. It is obvious that no such consensus currently exists. -- LCU ActivelyDisinterested «@» °∆t° 19:54, 1 November 2024 (UTC)Reply

Really? Their definition is "to publish (a book) using the author's own resources".

Who writes that website? The Coca-Cola Company, through the method of hiring employees to do so. Whose resources are used to publish it? The Coca-Cola Company's own resources.

I think you have made the twin mistakes of ignoring corporate authorship and assuming that "own resources" means "an individual human's personal money".

See also definitions such as

• "publish (a book etc.) oneself rather than through a publishing house" – The Canadian Oxford Dictionary
• "to arrange and pay for your own book to be published, rather than having it done by a publisher" – Cambridge Dictionary
• "To publish a work independently, without the assistance of a publishing company." – wikt:self-publish
• "to publish or issue (one's own book or other material) independent of an established publishing house" – [3]
• "Self-publishing is the process of publishing a book by the author without the involvement of a traditional publishing house or company." [4]
• "Self-publishing is the process of independently publishing and distributing a book, e-book, or other written work, without the involvement of a traditional publishing house or company." [5]
• "Self-publishing refers to the process of independently publishing and distributing your own work, such as books, e-books, or other written materials, without relying on traditional publishing houses or companies." [6]

WhatamIdoing (talk) 19:43, 1 November 2024 (UTC)Reply

No. What's plain you are creating your own illogical wikijargon and requirements that are baseless and factually inaccurate. No, the employees are not self publishing, they are being published by the corporation, and its the corporation's resources that are being used. Alanscottwalker (talk) 20:20, 1 November 2024 (UTC)Reply

So why does WP:SPS say that "the material contained within company websites" is an example of self-published content? FactOrOpinion (talk) 20:35, 1 November 2024 (UTC)Reply

We know why, someone wants to make a black and white but false statement. And it does not practically matter, anyway, we will use what's found on its website all over the pedia when we talk about the company and what's related to it.Alanscottwalker (talk) 20:49, 1 November 2024 (UTC)Reply

That sentence was added in October 2011 by Wifione. It doesn't look like the kind of edit that is trying to distort anything. WhatamIdoing (talk) 21:20, 1 November 2024 (UTC)Reply

The question isn't whether "we will use what's found on its website all over the pedia when we talk about the company", which is permitted for self-published sources in WP:ABOUTSELF. The question is only whether it's self-published. WhatamIdoing (talk) 21:23, 1 November 2024 (UTC)Reply

Come on. Yes we will use it to talk about living persons. And no it is not self published. Alanscottwalker (talk) 22:50, 1 November 2024 (UTC)Reply

I don't know who "we" is. I asked because I didn't know why, and because as a not-so-experienced editor, I'm using the policy to guide my actions, and this is part of the policy. That phrase was introduced in 2011, which is a long time for it to have been on the page if there's some consensus that it's wrong. FactOrOpinion (talk) 21:25, 1 November 2024 (UTC)Reply

BTW, re: it does not practically matter, anyway, it absolutely does matter, because the question of whether it is or isn't a SPS determines whether it can be used for statements about people who do not work for the company. FactOrOpinion (talk) 21:56, 1 November 2024 (UTC)Reply

That M-W definition seems problematic. By that definition, a tweet isn't self-published unless the person has paid for a blue checkmark, and the transcript of a congressional hearing wouldn't be self-published either, since it's the government paying for the publication. For that matter, WP wouldn't be self-published, as none of us pay to have our edits published. I largely agree with WP:SPS that "Self-published material is characterized by the lack of independent reviewers (those without a conflict of interest) validating the reliability of the content," but it's sometimes difficult to determine whether there's editorial oversight. For ex., there are a large number of people posting things to a university website, and some of the content has editorial oversight and other content does not. FactOrOpinion (talk) 19:47, 1 November 2024 (UTC)Reply

Well, I'd rather say that the M-W definition is focused on the idea of producing a book on paper.

I don't think that a university website has "independent" editorial control. There are the parts with no editorial oversight, and there are the parts written by the PR (and fundraising and recruiting) departments. The kind of editorial control that makes a reliable source (and a traditional publisher) is much more likely to be found at the student newspaper than anywhere on the university's own website. WhatamIdoing (talk) 20:09, 1 November 2024 (UTC)Reply

Currently I think there's two quite divergent opinions here, depending on what angle you come from.

• Narrow definition of published - something has to have some sort of recognisable, established publishing structure and a measure of independence or accountability. Book and journal publishers have independence, news publishers have some sort of accountability (corrections and complaints procedure and policy, regulatory body, something like that). Anything less than that is self-published.
• Narrow definition of self-published - something has to be published by an individual, with no oversight. Anything greater than that is not self-published.

Both perspectives lead to pathological cases in the middle (is a magazine self-published by the former definition? is a guest post on a blog not self-published by the latter?).

Personally, I would favour a narrower definition of published, and to consider everything that falls short to be self published, simply because it is the level of external accountability and oversight that a formal publishing structure gives that protects individuals from propagation of third party BLP claims on wiki.

Perhaps debating what "self-published" means is less important than deciding what is "sufficiently published" for a third party BLP claim. Void if removed (talk) 19:46, 1 November 2024 (UTC)Reply

@Void if removed, do you mean, for the first, "traditionally published"? We already have a definition of Wikipedia:Published that encompasses both the traditional publishing houses and self-publishing. WhatamIdoing (talk) 20:10, 1 November 2024 (UTC)Reply

Yes I would say so. And specifically I want to know what exactly we understand to be: "published with sufficient independent oversight and accountability for a BLP claim about a third party". Right now opinions differ, but this is a crucial distinction in practice. Void if removed (talk) 20:56, 1 November 2024 (UTC)Reply

Basically, the former approach says "BLP claims should only come via traditional publishers" which fits with the spirit of the guidance there.

The latter says "practically anything other than a random personal blog is in theory acceptable for a third party BLP claim", which IMO does not. Void if removed (talk) 21:11, 1 November 2024 (UTC)Reply

The opposite opinion is that the latter says "Sources that have editorial oversight aren't self-published", which matches the spirit and wording of the guidance.

BLP usage doesn't rely on this single aspect alone, as it has other requirements. So trying to use this to ban all bad sources just confuses the situation. -- LCU ActivelyDisinterested «@» °∆t° 21:21, 1 November 2024 (UTC)Reply

I think that "BLP claims should only come via traditional publishers" is the intention behind BLPSPS. Also, BLP would have a narrow definition of "traditional publisher" (e.g., excluding the government public records). WhatamIdoing (talk) 21:40, 1 November 2024 (UTC)Reply

I would agree with this and I think it is a straightforward interpretation. BLPs are supposed to be conservative. For 3rd party BLP claims, only once you clear the first hurdle of "is it a traditionally published source" do you then consider independence, secondary, due etc. Being traditionally published is a necessary - but on its own insufficient - first step. Void if removed (talk) 00:18, 2 November 2024 (UTC)Reply

Editorial oversight by that measure is meaningless. Any celebrity social media account has editorial oversight. Any publishing where one other person signs off counts. This is back to my first pathological case - a guest post on a blog.

And I'm aware there are other considerations beyond this - but no 3rd party BLP claims from SPSs is a very clear edict, so I would expect "what is an SPS" to be similarly clear, yet it appears not to be. Without that clarity, the rule is meaningless.

If it really is, essentially, single person blogs and not much else, that should be clearer than it is. Void if removed (talk) 21:43, 1 November 2024 (UTC)Reply

"Not an established or traditional publisher" is begging the question. You might as well say "it's not self-published if someone I respect published it".

There's simply no consistent way to do what you want to do. And this is Wikipedia policy, so even if I accepted that third-party sources agreed with you (they don't) we would be under no obligation to follow them. What a neutral point of view is on here is very different from the dictionary definition of "neutral". What is verifiable or notable are jargon definitions unique to us. Wikipedia policy is decided by consensus, not sourcing. Loki (talk) 21:21, 1 November 2024 (UTC)Reply

It is often true that we make up our own wikijargon, but that is not the intention for these particular words.

I would be interested in seeing the sources you would rely on. WhatamIdoing (talk) 21:37, 1 November 2024 (UTC)Reply

I think you have confused "established or traditional" with "reputable". So, for clarity, DMG Media, of WP:DAILYMAIL, is a traditional publisher. Larry Flynt Publications, publisher of Hustler porn magazine, is a traditional publisher. DC Comics, promoter of Superman, is a traditional publisher. Being a traditional publisher is about the business model. It's not about whether the publications you produce are accurate, educational, useful, interesting, or reliable. WhatamIdoing (talk) 22:06, 1 November 2024 (UTC)Reply

I responded to the post by 100.36.106.199 at 13:38, 1 November 2024 (UTC), a supposed quote from the Chicago Manual of Style, by correcting endnote 1 in the project page to show that this quote actually comes from "College of St. Catherine Libraries Guide to Chicago Manual of Style" (DEKloiber, December 1, 2003). Whether the material in that source is an exact quote, or even close paraphrase, from Chicago, I don't know. Jc3s5h (talk) 01:05, 12 November 2024 (UTC)Reply

Here's a table that might be useful:

I'm not sure that it's useful to list the NYT example, because newspaper publishing in the US has a tradition that could be argued either way, without changing the end result. But this should give you the idea. (Note that I've declared certain government reports/gray literature to be traditional publishers. I think that's the right thing to do, but if the community disagrees, then we'd have to change that.) WhatamIdoing (talk) 19:21, 1 November 2024 (UTC)Reply

Once again this is baseless and factually inaccurate, the corporation is the publisher, the corporation's resources are doing the publishing, and it has to be the publisher as it is liable for the website of the corporation.

And your arguments actually serve no purpose, except perhaps you want to self publish your own thoughts in an essay. This is on the verifiability page for a reason, and the reason is small, use in BLP's -- your arguments have nothing to do with reliability and barely anything to do with NPOV or OR. Literally, you fall back on an irrelevant to policy logical fallacy of appeal to authority (so called, tradition.) -- Alanscottwalker (talk) 20:33, 1 November 2024 (UTC)Reply

The corporations emoloyees ARE the corporation's (human) resources. The employees answer to the publisher. Therefore, it's self published. This is NOT the case with newspapers: the authors' work is the responsibility of the editor, and the publisher does not meddle with the editor's work. The publisher CAN'T, as that would be considered improper. Therefore, not self published.

And that is the difference. 73.2.106.248 (talk) 21:04, 1 November 2024 (UTC)Reply

No. Absolutely not. The resources of the corporation are not the employee's resources, nor is the web page the employee's. It is exactly like the same as the newspaper, in every way that matters, the author and the publisher are two different things, both liable. Alanscottwalker (talk) 21:08, 1 November 2024 (UTC)Reply

I can see that you're very committed to your view, but I wonder whether you can see the other viewpoint. For example, think about how editors talk about the Wikimedia Foundation. Have a look at https://wikimediafoundation.org/. Do you look at that and say "website designed by Greg, this page written mostly by the fundraising team, that section written mostly by one of the lawyers, this file was posted by Joe – see, none of that was written by 'the WMF', and none of it was published by 'the WMF', either"? Or do you look at that and say "Yup, that whole thing was written by the WMF and published by the WMF"? WhatamIdoing (talk) 21:29, 1 November 2024 (UTC)Reply

FWIW, I think WP:SPS is very broken and does not accomplish the thing it wants to, but yeah, clearly the WMF webpage isn't self-published either. The author and the publisher are different people.

Our issue here is that what self-published means and what we need it to mean are different things. What we want is editorial oversight and what we say we want is that someone who is not the author published the work, which are not the same thing no matter how much we want them to be. Loki (talk) 21:34, 1 November 2024 (UTC)Reply

Upon thinking about this more, I don't think my issue is with WP:SPS per se, it's with the definition in WP:USESPS, which is useless even without the line I suggested we strike. The definition of "self-published" is not whether the author and the publisher are different. For proof: who's the publisher of a self-published book? It's not the author, right? It's a company that self-publishes books. This is as different from the author as any other publishing company: the thing that sets them apart is not they are the same entity as the author but that they offer no editorial review over the author.

By the definition in WP:USESPS taken literally, YouTube and other social media would in fact be some of the only material on the internet that isn't self-published, because the author is an individual not employed by YouTube, and the publisher is YouTube. Loki (talk) 21:44, 1 November 2024 (UTC)Reply

You seem to have confused publisher with printer (or 'host', for digital media). The publisher is the person who decides that the book will see the light of day. The printer is the one who is paid to print whatever the publisher decides to have printed. WhatamIdoing (talk) 21:47, 1 November 2024 (UTC)Reply

Really, WAID? Your the one who posted a long wikispeak brief based on false claims and employing logical fallacies, so it's rather whether you can see it another way. As for your example, yes it appears to have a publisher the WMF, and have writers, who are not self-publishing. The WMF would be liable as publisher, unless it has a defense or immunity. The author's would be liable as author's and the reviewers would be held responsible for what goes wrong, whether they work at a newspaper or other company. It's nothing short of how out-of-touch your argument is when you go on about all the people involved in putting that page out, the more people involved is generally what we think of as great in a source. -- Alanscottwalker (talk) 22:46, 1 November 2024 (UTC)Reply

Alan, where in your definition(s) of self-publish did you find any reference to legal liability?

I can see the POV in which any entity with enough internal control isn't self-published. I don't agree with it as a basic/non-niche definition, and I don't think it is suitable for Wikipedia's purposes, but I understand that some people have a narrow definition. In their model, single-person personal blogs are always self-published, and once you get above a certain size, then it starts being a publishing business. I've never seen any editor try quantify the size – the closest we've gotten is "if they have enough lawyers", usually with this not-in-any-definition value of minimizing libel risk.

BTW, when we were both new editors, the definition in according to ArbCom was "A self-published source is a published source that has not been subject to any form of independent fact-checking, or where no one stands between the writer and the act of publication." BLPSPS took its present form shortly after that case.

That definition exempts traditional publishers (news orgs and similar publications use independent fact-checking; peer review and traditional book publishing processes have someone standing between the writer and publication). However, it can only be seen as exempting corporate websites if you think that coca-cola.com is "independently" written by Emily Employee (after having been not-so-independently told to write it), and that Cal Corporate (the same boss who told her to write it) is somehow going to prevent Emily from doing what the boss told her to write. WhatamIdoing (talk) 23:30, 1 November 2024 (UTC)Reply

I responded to most of this below,[7] in short, you have misrepresented the definition of self publishing. And to answer your first question, no I was not up there talking about definitions -- the law discussion represents the fact that the corporation is the publisher. Alanscottwalker (talk) 12:01, 3 November 2024 (UTC)Reply

The idea that "PepsiCo employees" are the publisher of pepsico.com is beyond bizarre. A corporate entity is not coextensive with its employees. --100.36.106.199 (talk) 21:05, 1 November 2024 (UTC)Reply

Can you think of a way for information to appear on pepsico.com that doesn't involve an employee taking some action? Remember, we're working with a definition of published that means "made available to the public", and not, e.g., "accepting legal liability for the contents". WhatamIdoing (talk) 21:46, 1 November 2024 (UTC)Reply

I can't think of a way for the US census to appear which doesn't involve USCB employees taking action, or the NYT to be printed without NYT employees taking action, but you haven't listed "USCB employees" and "NYT employees" as the publishers in those cases. I doubt that A.G. Sulzberger personally takes any action to publish individual editions of the NYT, yet you have credited him as the publisher there. Caeciliusinhorto (talk) 09:06, 2 November 2024 (UTC)Reply

I credit Sulzberger as the publisher because the reliable sources do. As I noted above, the US newspaper tradition around publishers is sufficiently quirky that it may not be a good example. WhatamIdoing (talk) 17:43, 2 November 2024 (UTC)Reply

Why do you credit Pepsi employees as the publisher rather than PepsiCo? The more I think about this, the more I puzzle over how one determines who the publisher is. Is the publisher the person who makes content viewable on a website or submits it to a printer? If so, then the publisher is an individual and not, for example, an entity like Oxford University Press, and Sulzberger would not be the publisher of NYT articles, and you might have different people identified as the publishers of the same content if it's published both online and in hardcopy. Is the publisher the entity that would be held legally liable for libel? If so, then the publishers of PepsiCo and NYT content are likely PepsiCo and the NYT Co., not the employee or Sulzberger (though perhaps I'm wrong, and both he and the NYT Co. would be defendants). Is the publisher the person identified as publisher on a masthead? If so, the the publisher of the NYT is Sulzberger, but PepsiCo doesn't have a masthead, so would it instead be the CEO? Is the publisher the holder of the copyright? If so, that would again point to PepsiCo and the NYT Co. as the publishers. FactOrOpinion (talk) 18:42, 2 November 2024 (UTC)Reply

You're right; I should ignore the fact that the corporation's actions are implemented by the employees as a practical matter. I would credit PepsiCo as the publisher of their corporate website for the same reason that I put |publisher=Oxford University Press in a citation template, instead of |publisher=Whichever unnamed employee and/or committee at Oxford University Press gave the final approval for publishing this book.

We could go further down the rabbit hole: If the publisher is not OUP, but an individual employee, then is it "really" the person who approved it, or the assistant who did all the necessary paperwork? It couldn't be made available to the public without the actions of the printer, so maybe it's the printer. Or the ship that carried the hard copies from China to the Port of Los Angeles. Or the truck driver who took the container to the wholesale distributor. Or the shipping department at distributor's warehouse, because they shipped a box of books to a bookstore, and the public can't get it if it doesn't reach a retail establishment. Or maybe it's the UPS driver who wheeled the box into the store. (I must remember to tell my driver; he'll be tickled to learn that he's a publisher now.) Or, "really", it must be the bookseller, because how else would the public actually get their hands on the book, if nobody sold it to them? So there you have it: the low-paid employees of every bookstore in the world publish all the books they sell. We leave it to the lawyers to determine whether their action in publishing so much garbage should be considered joint or severable. WhatamIdoing (talk) 19:58, 2 November 2024 (UTC)Reply

Is it possible that a distinction we've glossed over is that individual employees aren't generally identified or credited as authors of PepsiCo publications? If Alice Pepsico doesn't get to publicly exist, we should only evaluate the context of her work as presented, without assuming an editorial distinction that is more hypothetical than usual. Of course, many news articles have anonymous or mass bylines—but that matters, doesn't it? I think there is a case we should treat a piece credited merely to the NYT editorial board as being "slightly more self-published" than one with identifiable, credited authors. Remsense ‥ 论 11:11, 2 November 2024 (UTC)Reply

IMO we could make a rational argument for anything in the NYT that's written by Sulzberger as being "self-published", though in the case of the NYT editorial board, they're going to be running their opinion pieces past Sulzberger in at least a general way. (See also the recent kerfuffles about certain newspaper owners vetoing presidential endorsements.)

I don't think that bylines matter. Self-publishing vs traditional publishing is about the publication process rather than the qualifications of the author. WhatamIdoing (talk) 17:55, 2 November 2024 (UTC)Reply

Let me show you the issue with your table:

Loki (talk) 21:29, 1 November 2024 (UTC)Reply

I don't think that's helpful. First, if we assert that "Alice" is the author – that's fine; some things really are written by a single person – then who exactly puts Alice's content on the website? An abstract corporate entity can't upload files to a computer any more than it can write a paragraph. So why would we say "Alice is the sole author, not the company in general" but not say "Bob is the sole person making the content available to the public ("published" according to this policy), not the company in general"?

Second, since "traditional publisher" is a concept that comes from reliable sources, why is it subjective, and why do you think subjectivity is bad? WhatamIdoing (talk) 21:57, 1 November 2024 (UTC)Reply

I don't think that's helpful Yes but maybe at some point you should stop repeating yourself and try to understand a little bit why hardly anyone agrees with your point of view? 100.36.106.199 (talk) 01:23, 2 November 2024 (UTC)Reply

I think that quite a lot of editors agree with my view. Several of the people in this discussion today, including yourself, seem to be focused on a situation in which Hard cases make bad law: they want a particular result (GLAAD's GAP pages can be cited for BLP material), and although there are multiple paths they could take to make that happen, the only path they have explored is to reject the real-world definition of self-published and change the on-wiki definition. WhatamIdoing (talk) 01:59, 2 November 2024 (UTC)Reply

Agree... I think the contentiousness of GLAAD being used to label someone a transphobe or homophobe is possible reason for fuel to fire.

I think similar situation with potentially opposite leanings is using Society for Evidence-Based Gender Medicine to declare a "gender-critical" doctors as leaders in the field, and gender-affirming doctors as, whatever insult they can think of.

I don't think I still agree on Whatamidoing SPS definition, but folks should def consider that Whatamidoing is right that if grey lit/advocacy is not SPS, we cannot use WP:BLPSPS to disqualify a think tank/advocacy group that we do not like, and that it opens up every think tank/advocacy group to say whatever about a person, unless if it get reposted by CNN/NYTimes or another traditional publisher. Bluethricecreamman (talk) 02:09, 2 November 2024 (UTC)Reply

Well, you're basically right: Whatever rules we put out for the Honest Brave and True Organization will also apply to the Scurrilous Knaves Society. If we say that GLAAD is non-self-published because their staff use an editorial process to decide what to post, and next year SEGM tells their staff to use the same editorial process to decide what to post, then both of them would be exempt from BLPSPS. We shouldn't, and probably can't (because editors will revolt), have a rule that actually says "This publication process is non-self-published if I approve of the source's POV but self-published if I don't". WhatamIdoing (talk) 02:26, 2 November 2024 (UTC)Reply

SEGM would still be fringe. I don't know, WhatamIdoing, I feel like this argument really has more to do with whether we consider these sources reliable rather than whether we consider them self published, and self-publication is only one factor there. Alpha3031 (t • c) 02:30, 2 November 2024 (UTC)Reply

Correct, that's what we have the separate reliability criteria for. And even today, there are some editors (and often brand-new SPA accounts that pop up and regularly get banned or blocked) that try to push WP:FRINGE content, some of which does even sometimes get coverage on otherwise RS media, but that's where our other reliability criteria come in to stop such content from failing our other NPOV policies. And while sometimes such editing has gone unnoticed for longer (such as the case from a few years ago); by and large, typically our editor community is able to discern it and keep such misinformation in check, earlier or later (though earlier is preferred). Raladic (talk) 03:07, 2 November 2024 (UTC)Reply

Agreed with all of this. WP:SPS is just one part of our overall reliability guidelines. Loki (talk) 05:22, 2 November 2024 (UTC)Reply

On what basis? Right now this is a subjective assessment. There are no objective sources declaring this to be forever and always true.

Consider instead a large powerful advocacy org the Heritage Foundation deciding to call a prominent figure something inflammatory and damaging for their political stance on some hot button issues.

Right now such a thing is unlikely to be added to a BLP because of all sorts of subjective interpretations of rules by editors. It isn't due. They're biased and unreliable etc. But those subjective interpretations can change. I'd ask editors who confidently believe their own personal views about a subject to be the majority, to consider a situation where they are not.

Whether something is traditionally published or not is the only objective test between an inflammatory claim, and a 3rd party BLP. If that criteria for inclusion is watered down and subjective too, then there's no serious barrier. IMO BLPs are too important for the subjective views of editors to govern the entire process, and forcing deference to traditional publishing structures that provide some level of external quality control on this specific matter is in the spirit of BLPSPS. Void if removed (talk) 06:50, 2 November 2024 (UTC)Reply

There is no way to cut subjective judgments out of this. The Daily Mail is a traditional publisher, but it's deprecated because people have judged it to be generally unreliable. FactOrOpinion (talk) 15:05, 2 November 2024 (UTC)Reply

Yes, but if we say that advocacy orgs are not "a publishing house" in the meaning of the dictionary definitions for the word self-publish, then we would ban that use on both subjective ("it's not DUE") and objective ("besides, it's self-published, so you have to find a non-self-published source that quotes them") grounds. WhatamIdoing (talk) 17:58, 2 November 2024 (UTC)Reply

Are you sure that there's nothing subjective about determining whether a publisher is a traditional one? For example, you wrote that the Census Bureau is a traditional publisher. Is (government)=(traditional publisher) true for all government publications, and if not, how do you figure out when it's a traditional publisher and when it isn't? Or consider a university's website, would you say that all of it is self-published (because the only part that might be a traditional publisher is a university press, which is now generally split off as a separate entity, like Univ. of CA Press), and if not, what parts aren't? Does your exemption that an employer's website can be used for statements about its employees mean that a university's website can be used for statements about current faculty, but not for statements about students, unless they're student employees? FactOrOpinion (talk) 21:08, 2 November 2024 (UTC)Reply

They previously stated, and I would agree, that it is partially subjective—or rather, there is no single definition applicable in all cases. I do not see how it is any more or less subjective of a judgment call than any other we have to make in assessing the reliability of sources. Remsense ‥ 论 21:44, 2 November 2024 (UTC)Reply

I agree with Remsense, and FOO, I give these specific answers to your list of questions:

• Yes, you're going to have to use some good judgment to figure out whether a given government agency, or a given publication by that government agency, should be considered a traditional publisher.
• You determine whether an entity is a traditional publisher by looking at their main business (e.g., Oxford University Press does little else; United States Department of Defense does quite a lot that isn't publishing) and by looking at their publication process (e.g., Do they have a reputation for responsible publishing practices that is similar type of undisputed publisher?).
• Yes, I would start with the assumption that everything in a university website is self-published.
• I would say, wrt to the university website, that ABOUTSELF applies to everyone who belongs to the university in some fashion, including administration, faculty, staff, students, and guests, but not people unconnected. For example: if they post a note about Joe Film speaking at the graduation ceremony, then that's self-published but acceptable, but if they post an otherwise identical note saying that Paul Politician, alumnus of Some Other Place and representing a different district in the legislature, voted the wrong (or right) way on their budget, then that's self-published but unusable (for statements about him).

WhatamIdoing (talk) 00:05, 3 November 2024 (UTC)Reply

Thanks for answering my questions. FWIW, many federal government documents are published by the Government Publishing Office. The GPO strikes me as a traditional publisher even when it's printing DoD documents.

Re: ABOUTSELF, I clearly don't understand it well enough, in particular, who counts as a third party when it comes to interpreting "it does not involve claims about third parties." If a professor's statement about a guest speaker's work is not a claim about a third party, that's a much more expansive interpretation than I'd been using. Consider a congressional hearing transcript, which I consider a SPS by Congress, as it has no editorial oversight even though it's published by the GPO. If a Rep. makes a statement about one of the people testifying or about another Rep., I would have thought that that's a claim about a third party and disallowed, but it sounds like it falls under ABOUTSELF for Congress, since the people involved are members and a "guest" of Congress. FactOrOpinion (talk) 04:05, 3 November 2024 (UTC)Reply

I think that 'guests' is debatable; reasonable people could disagree. However, if you {{cite press release}} for something like "Joe Film was invited to speak at the graduation ceremony", I wouldn't expect editors to object even though it's self-published.

I would also lean towards considering Congressional transcripts as self-published (because nobody realistically has any editorial control over what's printed in them). I assume that Question time in the UK has the same challenge: the clerks are required to report whatever was said, and if someone says something regrettable, well, that's too bad. There is no editorial process for the staff to decide to omit things. In the US, elected legislators can petition the chair for permission to 'revise and extend' your remarks, so that you can speak for 90 seconds and have a much longer and more coherent-sounding speech printed.

Because transcripts are primary sources, and because editors generally have good sense if they're not trying to sling around shortcuts in an effort to 'win', I don't think that agreeing on the exact classification for legislative transcripts is actually important. (Specifically for hearings, they're also WP:Interviews.) WhatamIdoing (talk) 04:24, 3 November 2024 (UTC)Reply

I would say, having revisited the past discussions on SBM, that's my reading of what happened in that case. The precedent set by considering a group blog not self published as long as it was debunking quackery is well-meaning but I think both unnecessary and unsound. It could be argued that the sponsorship of SBM by the NESS makes it not self published, but since both of these are directly controlled by the same people - and that those people can publish with no prior review - that is nothing like a traditional publishing arrangement. Certainly the last discussion started by calling that whole classification into question. By my reading there were some good points in favour of it being reconsidered as SPS that mostly went unanswered, but again responses ended up conflating RS with SPS when these are separate issues and no changes were made.

https://en.wikipedia.org/wiki/Wikipedia:Reliable_sources/Noticeboard/Archive_351#Science-Based_Medicine

We've now got a situation where the argument really is that people are considering internal review and approval mechanisms of any kind to be sufficient to not be an SPS and to update SPS accordingly to retrospectively account for SBM, but I don't see how the arguments for SBM don't also apply to something like The Free Press. I don't think 3rd party BLP material on the Free Press should be a subjective local consensus away from inclusion on a BLP, the policy in BLPSPS becomes largely meaningless in that situation. Void if removed (talk) 10:35, 2 November 2024 (UTC)Reply

I'm fairly sure prior consensus at RSN was that Bari Weiss's substack newsletter was no longer a SPS. They seem to want to transition to being a NEWSORG anyway (they now have a media company and everything!), so they would be exempt from being considered an SPS even under the broader, objective definition that you're proposing, as far as I can tell. Alpha3031 (t • c) 10:50, 2 November 2024 (UTC)Reply

Not what I see in the discussions, pretty clear opposition to considering it anything other than an SPS, and here too.

And if they transitioned to a traditional publishing structure, then obviously they'd be a traditional publisher, but right now that's not the case. Void if removed (talk) 11:53, 2 November 2024 (UTC)Reply

My mistake, Void if removed, it does appear that it's closer to no consensus, I had only really paid a little attention while it was happening and I suppose I must have perceived the people arguing that it was no longer self published as receiving more support than they did. You seemed to have missed the Archive 425 (February 2024) discussion though, and I believe that one was the one that most informed my recollection: Notably, most people opposing use in that discussion focused on GUNREL in general rather than whether it was specifically a SPS.

I also really don't see how your excluding them under your definition, could you elaborate on how you see their publishing structure, versus, say Vox or HuffPost? Alpha3031 (t • c) 14:02, 2 November 2024 (UTC)Reply

Vox is published by Vox Media, the mass media company that publishes titles like New York Magazine, and a bunch of others. There is a separation between who commissions, authors and oversees the content of the title itself, and who takes legal responsibility for publishing it. Similarly, Huffington Post is published by Buzzfeed Inc.

What makes them not self published is not quality or reliability or independence or editorial oversight (all of which are quite questionable in some of these cases) but the existence of separation of concerns between creator and publisher. That a publisher is willing to take responsibility for the published material provides some measure of confidence in the editorial process. Sometimes what a publisher does - particularly in new media - is not terribly visible or obvious to someone external in the way it is with print media, but it can encompass investment, legal responsibility, marketing and distribution.

Determining that the creator and the publisher are not the same person or entity is all that's required to decide if something is a SPS. The Free Press, by contrast, publishes its own content itself.

Being a non-SPS doesn't automatically put Vox on a par with a scientific journal or an academic textbook publisher. It just places it in a different category of source, that can still be unreliable.

The mistake I think is in believing that what is in effect a blog or website which accepts posts from contributors becomes an external publisher for that content, when it is not. The blog in its entirety is the creation - the posts within it are not themselves separate, divisible publications. In exactly the same way, the publisher for the entire NYMag website and all articles on it is Vox Media. That the owner and creator of a blog sources and collates content that may have been written by other people does not make them an external publisher of that content, rather they are still self-publishing their own creation in the form of a singular blog or website. Just as the owner and creator of a corporate website self-publishes the site in its entirety, from mass contributions from its employees, whose work product the company owns. Just as the website of a non-profit advocacy org is the entire self-published creation of the org itself. Void if removed (talk) 17:35, 2 November 2024 (UTC)Reply

Barri Weiss's The Free Press supposedly has a company that takes legal responsibility for publishing it too. Is the difference for you that they're both owned by Weiss? Alpha3031 (t • c) 01:16, 3 November 2024 (UTC)Reply

No. Having a corporate entity does not make the results non-self-published.

(Incorporation is a method of limiting liability: if you, acting as an individual, post something inappropriate on the web and lose a lawsuit over it, they can take your house, your car, your bank accounts, your dog, etc; if you, acting as the employee of a corporation, post exactly the same thing, they can only take the corporation's assets, not your own personal money.) WhatamIdoing (talk) 01:52, 3 November 2024 (UTC)Reply

Restating my question again, since you keep ignoring it: here are the quotes currently in the footnote at WP:V that makes the conflation you're defend: The University of California, Berkeley, library states: "Most pages found in general search engines for the web are self-published or published by businesses small and large with motives to get you to buy something or believe a point of view." (my emphasis); The Chicago Manual of Style, 16th Edition states, "Any site that does not have a specific publisher or sponsoring body should be treated as unpublished or self-published work." Why do you think these expert guides specifically distinguish these situations if they're really just the same thing? 100.36.106.199 (talk) 15:48, 3 November 2024 (UTC)Reply

I don't think they distinguish these situations. There's no reason to believe that these sources are using the word or exclusively (i.e., to mean that it is impossible for something that is self-published to be something published by a business, or that it is impossible for something published by a business to be self-published). If you read or the way we use it in MOS:ANDOR, then that's an and/or, and they're both.

The CMOS page says "should be treated as" which is not the same as "is". (Also, they're apparently using a definition of "unpublished" that includes material that has been WP:Published.) WhatamIdoing (talk) 04:34, 4 November 2024 (UTC)Reply

So your answer is that you think they explicitly distinguish them because actually they want readers to think they are the same, or that one is a special case of another? I would think that authors of such documents would be familiar with phrases like “or, in particular”, but I guess I am less talented at ESP than you are. 100.36.106.199 (talk) 12:06, 6 November 2024 (UTC)Reply

I have decided that saying a microbusiness website that is written and posted by a single person ("Hello! I'm starting a new business, and I hope you will be my first customer") is obviously self-published, and therefore it would be illogical to say that "self-published or published by businesses small" means that small business can never self-publish their advertising materials. You, of course, are welcome to think that when someone puts up a website saying that they love coffee, that's self-published, but when the same individual puts up a website saying they sell coffee, that's non-self-published, but I will not agree with you. WhatamIdoing (talk) 16:43, 6 November 2024 (UTC)Reply

To be clear: the UC Berkeley page from which that quote was taken was entirely about evaluating websites and not about any other kind of material, nor did the page make any attempt to define "self-published." The current corresponding page does not contrast self-published with businesses of any kind. FactOrOpinion (talk) 18:12, 6 November 2024 (UTC)Reply

The current one says:

• Were there any apparent barriers to publication?
  • Was it self-published?
  • Were there outside editors or reviewers?

I think that the CMOS quote (which is actually from a PDF that says "College of St. Catherine Libraries Guide to Chicago Manual of Style", so this might not be in CMOS at all) and the UC Berkley page are not really helping, and I'm inclined to remove them. WhatamIdoing (talk) 20:22, 6 November 2024 (UTC)Reply

To be honest, I'm not really sure we should see the CMOS as authoritative on a definition of self-publishing that's useful on Wikipedia. I would expect the definition of self-publishing in style guides to focus more on the format of citations to those publications. Alpha3031 (t • c) 11:08, 5 November 2024 (UTC)Reply

These sources are drawn directly from footnote 1 in the policy; I did not put the, there. 100.36.106.199 (talk) 12:03, 6 November 2024 (UTC)Reply

Alpha, I suspect that you're right, and that "treated as" means "the footnote should be formatted the same as". @SMcCandlish, do you have a copy of CMOS handy? Can you tell us whether the quoted phrase is about how to format the citation? WhatamIdoing (talk) 17:07, 6 November 2024 (UTC)Reply

I do. The material is actually about how to assess the source, not just about how citations are formatted. From 17th edition:
14.14 Authority and permanence. Much as they do for printed publications, authors must weigh the authority of any electronic sources they choose to cite. Electronic content presented without formal ties to a publisher or sponsoring body has the authority equivalent to that of unpublished or self-published material in other media. Moreover, such content is far more likely to change without notice—or disappear altogether—than for­mally published materials. On the other hand, self-published material from an authority on a given subject can usually be relied on. Authors should note that anything posted on the internet is "published" in the sense of copyright and must be treated as such for the purposes of com­plete citation and clearance of permissions, if relevant (see 4.2, 4.64-69).
This is reasonable enough as observation, but the point above is correct that, basically, CMoS doesn't determine how WP weights sources. No third-party publication does that for us, in whole or in part. The editorial consensus of our community is what does that, through this policy and related P&G material (WP:RS, WP:NOR, WP:MEDRS, WP:FRINGE, etc.). We may take into consideration any/all such third-party publications' ideas during our internal consensus-formation processes, but it is not possible for our consensus to be "trumped" by something said in CMoS or any other work. CMoS isn't even about research assessment of the reliability of published claims; it's a style guide, pure and simple. So, it is entirely irrelevant.  — SMcCandlish ☏ ¢ 😼  00:26, 12 November 2024 (UTC)Reply

Thank you, @SMcCandlish.

Would you recommend removing that particular part of the footnote? WhatamIdoing (talk) 06:25, 12 November 2024 (UTC)Reply

On more reflection and on reading the above mega-thread in more detail, I think not, because the purpose isn't to determine policy, but to illustrate or elaborate upon a principle, using pretty reliable editorial sources for doing so. There's no question that Chicago University Press has to make this sort of judgment on a regular basis and they have a century+ of experience at it, so their editorial view on the question is pertinent, even if it happened to be found inside their style guide. The issue of WP policy not being set by external publishers isn't as central as I thought at first, because we're asking editors to consider CUP's editorial experience on an assessment approach, not consider their definition of something to be forcibly normative on WP. All that said, it's possible the consensus of some RfC[s] eventually coming out of this discussion could render a WP definition/determination about something to do with this which then ended up incompatible with CUP/CMoS's approach, in which case the footnote would need to go. But probably the entire thing; we would have no need to keep one semi-random publisher's opinion on such a matter and excise the other.

In going over this subsection more closely, I find that WhatamIdoing's "How to determine whether a source is self-published" table is much closer to objectively accurate than Loki's (objectively in the sense of accurately capturing actual en.WP community practice). Loki's version would be quite erosive of our ability to hold any source to be "self-published", to treat major mainstream publishers as probably reliable, to distinguish between corporate actors (who necessarily act through employee hands, at least until the AI revolution is complete) versus writers with some authorial independence, to treat major academic publishers as being as reliable as newspapers of record, and several other faults. As far as I can see, WAID's version is only problematic in treating a government agency as equivalent to a newspaper of record or a respected academic publisher, when it is necessarily in a category by itself that has to be assessed for special kinds of skew. That'll be more true the less democratic and fact-based the government, an issue that is going to increasingly be a problem for us with regard to American and many European agencies/ministries as it long has been for various others under some shade of authoritarian control. Or to not put too fine a point on it, given the coming Trump administration's xenophobia, racism, voter suppression, and other demographic manipulation and falsehood tactics, and the plan to replace any "resistant" government administrators with Trump toadies, the US Census Bureau almost certainly stops being a reliable source for new publications in January 2025, and will continue to be dubious until no earlier than Jan. 2029.  — SMcCandlish ☏ ¢ 😼  01:58, 13 November 2024 (UTC)Reply

SMcCandlish, thank you for quoting the current CMoS. Reading the heading and the longer quote and then looking back at the footnote reinforced for me that (a) all three of the sources quoted in that footnote are talking only about web content, (b) the longer CMoS quote is focused on the question of authority/permanence and treats the concept of self-published as understood, (c) it includes the idea of a sponsoring body, which seems to capture the kinds of entities (e.g., advocacy organizations, universities, governments, learned societies) where we're trying to resolve when their material should be considered self-published, and (d) if there's a sponsoring body, the web content inherits its authority, but we might or might not consider that sponsoring body to be WP:RS. FWIW, there is an RfC about this as we speak: Should grey literature from advocacy groups and other similar orgs always be considered WP:SPS and therefore subject to WP:BLPSPS?. FactOrOpinion (talk) 16:03, 13 November 2024 (UTC)Reply

I think the "sponsoring body" is meant to cover situations like Blood (journal) being sponsored by American Society of Hematology. I don't think it is meant to cover situations like an advocacy group posts on their own website. WhatamIdoing (talk) 16:36, 13 November 2024 (UTC)Reply

That's not so clear to me. I found a partial copy of the CMoS to skim, and am pretty sure that the text SMcCandlish quoted is part of a section on citations, and elsewhere in the citations section it uses phrases like "the owner or sponsor of the site" (for websites), "sponsor (network, studio, label, etc.)" (for performances), and "sponsorship" (for lectures and papers or posters presented at meetings). It gives examples where entities such as the Microsoft Corporation, the Alliance for Linguistic Diversity, and the Wikimedia Foundation are listed as the owner or sponsor of a website. FactOrOpinion (talk) 19:20, 13 November 2024 (UTC)Reply

If all it takes to make a website non-self-published is for "the owner or sponsor of the site" to exist, then no websites, include single-person fansites, are self-published, because all websites have "an owner". WhatamIdoing (talk) 00:57, 14 November 2024 (UTC)Reply

I concur strongly with WhatamIdoing. FactOrOpinion's interpretation would make every self-publisher a "sponsoring body", which would render the phrase "sponsoring body" a ridiculous obfuscation that serves no purpose. If we know anything about CUP, and CMoS's editors more narrowly, it is that they write with a great deal of precision and would not do that, so they cannot mean that the self-publisher is always a "sponsoring body"; they would have just said "self-publisher". The very phrasing indicates that they mean an established body of some kind with a public reputation and capable of sponsoring research and signficant publication. CUP's tendency to trust something like Microsoft Corp. as a "sponsoring body" is probably further than WP would be willing to go (at leaset on most subjects), but it's clear from the other examples what they mean by "sponsoring body": a prominent organization with a worthwhile reputation.  — SMcCandlish ☏ ¢ 😼  01:26, 14 November 2024 (UTC)Reply

You (pl.) have convinced me that CMoS's editors must have meant something narrower by "a sponsoring body." I'm still uncertain how they bound "sponsoring body," but I'll retract parts (c) and (d) in what I wrote. FactOrOpinion (talk) 02:16, 14 November 2024 (UTC)Reply

if there's a sponsoring body, the web content inherits its authority, but we might or might not consider that sponsoring body to be WP:RS: The second half of that is clearly correct, but the first is already not the case; WP:RSNP is chock-full of examples of blogs with what someone might want to call a "sponsoring body" (e.g. a major news publisher) that are nevertheless deemed categorically unreliable or questionable because their editorial practices markedly differ from those of the main publication(s) of the "sponsoring body". The canonical example is Forbes magazine (and the online Forbes.com equivalent) content written by their editorial staff versus "Forbes Contributor" material by rando freelancers subject to lower editorial control and review standards. That is to say, web content from "sponsoring body" might inherit the authority of that body, if it is essentially equivalent to other material written by that body, and not material for which the body is simply acting as a distribution/publishing conduit with lower editorial standards. A quasi-analogy would be that material that is written and published by YouTube/Google/Alphabet staff at YouTube.com about how to develop content and apps for YouTube (e.g. [8]) is a reliable (albeit primary) source for how to develop content and apps for YouTube. If I create a video about the same subject and post it on YT, that would not be as reliable a source [unless I'm an acclaimed expert on the subject], and YT/Google/Alphabet would not be a "sponsoring organization" of it, since they did not vet that content by the same standards as their own internal documentation output by staff (if they vetted mine at all, other that by automated "AI" detectors of forbidden content). They're just acting as a medium for my self-publication, as with Forbes and its "Contributors".  — SMcCandlish ☏ ¢ 😼  01:53, 14 November 2024 (UTC)Reply

Well, hard cases make bad law cuts both ways. I think that some people, above, are looking for a magical dividing line that will let them discount GLAAD in this context without having to examine its reputation in depth. Clearly the desire for a hard line here is partially driven by the fact that some people feel GLAAD is obviously unusable and that any policy that wouldn't clearly indicate that must be fatally flawed. I don't think that that hard dividing line people are looking for is ever going to exist - ultimately, reputation for fact-checking and accuracy trumps everything else. It's the core of what makes an RS citeable for statements of fact and, when it's strong enough, will override anything we write elsewhere, which means that people will always have to confront a source's individual reputation. I don't even know if GLAAD passes that bar (we haven't really looked in-depth because we've been spending our time quibbling over this instead), but I stand by my assertion that SPS is a category of non-WP:RS, outside of a few limited exceptions; it is literally a category of WP:NOTRELIABLE - the purpose of SPS is to say "these things will usually lack a reputation for fact-checking and accuracy". It's not intended to give us an easy test that can discount a source without regard for its reputation, because no such test can exist. Reliability is contextual, but a strong enough reputation overcomes everything else. --Aquillion (talk) 14:58, 2 November 2024 (UTC)Reply

I think you're right: in the case of GLAAD, this is all being driven by a desire to use the source. If BLP said that only websites with green backgrounds can be used, we'd doubtless have editors talking about the fuzzy Blue-green distinction and that this blue website is really green and should therefore be acceptable.

I don't want to re-write definitions to make this (or any specific) source usable. I say this even though I'm sure that I've cited that website myself and consider it generally reliable. Wikipedia:If a rule prevents you from improving or maintaining Wikipedia, ignore it. If citing GLAAD would improve any article, and BLPSPS prevents us from doing that, then we should ignore that rule and improve Wikipedia. We should not be trying to re-write basic definitions so that corporate, advocacy, and political campaign websites are all non-self-published just because we want to use one (1) website. WhatamIdoing (talk) 18:07, 2 November 2024 (UTC)Reply

I'm not sure its the reason (after all the original page wasn't even a BLP, unless people decide BLPGROUP applies) but its certainly a precipitating factor. After a few years of doing my best here I find the rules arcane and unevenly applied, so when I do my best to read, understand and then cite a policy in a CTOP, I find it quite unsettling to be outnumbered by editors calling my interpretation of the policy wrong. I raised it here because I want to know why, and where I erred.

However the result of all this discussion really seems to be that I didn't err, and that editors are moving to change either the policy or the explanation to retroactively endorse a practice they prefer. Whatever the policy is, all I care about is that it is something that I can at least try and follow in good faith. Void if removed (talk) 20:39, 2 November 2024 (UTC)Reply

Your essay is about Wikipedia articles, I don't think Wikipedia policy is the same thing (though I suppose if we consider "discussions Wikipedia users have about policy" to be a reliable source on Wikipedia policy, we could theoretically make COPO apply to our PAG as well). I do see subjectivity as sometimes necessary in policy, but at present, we don't have to adhere to WP:YESPOV when writing our policies and guidelines, and introducing subjectivity in standards where not necessary can make our guidelines less useful. Alpha3031 (t • c) 02:39, 2 November 2024 (UTC)Reply

All sources are reliable for something, and a discussion about policy would be reliable for certain statements, and such discussions might even be cited in articles such as English Wikipedia Arbitration Committee. If you'd like to make the core content policies apply to policies and guidelines, then you will have to get WP:NOTPART revoked. WhatamIdoing (talk) 18:11, 2 November 2024 (UTC)Reply

My point, WAID, is that YESPOV, and thus your essay about how subjectivity is required to meet that, does not currently apply to our PAG. Alpha3031 (t • c) 01:18, 3 November 2024 (UTC)Reply

I linked to that explanation as an example of why subjectivity should not be considered an inherently bad thing, and maybe even in the hope that some people might read it and learn what that word means (because it's not just about opinions). WhatamIdoing (talk) 01:54, 3 November 2024 (UTC)Reply

GLAAD should be largely discounted as an RS; not entirely, but as self-published advocacy material on the same footing as other such material, i.e. as something we can give WP:DUE consideration to in balance with other viewpoints (which on many issues may be short shrift to the other side, since it will generally lack evidence and be faith-based or otherwise irrational). I say this as someone under the LGBT umbrella, who worked closely with a GLAAD leader (in one of their side roles at another organization). It's not the subject of the advocacy that matters, or the reputation of the advocacy group within the group they are advocating for. It is simply the fact that it is advocacy material. Claims by the ACLU or the NRA about a US constitutional law question should be treated the same way: as socio-politico-legal positioning and opinion, not as factual reportage. Same of course goes for religious bodies. And for governmental agencies/ministries if they have any potential for spin (e.g. to back up their policy decisions or advance the interests of the controlling political party). A passing example would be that the state of California has declared various species and subspecies of things "endangered", yet some of them are quite common and not under any threat (they just turn out to have very small populations within the state due to their natural range barely crossing the state line, while they are all over the place in Oregon or where ever.) CA state publications often are not reliable sources for the conservation level of animal or plant species, not even at a state level, because they tend not to draw a clear distinction between a [sub]species in decline and one that simply barely has a range within the state. And this isn't even a politicized example; wait until you seem what's going to come from US government agencies and those of many "red" states under the Trump adminsitration.  — SMcCandlish ☏ ¢ 😼  01:53, 14 November 2024 (UTC)Reply

I think that one of the challenges in defining GLAAD as non-self-published is the community's preference for even-handed rules that do not overtly engage in viewpoint discrimination. If GLAAD's website is written and published by its employees, and we deem that a non-self-published model for publishing information, then so is any other website that follows a similar publication process, including anti-LGBTQ+ advocacy groups. WhatamIdoing (talk) 02:19, 14 November 2024 (UTC)Reply

Traditional publisher is not an iota less well-defined than publisher. Remsense ‥ 论 02:27, 2 November 2024 (UTC)Reply

It worryingly sounds a lot like "publishers who meet the definition but get a pass in our opinion". I would be happier with something a lot better defined. Would a think tank who changes the wording on their website to say their a news organisation suddenly become a traditional publisher? If not do start up news organisations still count as self-published? What if a think tank setup a new news website to basically publish the same details they have on their own website?

None of this would matter if the definition of what a self-published source is was kept simple, there are other parts of policy/guidelines that would still need to be assessed on whether the source was reliable. -- LCU ActivelyDisinterested «@» °∆t° 15:21, 4 November 2024 (UTC)Reply

I think that the The Canadian Oxford Dictionary has a simple definition:

• "publish (a book etc.) oneself rather than through a publishing house"

that covers everything we care about, including explaining why a small newspaper isn't self-published even though it's a few people publishing content they write, as well as why an equally small website, e.g., for a prisoner whose friends outside put up a website to publish his writings, is still self-published. Furthermore, a simple "change the website" doesn't get around this definition (because it must actually be a publishing house, not merely claim to be one).

If you are looking for a rule that does not require editors to determine which alleged news outlets are actually news outlets, then I don't think that is possible. Even if one didn't need to make that determination for self-publishing status, you'd still need to make that determination (or an even more complex one) for reliability. WhatamIdoing (talk) 18:55, 4 November 2024 (UTC)Reply

But news organisations don't publish through a publishing house, they publish themselves. So we're back to arelyinh on indefinable "traditional publisher" to separate them. If we're for allowing people to consider what is or is not self-published, they have and it doesn't match what has been put down in USESPS. "Change the website" was one point in a chain, not a point on it's own. -- LCU ActivelyDisinterested «@» °∆t° 19:26, 4 November 2024 (UTC)Reply

News organizations are publishing houses. Publishing the news is their main activity. A news organization that doesn't publish the news is an oxymoron. WhatamIdoing (talk) 20:38, 4 November 2024 (UTC)Reply

Yes, they're publishers... of their own work. That's the issue here. Loki (talk) 08:18, 5 November 2024 (UTC)Reply

Yes: they're publishing through a publishing house (their own publishing house, but it is still "a" publishing house), which means that they are not self-publishing according to a definition that says self-publishing is publishing on your own "rather than through a publishing house". WhatamIdoing (talk) 22:29, 5 November 2024 (UTC)Reply

@LokiTheLiar, @WhatamIdoing, @Alanscottwalker @Alpha3031, I am making an RFC draft based on Alpha's initial wording. It needs significant wording and shortening I think.

What are some ideas on how to clean this draft up? Bluethricecreamman (talk) 21:36, 1 November 2024 (UTC)Reply

Feel free to edit the draft, its very rough and i don't have much experience drafting useful RFCs. I'm going to come back to it later tn for another passthrough myself after reading the discussion a bit more. Bluethricecreamman (talk) 21:38, 1 November 2024 (UTC)Reply

@Bluethricecreamman, I don't think that draft is even close to ready for discussion. For example, you suggest that "some form of formal review process independent of the original author" is the key distinction. Can you tell me whether an article that goes through the Wikipedia:Articles for creation is now "grey literature"? AFC is "some form of formal review process independent of the original author".

For another scenario, imagine that I run the advertising department for a large consumer-goods manufacturer. I hire three ad agencies to produce a 30-second commercial spot. I will put them all through "some form of formal review process independent of the original author" and run the best one during the Super Bowl. Does my advertisement now become "grey literature"? WhatamIdoing (talk) 21:53, 1 November 2024 (UTC)Reply

I was really intending for more people to edit it before anyone actually tried to use it, but... well... Wikipedia isn't a traditional publisher, so by conventional definitions it could be considered grey literature. The current footnote in SPS also has validating the reliability of the content, so that should probably be added, though I think I've been overly focused on what the policy currently says. I am actually inclined to merge the well established reputation part from the second sentence into the first one (so only review processes that have a positive reputation for validating reliability are considered sufficient, and everything else would fall under the minimal or no in the last sentence). Alpha3031 (t • c) 02:28, 2 November 2024 (UTC)Reply

Wikipedia is self-published. You publish your changes, and I publish mine. That's why the big blue button says "Publish".

I don't think the current footnote is the best we can do. For example, "validating the reliability of the content" uses words the conflict with our jargon (WP:Reliability), and it's also not an activity universally undertaken by traditional publishers. (How exactly does one "validate the reliability" of a book of poetry, or a fantasy novel? And yet we cite such works, and not exclusively in WP:PLOTSUM sections.) WhatamIdoing (talk) 18:16, 2 November 2024 (UTC)Reply

Wikipedia is self-published

If you'd be willing to go over my comment again, I did not say that it wasn't self published. Alpha3031 (t • c) 01:05, 3 November 2024 (UTC)Reply

We agreed on that point: You said it was not a traditional publisher, and I agreed that it is self-published. This is an X-and-not-X determination: any document that is not traditionally published is self-published. All published documents must be one or the other (though in borderline situations, reasonable people could disagree over which one it is). They cannot be both, and they cannot be neither. Either someone is acting with independence towards the author/their work, in which case they're using the traditional publishing model, or they're not, in which case it's self-publishing. If you think there's a third category, please provide a name for it (and some sources). WhatamIdoing (talk) 01:48, 3 November 2024 (UTC)Reply

I just took a look at what GreyNet considers grey literature, and others might also want to do that, since the draft RfC is worded in terms of grey literature. The goal of that list is not to distinguish self-published from non-self-published, and it strikes me as a mistake to frame the issue around grey literature generally being SPS or generally being non-SPS. The two options also seem to boil down to text that's already in footnote 1 for WP:SPS: "Self-published material is characterized by the lack of independent reviewers (those without a conflict of interest) validating the reliability of the content." If that's the WP determinant for non-SPS, then it shouldn't be in a footnote, and WP:SPS should make it clearer that (1) some kinds of publications (e.g., newsletters, books, website content) might or might not be self-published, depending on the nature of the editorial oversight, and (2) a given organization might create both SPS and non-SPS content (e.g., both unedited podcasts and edited reports), so the WP editor's task is to assess whether a specific source is/isn't SPS rather than whether a larger entity (e.g., a university website, the federal government, an advocacy organization) is/isn't SPS. WP editors are still faced with figuring out whether there are independent reviewers validating a source's reliability; some cases will be obvious and others won't. FactOrOpinion (talk) 23:56, 1 November 2024 (UTC)Reply

Yeah, agreed not every bit of grey lit is remotely usable for wikipedia citation. As per Grey literature, the wikipedia article indicates that the definition is supposed to be very hazy.

I think its a good catch-all term.

I think I see someone added this sentence If an organisation produces multiple forms of content, the level of review may vary, and sources subject to minimal or no review, or with a poor reputation for fact checking, should be considered self-published and only useful in the same situations as other self-published sources. Bluethricecreamman (talk) 04:58, 8 November 2024 (UTC)Reply

Bluethricecreamman, if you're still thinking about an RfC, my sense is that this discussion has revolved around these two alternatives:

• Unless it is published by a traditional publisher (e.g., news organizations, book publishers, academic journals, record labels), a source is considered self-published. However, a small amount of content published by traditional publishers, such as outside ads and marketing, is also considered self-published.
• The only sources that are considered self-published are those that lack independent editorial review (e.g., individual or group blogs, open wikis, internet forums, comments added to news websites, vanity press books, social media posts, podcasts).

I'm not proposing that this be the specific wording of the RfC, only saying that I'd be inclined to frame the RfC as a choice between these two kinds of alternatives. I'd actually invite the people who take View 1 to word the first and the people who take View 2 to word the second. In the discussion, I'd add the following at the top:

• Issues such as whether a source is reliable, independent, or primary are distinct from whether it's self-published.
• The decision about how to define self-published has significant implications for what falls under WP:BLPSPS.
• Among the kinds of sources that are at issue: electronic and print information from advocacy organizations, think tanks, universities, corporations, governmental entities, professional societies.
• An organization might have both non-self-published and self-published content (e.g., a NYT article vs. comments on that article, a government report vs. a government hearing transcript), so in determining whether a source is or isn't self-published, people should focus on the specific source and not the organization. FactOrOpinion (talk) 19:29, 3 November 2024 (UTC)Reply

I agree this is roughly the split, but disagree that the second choice is quite accurate or adequate as a reflection of the arguments made. The arguments here include that corporate websites are not self published because they are published by the company and written by is employees. This is not independent editorial oversight, and restricts self publishing solely to an individual endeavour. This IMO bears no relation to the real world but that's what's been put forward and it should be worded unambiguously as such if that's what is proposed. Void if removed (talk) 19:53, 3 November 2024 (UTC)Reply

I think that for View #2, different proponents have argued that the material is written by employees but published by the organization (corporation, advocacy group, etc.) or that there is sufficient in-house editorial oversight of publications. For ex., this RfC seemed to focus on the latter. FactOrOpinion (talk) 20:25, 3 November 2024 (UTC)Reply

Even if you strongly believe that the lowly copywriter's work is being reviewed by the marketing head in a way that makes it non-self-published (a view I don't share), I don't think any of us would call that "independent" editorial review. WhatamIdoing (talk) 04:39, 4 November 2024 (UTC)Reply

I don't know how many editors are independent in the no-conflict-of-interest sense that's currently in the Verifiable policy: Self-published material is characterized by the lack of independent reviewers (those without a conflict of interest) validating the reliability of the content. For example, does the simple fact that the editor and copywriter work for the entity create a COI? Does a publisher like Sony Music or Nintendo have someone who serves the function of an editor when deciding what music or games to publish? (Though I just realized that the WP:V text is worded in terms of "reviewers" rather than "editors," so View #2 should likely be worded that way instead.) The potential problem with possible definitions is that they may rely on terms that themselves could be understood in different ways (e.g., who is the "author," who is the "publisher," is the reviewer "independent," what is a "traditional" publisher).

If there's going to be an RFC, I'd rather that the people who hold View #1 (e.g., you and Void if removed) write the text for View #1, and the people who hold View #2 (e.g, Loki, Alanscottwalker, Jc3s5h) write the text for View #2. I no longer know what my preference is, but I know that I want there to be a clearly worded policy, so that most of the time, people are clear about whether a given source is or isn't self-published. If you have suggestions for things to add to the list in "Among the kinds of sources that are at issue: electronic and print information from advocacy organizations, think tanks, universities, corporations, governmental entities, professional societies," that would be helpful. For example, I saw that you'd raised the issue of whether a campaign website is/isn't self-published, and Void if removed introduced publishers of games and music. I've now realized that I need to reword "electronic and print information" so that it includes non-print creative works like music, games, and images. FactOrOpinion (talk) 17:52, 4 November 2024 (UTC)Reply

I wonder if people holding the narrower view would like something else entirely, like "Self-publishing means one (or maybe two) humans writing and publishing something without anyone else being able to stop them. This includes, for example, Donald Trump posting on Twitter but not his campaign staff posting on the campaign website, because they have lawyers and other professionals involved." WhatamIdoing (talk) 20:48, 4 November 2024 (UTC)Reply

I've already posted what changes I think should be made to WP:SPS. If I had to write it as a paragraph to be glommed onto the end, maybe:

Because many sources on Wikipedia are not books, what is meant by a "self-published" source here does not necessarily match how that phrase is used outside Wikipedia. In particular, many sources (such as newspapers) that inarguably publish their own work are not self-published for Wikipedia's purposes, while other sources where the author and publisher are separate (such as, for instance, the ordinary process of self-publishing a book) are self-published for Wikipedia's purposes. To determine whether a source is self-published, rely on whether the publisher and the author both have the authority to independently fact-check the work, not whether the author has literally published the work themselves.

I also think that maybe there should be three options because this is a shift from the position I came in here with. I do think the Trump campaign website and the Coca Cola website are self-published, though GLAAD and most advocacy organizations still aren't. Coke both provides the copy of its own website and publishes it: to the extent that the author of that copy is an individual they do not have independent fact-checking authority. Loki (talk) 08:50, 5 November 2024 (UTC)Reply

In your discussion, if material is published by an organization, is that organization always the publisher? (For example, at the NYT, is the NYT Co. the publisher rather than A.G. Sulzberger?) In some cases, the organization may do no or minimal fact-checking. For example, in peer-reviewed journals, the editor may not fact-check the article but instead rely on comments from outside reviewers, and those reviewers might only fact-check glaring errors and be more concerned with whether the work is sufficiently interesting/impactful/makes good use of existing literature. When publishing creative work (e.g., music, fictional books or movies, games, to a lesser extent editorials), neither the author nor the publisher are concerned with fact-checking. Also, by "independently," do you mean "without a conflict of interest" (text from WP:V, which refers people to WP:COISOURCE for more info)? If so, it's not clear to me why a secondary fact-checker at GLAAD has no COI, but a secondary fact-checker at, say, Microsoft does. (And of course this is an oversimplification; for almost any definition we settle on, a given organization might publish a combination of self-published and non-self-published material, as when the NYT publishes ads.) FactOrOpinion (talk) 19:49, 5 November 2024 (UTC)Reply

In your discussion, if material is published by an organization, is that organization always the publisher?

The way you've phrased it, tautologically yes.

For example, in peer-reviewed journals

Peer review is always enough to satisfy this. The two fact-checkers are the reviewer(s) and the author.

When publishing creative work (e.g., music, fictional books or movies, games, to a lesser extent editorials), neither the author nor the publisher are concerned with fact-checking.

This is true, and fiction is weird and special here, in that it can only really be a source on itself. I don't think WP:SPS should apply at all to fiction, because we're not gonna use Harry Potter or Homestuck as a source for facts, and we can equally use Harry Potter or Homestuck as a source for details inside the fiction, so whether it's self-published makes no difference.

Also, by "independently," do you mean "without a conflict of interest" (text from WP:V, which refers people to WP:COISOURCE for more info)?

No, not really. I meant "independently of each other". If they both have the same COI they wouldn't be independent. But, for instance, if you publish an ad in a place that fact-checks its ads, that's not self-published even though it's still not a reliable source for WP:INDY reasons.

I'll also say that just having the same bias or opinion towards the topic is not a COI, which is what's going on with GLAAD. GLAAD has a COI about GLAAD the specific organization but it doesn't have a COI about gay rights in general. Loki (talk) 20:24, 5 November 2024 (UTC)Reply

> In your discussion, if material is published by an organization, is that organization always the publisher?

FOO, I don't think that's quite captured it. It sounds above like @LokiTheLiar is saying that if the material is printed (either on paper or digitally) by a self-publishing service provider, then the service provider is the publisher.

Consider Kindle Direct Publishing: you write a book, you upload your book, and Amazon sells and distributes it. I'd say that Amazon is the printer but not the publisher. It sounds like Loki would disagree with me, and call Amazon the publisher. WhatamIdoing (talk) 23:41, 5 November 2024 (UTC)Reply

Yes: I would say that the author is actually directly the publisher if and only if the author themselves controls the means of distribution. So like, your own personal website would be self-published in the literal sense.

Otherwise, the printer is the publisher and is just choosing to exercise almost no editorial control, which IMO should make us consider the source self-published even though it is not literally published by the author. Note that it's not actually zero even in this case: Twitter will sometimes remove tweets, just not for bad facts alone, which is what we care about. Loki (talk) 00:02, 6 November 2024 (UTC)Reply

Distribution is not publication. Otherwise, the post office is a publisher.

Think about the implications of the first sentence of List of book distributors: "This is a list of book distributors, companies that act as distributors for book publishers, selling primarily to the book trade." If distributors are the true publishers, then that sentence is nonsensical. WhatamIdoing (talk) 00:46, 6 November 2024 (UTC)Reply

Okay, it's clear that you have experience in book publishing, and part of my argument is that we should avoid mapping terms to the ways these are used in literal book publishing because most sources aren't books and don't work like books, so I think we're talking past each other a bit.

I meant "the means of production and distribution", sorry if that's not clear. Loki (talk) 01:13, 6 November 2024 (UTC)Reply

I agree with you that when the printer...is just choosing to exercise almost no editorial control, that's a case of self-publishing. I suggest to you that "choosing to exercise almost no editorial control" is exactly what's happening with Amazon's Kindle platform. WhatamIdoing (talk) 03:31, 6 November 2024 (UTC)Reply

Yeah, I didn't word it well. I was trying to get at the following: let's assume that the person creating the material engages in fact-checking. The organization is not a person and so is not the entity doing the secondary fact-checking; rather, it's one or more people who work for (or volunteer for) the organization doing the secondary fact-checking. My understanding is that as long as someone at the organization has delegated the secondary fact-checking to those people, Loki is saying that the organization is the publisher, and those who do the secondary fact-checking and sign off on it for publication are not the publisher. If an organization's author does no fact-checking or the material doesn't go through a secondary fact-checker or the author and secondary fact-checker share a COI, then the material is self-published. Examples of the latter: a staffer is told to write text promoting XYZ but only works on the wording and doesn't have responsibility for checking whether XYZ is accurate, an organization posts an unedited podcast of a discussion among some knowledgeable employees, an employee live-blogs an event, the material is marketing material. I'm guessing that Loki concluded that Coke's material is self-published because almost all of it is marketing. FactOrOpinion (talk) 01:44, 6 November 2024 (UTC)Reply

Just in case it is not clear, Sulzberger is the publisher through a public corporation (New York Times Company). The corporation is still responsible and liable for publishing. Alanscottwalker (talk) 21:50, 5 November 2024 (UTC)Reply

LokiTheLiar, I'm not sure whether you saw my reply to WhatamIdoing about your text. Since the organization is not a person, I'd modify rely on whether the publisher and the author both have the authority to independently fact-check the work to something like "rely on whether the publisher uses an editor, and if so, whether the editor and the author both have the authority to independently fact-check the work." [Edited to add: Now I'm unclear how your proposal is different from the existing WP:V text "Self-published material is characterized by the lack of independent reviewers (those without a conflict of interest) validating the reliability of the content."] Also, am I correct that you take the Coca Cola website to be self-published because it's primarily (or all?) marketing, whereas for a company like Microsoft, you'd say that much of it is marketing and so is self-published, but some of it (e.g., the technical documentation) isn't? And what would you say about a situation in which the editor doesn't do fact-checking but only edits for readability/grammar and assesses whether the text will be of sufficient interest to readers? Thanks, FactOrOpinion (talk) 19:49, 7 November 2024 (UTC)Reply

I'd modify...

That makes sense, but as-worded it excludes situations where the publisher is the editor in a way I don't really mean to. But you're right that I'm really looking for the presence of an editor and not a publisher per se.

Also, am I correct that you take the Coca Cola website to be self-published because it's primarily (or all?) marketing, whereas for a company like Microsoft, you'd say that much of it is marketing and so is self-published, but some of it (e.g., the technical documentation) isn't?

Hmm, that's an interesting question. I think I would actually say that, say, the Azure documentation is self-published. To me it feels a lot like fiction in that it's reliable for facts about itself where "itself" is construed somewhat broadly to include Azure itself.

And what would you say about a situation in which the editor doesn't do fact-checking but only edits for readability/grammar and assesses whether the text will be of sufficient interest to readers?

Still self-published. Loki (talk) 02:02, 9 November 2024 (UTC)Reply

Honestly, it's better than many RFCs I've seen already.

The advice I'd give is that the most useful and actionable RFCs propose specific wordings of a specific section of a page, and usually a short section. So keep in mind where this text would actually go.

Personally, what I'd propose is replacing the first two sentences of WP:SPS with (changes bolded):

Anyone can create a personal web page, self-publish a book, or claim to be an expert. That is why material published without independent editorial review such as books, patents, newsletters, personal websites, open wikis, personal or group blogs (as distinguished from newsblogs, above), content farms, podcasts, Internet forum postings, and social media postings are largely not acceptable as sources.

and then also add the following section:

Advocacy organizations

Some sources are written by political parties, think-tanks, or other organizations with a clear agenda. Whether these sources are self-published depends on whether the organization has done independent editorial review on the source, in the same manner a WP:NEWSORG would fact-check an article before publication. Even if it has, assume material put out by an advocacy organization is WP:BIASED and attribute it.

Loki (talk) 22:01, 1 November 2024 (UTC)Reply

Your first suggestion has a syntax problem (it inadvertently defines books, etc. as being material published without independent editorial review). It also doesn't address some editors' uncertainty over whether "independent editorial review" applies to ordinary newspaper models. Is the review really "independent" if the journalist/author and the editor/reviewer both get paid by the same entity?

For the second, although it addresses the "independent editorial review" problem (at least partially; hopefully sufficiently), I don't think it's helpful to define them as non-self-published. If the goal is greater internal consistency, then I think we want a system like this:

• Self-published means what it says in the dictionary: one person/group writes it, and the same person/group controls whether it is made available to the public – but traditional publishers are exempt.
• Traditional publishers either work like a mid-century book publisher (the author mails in a manuscript, the publisher rejects it) or like a traditional news org (the publisher/publication hires editors and journalists; the editor assigns stories; the journalist writes them; the editor and publisher/publication representatives decide whether to publish what the journalist wrote) or like a peer-reviewed organization (authors submit papers; editors send papers for external review and uses that information to decide which ones to publish).
• Any entity that editors believe is behaving like a traditional publisher is non-self-published (per explicit exemption in the definition). Note that most traditional publishers have both non-self-published content (e.g., the book, the news article) and self-published content (e.g., marketing materials, investor relations reports, advertising rate sheets).

I think this is internally consistent. The next problem will be editors claiming that almost all corporate and political/campaign materials are non-self-published, because They Have an Editorial Process. WhatamIdoing (talk) 22:20, 1 November 2024 (UTC)Reply

It seems to me that what you're essentially proposing is to create a whitelist of acceptable publishing methods (essentially, news organizations, book publishers, and academic journals) and declare everything that isn't published via one of those models self-published, or at least to be presumed to be self-published. Would that be correct? That is to say, can you think of anything that isn't published via one of those models that you would consider not self-published? If that's the case, it might be possible to word this proposal more succinctly. Personally, I would focus more on potential conflicts of interests among reviewers - I don't think that a publication by a think-tank becomes a RS just because they structure it like a magazine, book, or newspaper. It's ultimately the existence of editorial controls and fact-checking capable of producing a reputation for fact-checking and accuracy that defines an RS; my reading of policy has always been that SPS is a category of non-RS, ie. a way in which a source might lack these things. That key point (they need a reputation) is what bars rando political organizations from just saying "look, we have an editorial process!" while having it be fatally flawed. I do think that there should be a harsher presumption that think-tanks, advocacy orgs, and so on lack a functional process, but if it's unambiguous that they have a reputation for one then we can't override that by splitting hairs with categories. I also don't really agree with the "plain english meaning" argument about SPS; to me, the normal English meaning of SPS only really refers to individuals publishing things themselves and is therefore not useful to us. No one would normally refer to anything else as being "self-published" outside of Wikipedia. I believe that that is what was intended and envisioned when SPS was written - ie. one individual crank throwing out newsletters, not organizations. I don't think that we should necessarily confine ourselves to that but it's important to keep in mind. --Aquillion (talk) 14:36, 2 November 2024 (UTC)Reply

my reading of policy has always been that SPS is a category of non-RS

I think it is not generally reliable. So something with an external publisher likely starts from a presumption of reliability until demonstrated otherwise by lack of independence, poor fact checking etc, while something self-published starts from the other end but can be demonstrated reliable given sufficient expertise and track record, USEBYOTHERS etc.

In practice I think the burden of evidence and direction of travel is different, but not insurmountable - with the sole exception of 3rd party BLP claims, which remain the preserve of traditional publishing because of the sensitivity of the subject. Void if removed (talk) 17:58, 2 November 2024 (UTC)Reply

The point behind WP:SPS is to define the situations in which a self-published source actually is reliable. The canonical example is that if Einstein posted something on his blog about physics, that would be a usable/reliable source. It wouldn't be as good of a source as a peer-reviewed paper saying the same thing, but still: Einstein is a previously published subject-matter expert, and it would be possible to use it.

I do think that the main problem here is that humans do a poor job of separating components. Just like WP:Primary does not mean bad, self-published does not mean bad. We look at multiple characteristics in a source, including the context of the sentence/claim it's being used to support, to determine whether a source is desirable. Just like a primary source can sometimes be the best source, a self-published source is sometimes highly reliable for the sentence/claim we're supporting.

But because of our WP:UPPERCASE culture, we have taught editors not to say "Eh, I don't think that's such a good source; we should probably avoid that", and instead to spew a bunch of shortcuts to "prove" that WP:THE WP:SOURCE WP:IS WP:BAD WP:PER WP:POLICY. WhatamIdoing (talk) 19:31, 2 November 2024 (UTC)Reply

WAID:

1) You again misrepresent the definition of "self publishing". You previously claimed that "traditional" was in the dictionary definition, and you were immediately shown that was false. Once again here's Miriam Webster [9]. (Publish [a written work] using one's own resources). "Traditional" is not there. You then produced a few dictionaries none of which even say "traditional". So what this should say is not your made up terminology, rather, something like:

'Self publishing is where a writer (or creator of work) uses their own resources to publish (such as through social media, gathering a group to publish, or paying to be published (as with a self-publishing company or printer). Examine sourcing through WP:SOURCE: all sources must be examined carefully and in compliance with all policy. If there is dispute about a source, seek consensus. [Don't use self-published sources in BLPs, etc]'

2) So, then you really revealed where "traditional" comes from, a few what look like weak blogs selling self-publishing or pushing self-publishing services (eg. [10], and they don't even support your made up claims: they seem in context to use "traditional", as 'not the business of self-publishing,' and 'self-publishing' as, 'a business where you pay and have control over the publishing of your work.'

3) So, you take the word "traditional" like a logical fallacy as argument from authority and make it up. (In your essay, you also seem to misrepresent the University of Chicago source: 'Any site that does not have a specific publisher or sponsoring body should be treated as unpublished or self-published work.' The Foundations (like Mac Arthur Foundation), Coca Cola, the university, and think tanks, and governments have specific sites they sponsor and are publisher for (and for example, we use Mac Arthur's website, all over the place for BLPs).

4) Finally, you ignore facts: liability fixes the fact that it is a corporation (what you call traditional or not) that is generally the publisher, it is the corporation that has control and pays, and that is at risk as the publisher, for example in claims of defamation.

5) I totally get the impulse to restrict all sourcing to WP:BESTSOURCES, and perhaps I would even be up for doing that. But that's not done by making up our own definition self-publishing, and traditional, here. Alanscottwalker (talk) 15:01, 2 November 2024 (UTC)Reply

See the dictionary definitions:

• "publish (a book etc.) oneself rather than through a publishing house" – The Canadian Oxford Dictionary
• "to arrange and pay for your own book to be published, rather than having it done by a publisher" – Cambridge Dictionary

What do you suppose the dictionary writers mean by "a publishing house", if not the sort of business typically and even "traditionally" involved in the word of publication? Do you think that "rather than having it done by a publisher" includes posting the content online without the involvement of an external organization? I don't.

The absence of the exact words "traditional publisher" from most dictionary definitions does not mean that they're saying corporate PR departments are "publishing houses". It is true that I tend to WP:Use our own words when not providing direct quotations; I recommend that practice to all editors, especially in the mainspace.

Legal liability says that a gossipy old woman is also the "publisher" of whatever slander she spreads. But you are a publisher in copyright terms when you put something in a fixed form and make it available to the public, even if you are not legally liable for it. IMO copyright is a more relevant legal model for Wikipedia than libel laws. WhatamIdoing (talk) 17:40, 2 November 2024 (UTC)Reply

Dictionary definitions are problematic sources here, because self-publishing companies are very clearly publishers, and yet the point of the term "self-published" is to separate them from the normal method of publishing things.

What we're actually doing here is trying to extend a word that has an ordinary and sensible meaning when limited to the world of books to all possible sources, and consequently creating jargon that is not present anywhere else on the internet. So it's up to us, and not the dictionary, to decide what that jargon means. Loki (talk) 18:21, 2 November 2024 (UTC)Reply

That's simply not true, self-publishing as opposed to having a publisher is a commonly understood distinction across all kinds of media. Void if removed (talk) 18:37, 2 November 2024 (UTC)Reply

Self-publishing companies are clearly not publishers, as they have no say in whether the content gets made available to the public. WhatamIdoing (talk) 20:12, 2 November 2024 (UTC)Reply

I'd love to read a discussion of self-publishing that addresses it across diverse kinds of media. Would you link to one (off-wiki)? FactOrOpinion (talk) 22:04, 2 November 2024 (UTC)Reply

A couple of perspectives in video games where distinctions and cost/benefits are very clear:

https://www.lurkit.gg/blog/game-publishing-should-you-self-publish-or-work-with-a-publisher

https://www.gamesindustry.biz/a-beginners-guide-to-bringing-a-game-to-market

This is particularly interesting because it makes it obvious that it is a studio that is self-publishing, and any notion that "self-publishing" is restricted to individuals is obviously untrue.

One for board games which has a clear distinction but less analogous because the traditional publishing model is much more about licensing a design:

https://www.pineislandgames.com/blog/should-i-self-publish

Music industry has always been notorious for exploitative practices by record labels, but the model persists because what you get is marketing and connections:

https://www.vampr.me/faq/self-publishing-vs-record-labels-understanding-the-difference/

At the present time where there's so many self-publishing platforms for electronic content like games, music, video and blogs, it is easy to succumb to the idea that simply "making it available online" is all there is to publishing, and believe that simply adding editorial oversight transforms that process in some fundamental way. Invariably, an arrangement with a separate publisher is a business setup where both parties bring something (marketable content vs marketing infrastructure, connections, brand recognition etc) that is not always noticed by the consumer, who simply accesses the end product with equal ease regardless of how it got there. Void if removed (talk) 23:17, 2 November 2024 (UTC)Reply

https://www.jstor.org/stable/26397544 (2016) says that self-publishing musical scores has been done for centuries, and that about half of the modern composers are engaged in it at some level (ranging, e.g., from some of their compositions to all of them, and from a simple "contact me if you want to buy a copy" note to websites with automated payment and delivery systems). WhatamIdoing (talk) 00:27, 3 November 2024 (UTC)Reply

I see that I wasn't clear enough in my question. I wasn't looking for multiple discussions, each of which focuses on a single type of commercial product (e.g., video games), but for a single discussion that addresses several kinds of media and includes non-commercial publishing as well. But it was interesting to skim the links you provided, thanks. Do you consider Sony Music, for example, as a traditional publisher for music, but a self-publisher info about the musicians whose music they publish? FactOrOpinion (talk) 02:20, 3 November 2024 (UTC)Reply

I can see an argument for treating an "About this musician" page on their website (e.g., https://www.sonymusic.com/artists/) as being equivalent to a book jacket blurb, and therefore within the realm of the traditional publisher. I would personally lean towards treating it as self-published (also non-independent, etc.), but I don't think that anyone holding the opposite view is unreasonable.

However, I would definitely treat their press release section https://www.sonymusic.com/news/ and pages such as https://www.sonymusic.com/careers/ as self-published, and I would think it unreasonable of anyone to say that this was not self-published. WhatamIdoing (talk) 03:47, 3 November 2024 (UTC)Reply

Well games are an interesting example of mixed media publishing. For example there is the title itself, which can be self-published, or it can be published by a third-party publisher - and that process has nothing to do with whose employees physically press buttons and make content available on appstores, it is a legal arrangement as to who takes responsibility for the published product. Then there are ancillary concerns like marketing materials (which may be produced by the publisher itself and are thus self-published), or an associated website (which may be self-published by the studio, or published by the publisher in a similar way to marketing materials). Then there are community hubs like discord, which feature self-published posts by community managers associated with the studio or the publisher with no editorial oversight beforehand. And then there is the fact that publisher can vary by platform and by country. A PS5 version of a game may be published by Sony. A Switch version may be self-published, or may be prohibited by prior contract with an existing publisher. Publishing in China requires a Chinese company by law - a US-based game studio simply cannot self-publish in China.

This then brings us to the territorial concerns of publishers of the sorts of sources Wikipedia relies on which may not be apparent. When people here argue that news org websites self-published, they perhaps don't consider the warning those of us outside the US sometimes see when trying to read or watch their content: the publisher has not made this content available in your region. That was not the person who pressed the button. That was not the editor or the fact checker who is responsible for that decision. The publisher did that.

I think this really is quite simple - if a separate organisation is responsible for publishing the content, then it isn't self-published. Arguments to the contrary confuse the trivial matter of placing content online, with who has legal responsibility for that content. Publishers have ultimate legal liability, and it is settled law in the UK AFAIK that a complainant can decide who to sue, from the author up to and including the publisher.

Think about it this way: if The Guardian publishes defamation about me, I can sue the journalist, the Guardian newspaper, or Guardian News and Media Ltd (the publisher).

If the Free Press publishes defamation about me the buck stops at The Free Press. Void if removed (talk) 10:40, 3 November 2024 (UTC)Reply

if a separate organisation is responsible for publishing the content, then it isn't self-published Do you consider the Mueller Report to be self-published when published by the government but non-self-published when published with supplementary materials by Scribner? If the Free Press publishes defamation about me the buck stops at The Free Press Presumably you can sue the journalist as well. If Weiss were named the publisher (analogous to Sulzberger and the NYT) or formed Free Press News and Media Ltd (analogous to Guardian News and Media Ltd), would you consider the Free Press to not be self-published at that point? FactOrOpinion (talk) 17:11, 3 November 2024 (UTC)Reply

Do you consider the Mueller Report to be self-published if you obtain it from the government but non-self-published if you obtain it with supplementary materials from Simon & Schuster?

I'm not overly familiar with the Mueller Report or US government publishing but in the UK government docs are published by the national archives, is there some similar structure? The second case is definitely not SPS.

Presumably you can sue the journalist as well.

Yes, the point is, there is no higher publisher. The Free Press is both the creator of the publication, and the publisher.

would you consider the Free Press to not be self-published at that point

Yes, though somewhat like when Ariana Huffington was both head of the publisher and publication at The Huffington Post, it could be questioned on other grounds. A technical separation between publisher and creator is not a magic seal of approval, it is a description of a particular publishing structure. Void if removed (talk) 18:09, 3 November 2024 (UTC)Reply

In the US, who publishes government docs varies with the document. There's a Government Publishing Office (GPO) that publishes many documents, but as best I can tell, lots of federal government documents are published by some other entity. For example, I think that the Mueller Report was published by the Department of Justice (DOJ) and that the DOJ publishes diverse other electronic and print documents (e.g., reports, guidance, indictments, videos of press announcements, where those publications are likely overseen by a division or office within the DOJ, such as the Office of Public Affairs or the Civil Rights Division). Probably the same thing for other Departments, like the Department of Education. It looks like Congressional Research Service reports and congressional legislation are published by the Library of Congress (LOC), but it seems like congressional hearing transcripts are published by the GPO rather than the LOC; the latter are not edited even though the GPO might be considered a traditional publisher. FactOrOpinion (talk) 20:12, 3 November 2024 (UTC)Reply

There is also the National Institutes of Health, Center for Disease Control, National Economic Council (United States), National Institute of Standards and Technology, etc,, etc. All of which may well be considered useable in BLP's for living professionals they work with or recognize, but do not employ. -- Alanscottwalker (talk) 15:10, 4 November 2024 (UTC)Reply

WAID, You are not just using your own words, you making up a whole bunch of text based one one or two sourced words. That's no way to honestly represent a source. As for defamation, of course that makes sense to look to, since a primary concern here is living persons. Nor is the property law of copyright meaningfully different here, and we are not concerned with property rights here. -- Alanscottwalker (talk) 14:29, 3 November 2024 (UTC)Reply

Also, some including WAID seem really concerned about corporations for example, publishing about themselves, but they then do something that makes no sense, from any policy standpoint, they really, really must define that, the corporation talking about themselves, as 'self published'. No, you don't, it makes no difference. Alanscottwalker (talk) 15:25, 2 November 2024 (UTC)Reply

Copyright is a poor model for determining whether something is self-published or not. By default, the author holds the copyright even when it is published by some other entity. The default doesn't apply if it is a work made for hire, or if a contract is agreed to transferring copyright to the publisher. But there is no universal policy about whether the author transfers copyright to the publisher or not. It used to be customary for academic papers that the author retained the copyright when an article was published in a journal. Jc3s5h (talk) 20:04, 2 November 2024 (UTC)Reply

@Jc3s5h, I think copyright is a better model than libel law, and in particular, copyright recognizes a distinction between published and unpublished content, and that there is a specific publication date that divides the two. Somebody does something on that date to make the copyrighted material become published. Whoever is doing that 'something' is the publisher. WhatamIdoing (talk) 20:10, 2 November 2024 (UTC)Reply

Its also useful to note that on media sites like Vox or Huffington Post or The Guardian or The Times, the copyright notice invariably identifies the actual publisher. Void if removed (talk) 21:34, 2 November 2024 (UTC)Reply

Defamation also deals in fixed dates of publication. Copyright is concerned with property. Here, we are concerned with reputation and reliability, defamation is also concerned with reputation and reliability. Alanscottwalker (talk) 11:20, 3 November 2024 (UTC)Reply

If I understand @Void if removed: correctly, it would have been the duty of whoever added, to the article Epoch (computing) this source:

• Chen, Raymond (6 March 2009). "Why is the Win32 epoch January 1, 1601?". The Old New Thing. MSDN Blogs.

to consider the article self-published because it's published by a corporation which is not exclusively involved in publishing. Then, the WP:V policy would require that Raymond Chen's "work in the relevant field has previously been published by reliable, independent publications."

This extra burden would be placed upon the poor editor even though millions of people rely every day on Microsoft publications to figure out how Microsoft products work. Jc3s5h (talk) 21:00, 2 November 2024 (UTC)Reply

Why is that an issue? It's obviously self published, and obviously usable. To be clear, that's straightforward WP:ABOUTSELF IMO.Void if removed (talk) 21:16, 2 November 2024 (UTC)Reply

I agree. It's self-published and totally acceptable. Microsoft (in the form of one of its employees) wrote a blog post, and Microsoft (presumably in the form of some other employee) published it. Self-published doesn't mean bad. It just means self-published. WhatamIdoing (talk) 00:07, 3 November 2024 (UTC)Reply

That's wrong. Wikipedia:Verifiability#Self-published sources only gives one situation where a self-published source is acceptable:

Self-published expert sources may be considered reliable when produced by an established subject-matter expert, whose work in the relevant field has previously been published by reliable, independent publications.

This criterion can only apply to one or more identified natural person(s). Therefore, if Microsoft isn't a publisher, then no Microsoft work can ever be used if it doesn't identify one or more natural person(s) as author(s) and at least one of the authors must have work in the relevant field that was previously published by reliable independent publications. And the author who wants to use Microsoft must take time and spend money to read those reliable independent publications, or at least reliable summaries thereof. Jc3s5h (talk) 00:39, 3 November 2024 (UTC)Reply

The policy doesn't say that this criterion can only apply to natural person(s); you're assuming that. More pointfully, we don't have to distort a definition to make this policy do what we want it to do. We could change the policy to say something like, e.g., "Self-published expert sources may be considered.... Sources written for and published by an organization, such as corporate websites, product manuals, etc. may be considered as non-independent primary sources when...." Besides, most self-published sources that we want to use fall under "Self-published or questionable sources as sources on themselves" (e.g., Microsoft decided to have a blog post saying ____ about Microsoft's software, the company denied liability for the disaster, and so forth). WhatamIdoing (talk) 01:47, 3 November 2024 (UTC)Reply

You didn't read down far enough. WP:ABOUTSELF is the second situation where a self-published source is acceptable. FactOrOpinion (talk) 04:16, 3 November 2024 (UTC)Reply

Good example. This looks like an example when "self publishing" is a useless inquiry for us and a waste of time. We are concerned whether it is reliable, including whether the publication is qualified to speak on the matter. (We may sometimes be concerned with COI, but not with this source.) -- Alanscottwalker (talk) 11:52, 3 November 2024 (UTC)Reply

1000 times this: being self-published is a reason that certain sources deserve additional scrutiny but it's not a useful framework for understanding publications by advocacy organizations or corporations, and it is unhelpful to have an "explanatory essay" that conflates this irrelevant issue with the real reasons such publications might deserve scrutiny. 100.36.106.199 (talk) 15:43, 3 November 2024 (UTC)Reply

I agree that being self-published doesn't mean the source is automatically bad, but I don't understand why, if you believe that, you would object to having an essay that literally says WP:SPS#Self-published doesn't mean bad. WhatamIdoing (talk) 04:29, 4 November 2024 (UTC)Reply

@WhatamIdoing Well they're inherently lesser on account of the fact that they're completely banned for any and all BLP material, which is a sizable portion of the site. PARAKANYAA (talk) 12:00, 5 November 2024 (UTC)Reply

That's about 10-15% of the site. That portion must be written conservatively because Wikipedia is not the primary vehicle for the spread of titillating claims about people's lives.

Requiring that a source for this sort of claim be published via a traditional publisher is not in itself a judgement on the quality of a self-published source overall and is a very, very minor thing in the grand scheme of things unless you particularly want to get self-published gossip that no decent RS bothers to touch into Wikipedia. Why in the world would you want to add a claim about a 3rd party BLP that no traditionally published source has reported on? Because that's what is being asked for: a mechanism to add a claim about a living person that not a single reputable newspaper, magazine, book, or journal in the world has repeated. How does that fit with writing conservatively? Void if removed (talk) 12:44, 5 November 2024 (UTC)Reply

Because, conservatively involves providing reliable information, and you don't get to reliable by over expansive use of an unthinking "self-publishing" prohibition. If relevant, we should have no compunction informing readers that Raymond Chen says he confirmed what he says with Dave Cutler, either explicitly though saying that, or implicitly through linking to Chen's statement. Alanscottwalker (talk) 13:46, 5 November 2024 (UTC)Reply

Raymond Chen's self-published blogpost is not an acceptable source for a statement about a 3rd party which - in this case - is trivial and non-ecyclopaedic.

It is however a completely fine source for the information that's actually sourced to it, so bringing up the hypothetical addition of the email exchange is thus a straw man.

If you were to suggest adding the trivia that Chen confirmed this with Cutler, using this source, you would be falling over WP:NOTEVERYTHING hurdles before you even get to the obvious fact it is violating WP:BLPSPS, even if the information is seemingly innocuous. If you somehow managed to convince people it was due, you could say that Chen "in 2009 said he still had the email confirming this" because that is totally banal and permissible under ABOUTSELF.

Lets say Chen was instead saying Cutler once sent him a particularly inflammatory email, and that the only source for this was this self-posted blog. Would this still be an acceptable source? Do you really think you could use this to add a line saying that to Cutler's page? Of course not.

A self-published blogpost is a self-published blogpost, whether it is on your own server, substack, blogspot, or a microsoft corporate website. 3rd party BLP content is 3rd party BLP content, whether it is saying they sent you a lovely polite email or a horrible sweary one. Void if removed (talk) 15:31, 5 November 2024 (UTC)Reply

Once again unthinking and overbroad, it is not that he published a blog, its precisely how the blog was published by the corporation that makes it usable for anything. If Chen ever defamed Cutler in the blog the company published, the company is liable as publisher. As for the rest, I agree that "self-published" is a silly inquiry here, I already said that. If there are other reasons to keep it out, than keep it out. Alanscottwalker (talk) 16:10, 5 November 2024 (UTC)Reply

the company is liable as publisher

No, it is not so straightforward (certainly in UK law).

Website operators (which is drawn widely) are not necessarily "commercial publishers" and they aren't automatically liable for the act of publishing. Rather they can be held liable if they can't demonstrate they aren't responsible. This sort of defence applies to things like your MSDN blog - if the author posted something defamatory, Microsoft could take it down when asked and claim they had no knowledge of the posting before it was posted, and (if this were true) they would not be liable for the act of publishing, as a mere conduit for (in this case) an employee's self-posted content (though if they approved it, then they could be considered "editor").

This is all very much in the weeds though, but if you really don't consider a self-published MSDN blog to be an SPS just because it has a corporate logo on, then frankly BLPSPS is meaningless. Void if removed (talk) 22:57, 7 November 2024 (UTC)Reply

Here's an example. I tried to improve a BLP article for an NPROF, P. P's university, U, wrote about an award, A, that he'd gotten from an outside professional society, OPS1. OPS1 also wrote about awarding A to P, as did another professional society, OPS2, because A is notable within the field. But A is not something that newspapers notice/write about because they simply don't care about most academic awards, even when they're notable in a field. All of these sources are RS, but can I use any of them to note on P's article that he got award A from OPS1? Using your preferred definition of SPS, they're all self-published. OPS2's write-up is definitely out, as it's SPS and ABOUTSELF doesn't apply. Might U's and/or OPS1's write-up be permitted under ABOUTSELF? I initially thought that U's statement fell under BLPSPS, but some say that ABOUTSELF allows a U to make a statement about employees like P. However, U's write-up also mentions OPS1, which I'd say is a third party, and that suggests U's statement can't be used as a source for P getting A from OPS1. ABOUTSELF allows OPS1 to make a statement about the fact that they give the award, but OPS1's write-up also discusses P and his work, and P seems like a third party to OPS1, so that might nix using OPS1's write-up as a source. Would you conclude that I simply can't note the award on P's article? Or is the problem instead with how I interpret "third party"? (FWIW, 2 experienced editors told me that it's fine to use OPS1 as a source for a claim about P getting its award, and one of those editors even modified the text of BLPSPS to say so.) I'm not claiming that this situation is typical for a BLP claim, only pointing out that even a conservatively written BLP might reasonably include statements that no traditional publisher has reported on. FactOrOpinion (talk) 20:51, 5 November 2024 (UTC)Reply

The TLAs in your example caused some temporary MEGO, but here's my answer: Yes. The university's (self-)congratulatory announcement is ABOUTSELF, and the awarder's announcement is arguably ABOUTSELF, at least far as the fact of the award and their reasons for giving the award, are concerned. Even if you don't want to say that the awarder's announcement is ABOUTSELF, as of a few weeks ago, it's now (as you note) explicitly permitted in BLPSPS, which does not refer to a reputable organisation publishing material about who it employs or to whom and why it grants awards. I think that change reflects the community's actual views, so I don't see any reason why this is going to be contested (except for the who/whom thing).

I agree that the second professional society is not usable.

The fact that the university's publication mentions a non-BLP third-party is irrelevant. There is nothing that says you can't cite a university press release for claim #1 if the university press release also happens to include information about something else. There was a discussion a while back – I can't find it right now – about whether BLP rules prevent us from citing an article whose headline implied something derogatory about a BLP (political interference in an arrest, if memory serves), even though the news article was being cited for a perfectly innocuous fact (which city a politician lived in back then). I don't believe the editor was able to convince anyone that the source should be banned. WhatamIdoing (talk) 00:43, 6 November 2024 (UTC)Reply

Why is mentioning a non-BLP third party irrelevant? ABOUTSELF says that SPS "may be used as sources of information about themselves ... so long as ... it does not involve claims about third parties." My impression is that "it" refers to the WP edit, not the source material. That restriction isn't limited to third parties who are people. FactOrOpinion (talk) 02:16, 6 November 2024 (UTC)Reply

I agree: "it" refers to the Wikipedia edit, not the source material, and so it's irrelevant that U's write-up also mentions OPS1, which I'd say is a third party, and that suggests U's statement can't be used. U's statement can be used for information about P and A, even though U's statement mentions a third party, OPS1. If OPS1 were a BLP, then U's statement could be cited, but not to support information about OPS1. Since OPS1 is not a BLP, then U's statement can be cited for information about OPS1, too. WhatamIdoing (talk) 03:16, 6 November 2024 (UTC)Reply

I recognize now that if the third party is not a person, then the BLP restriction doesn't apply (I hadn't thought that through earlier, so thanks). But a self-published source can still only be used for a non-BLP claim if the source's author is an expert in the relevant field. The author of the U's statement is unidentified, so I can't know whether they're an expert in the professional society's field. The prof is an expert in that field, and I could have used a statement from him about it, though that might bump into the "unduly self-serving" restriction (or perhaps it's no more self-serving than for the U to say it). Ultimately, I'm glad that Andy Mabbett (Pigsonthewing) made that change to BLPSPS, as it enabled me to use the professional society's statement about it, and their statement was much more detailed about why the prof's work was significant in the field / why he merited the award. FactOrOpinion (talk) 04:41, 6 November 2024 (UTC)Reply

10-15% is a lot of the site! That would be an absolute nightmare. PARAKANYAA (talk) 10:05, 9 November 2024 (UTC)Reply

Most of this discussion leaves me cold. Why should we care whether particular well-defined classes of sources are "self-published"? What we should care about is whether they are reliable. Vague categories like "self-published" are only useful for cases that don't fit any of the well-defined classes. Zero^talk 02:01, 6 November 2024 (UTC)Reply

There's a strong BLP restriction on using self-published sources: "Never use self-published sources ... as sources of material about a living person, unless written or published by the subject of the article." That's why it matters that we have a means of determining whether a source is or isn't self-published. FactOrOpinion (talk) 03:01, 6 November 2024 (UTC)Reply

The endless discussion about what sources are self-published shows that this restriction is unsatisfactory as written. The policy should precisely define the main classes of sources that are intended here and only leave the vague class for the miscellaneous cases. Zero^talk 03:28, 6 November 2024 (UTC)Reply

We're having the discussion because there is no agreement on what the intended classes are. Initially people with different views were trying to convince each other, but at this point, I think the goal is to come up with clear wording for an RfC, so that the consensus policy is clearer. FactOrOpinion (talk) 04:59, 6 November 2024 (UTC)Reply

I think the new edition (18th) of Chicago Manual of Style can shed some light about what the rest of the English-speaking world means by self-published:

¶ 14.36 Self-published or privately published books. Books published independently by the author, like traditionally published books, should be cited according to the information on the title page or copyright page or otherwise known. Unless the book has been published under a publisher imprint name (in which case it can be cited like other books), "published by the author" can stand in for the publisher's name. (Because "author" is more logical than "self" in this context, Chicago now prefers this phrase in sources citations over the term "self-published.") The name of any self-publishing platform or distributor such as Amazon is usually omitted...

I think this very much leans toward a self-published work being one published by an author or group of authors, rather than regarding works published by an organization with editorial input from the organizations as being self-published. Jc3s5h (talk) 20:12, 3 November 2024 (UTC)Reply

Sure. For publisher, we would always put the corporation in that cite-field, not the employee author, the employee author goes in the author cite-field (or something like, "staff") -- Alanscottwalker (talk) 20:23, 3 November 2024 (UTC)Reply

We normally omit unnamed authors. Here's relevant examples from Template:Cite press release/doc and Template:Cite web/doc:

• "Sprint 4G LTE Lights Up in 11 New Cities and Counties in Indiana, Kansas, Maryland, Massachusetts, Pennsylvania, Virginia and West Virginia" (Press release). Overland Park, Kansas: Sprint Nextel. November 26, 2012. Retrieved November 28, 2012.
• "NFL Rules Digest". NFL Football Operations. National Football League. Retrieved November 4, 2024.

I would consider both of these to be self-published (and the second to be an authoritative source). WhatamIdoing (talk) 04:44, 4 November 2024 (UTC)Reply

A source that makes it unambiguous that self-publishing involves a natural person is the definition of self-published in The American Heritage dictionary of the English Language:

adj. Published by oneself or with one's own money: a self-published memoir; a self-published poet.

Two other sources could be understood to mean it has to be a natural person depending on whether one accepts the notion of a corporate author in this context.

dictionary.com (omitting pronunciation and alternate forms for all entries):

self-published

adjective

1. published independently by the author: self-published books.
2. having published one's onwn work independently: a self-published author.

merrian-webster.com

transitive verb

:to publish (a book) using the author's own resources

Jc3s5h (talk) 03:33, 4 November 2024 (UTC)Reply

Though Wikipedia is definitely influenced by definitions and words corresponding to what they mean, there is no requirement for us to follow the rules or standards developed by other organizations.

We use community consensus to decide what, for example, WP:RELIABLE means, not a dictionary. Bluethricecreamman (talk) 03:41, 4 November 2024 (UTC)Reply

Yes, it's true that our terms sometimes diverge from ordinary words, but self-published in this policy is supposed to be the ordinary dictionary definition. It is not supposed to be some kind of wikijargon (like "neutral", which looks biased to some people, or "notable", which can be achieved for subjects that ordinarily get overlooked).

@Jc3s5h, I wonder if you're reading too much into the wording of the American Heritage definition. If you take "one" to mean "a single human", then what would you call two humans – perhaps a husband and wife team – jointly publishing something using their own resources? Any entity could be "one", which (in its noun form) the same dictionary defines as "A single person or thing; a unit" and as "An unspecified individual; anyone". Both "thing; a unit" and "anyone" could encompass a group of people. One is a grammatically singular pronoun, but it does not necessarily refer to a single human. WhatamIdoing (talk) 05:00, 4 November 2024 (UTC)Reply

I'd argue the strict dictionary definition does not have a carve-out exception for traditional print publishing, while Wikipedia essays and community practice definitely does. Bluethricecreamman (talk) 05:13, 4 November 2024 (UTC)Reply

The dictionary definitions don't need a carve-out, because in traditional print publishing of books, the author submits to the publisher, and if the publisher accepts it, the publisher not only pays the publication expenses, but usually pays the author as well. But academic journals are a different matter; the author is usually associated with some institution, and the institution makes a contribution toward the expense of publishing. Jc3s5h (talk) 05:50, 4 November 2024 (UTC)Reply

It is perhaps easier to understand this if we look at the different meanings of "publisher".

"Publisher" has a specific meaning separate from the verb "to publish". See:

a person or company whose business is the publishing of books, periodicals, engravings, computer software,

UK defamation law makes this distinction clear. It considers the "publishing" of defamation to be a continuous, ongoing process of making that material available, and in that process it identifies the author, editor and publisher as entities liable. However, "publisher" is narrowly defined as:

“publisher” means a commercial publisher, that is, a person whose business is issuing material to the public, or a section of the public, who issues material containing the statement in the course of that business.

So even though a person or company can be the one who causes the publishing to happen, they are not the "publisher" unless they are specifically a business engaged in the publishing of material.

The word has two meanings. A publisher is both in the general sense "the person who publishes" and it is in a specific sense "a business whose business is publishing". And very simply, something becomes self-published, if it is not published by "a business whose business is publishing", because by the more general meaning of "publisher", "self-publishing" is a tautology.

The confusion arises because we have a form of business whose name in English is constructed from an everyday verb, creating confusion. Consider: cleaner. If I clean something, I am in a sense the "cleaner" of that thing. But I am not a cleaner, in the sense of that being my business. I have not hired in a cleaner to do my cleaning. If I clean my workstation at the office, I am the "cleaner" in the sense of the act, but I am not a "cleaner" employed by the company. Yet every act of cleaning is "self-cleaning" in the sense that a person does it, and if you take the view some here are by collapsing the two different meanings, there is no difference between me and a cleaner. Yet in the obvious, everyday sense, of course there is, because we understand that distinction.

The distinction is much more obvious in other languages. For example, in German a publisher is Verlag and quite specifically refers to a publishing company, but "to publish" is veröffen, and self-published is Selbstveröffentlichung. The distinction in English is the same, but because it is the same word, confusion arises.

You need to separate the act of publishing (which can seem deceptively trivial in this electronic era of self-publishing platforms) from the business of being a publisher (which encompasses far more than just "making a website"). Anything that does not go through some sort of specific publishing organisation whose purpose is publishing, is self-published. It makes much more sense to consider from that direction, and that means the only things that aren't self published are books, newspapers, magazines, journals etc. I think website material from advocacy orgs so obviously fails this standard it isn't even close to an edge case.

What is causing people issues falsely conflating SPS with unreliability, and then using the different meanings of "publisher" to try and sidestep something that doesn't need sidestepping. Lets consider a seemingly hard case like where a genuinely trusted source (say, a government body) produces an independent report from which we want to draw a 3rd party BLP claim (say, into the conduct of specific individuals, like the Rotherham scandal). If we wanted to cite the Jay Report to say that a specifically named person:

received a report by the Director of Targeted Services on the progress of arrangements to protect young people from sexual exploitation.

Would that be ruled out as "self-published"? Perhaps in that strict sense, yes it would be? Is that a big deal? It is still a completely authoritative source for absolutely every other claim about this subject (because self-published doesn't mean bad). If the 3rd party BLP claim is due, there will be secondary coverage, but if literally the only source of 3rd party BLP claims is a government report, then maybe it is absolutely right that editors shouldn't go mining for it to put it on Wiki. To properly judge if it is the sort of thing Wikipedia can say on the highly sensitive subjects of BLPs, a reliable, secondary source with an actual publisher will report it, and that is the whole point of BLPSPS. Void if removed (talk) 10:11, 5 November 2024 (UTC)Reply

You misread that law, the company that is engaged in making the publication available is the commercial publisher. They are issuing the material containing the statement in the course of their business, and as long as it is on their website, for example, continuing to make it available. (As an aside, governments don't generally fall under ordinary statutory law like that one, they are generally immune from such claims). -- Alanscottwalker (talk) 13:09, 5 November 2024 (UTC)Reply

@LokiTheLiar, @WhatamIdoing, @Alanscottwalker @Alpha3031, @Void if removed, I am planning on releasing this RFC sometime this weekend.

• I think I saw some ideas about making sure to include the obvious exceptions we would expect for SPS (i.e. public comments on news articles are usually SPS).
• I also saw some debate about whether to use the term grey lit and how nebulous to make this RFC. I think grey lit is nebulous enough, and rfc has caveats for when literature from an advocacy org/etc. is obviously SPS I think it should be fine. I removed grey lit from the title of the RFC anyways.
• the text in the proposed addition to WP:USESPS is just for an essay, to capture community consensus. IDK the process for editing a whole policy from wikipedia at this point.
• I don't think the debate dies with this RFC, but thought this is the question that captures the spirit of most of this debate, about whether coming from a non traditional publisher automatically makes material SPS. I was hoping to make an RFC that would not lead to absurd scenarios for either camp.

Should I put this out on WP:RSN or on it's own page? Is there anything fatally flawed with this RFC, or something else we need to change? Please feel free to give it a last lookover and edit if yall want, I tried to skim this discussion to incorporate feedback, but its a heck of a lot. Bluethricecreamman (talk) 05:37, 8 November 2024 (UTC)Reply

I disagree with this approach. I think this wrongly focuses on (politically expedient) exceptionalism when really we need an RFC clarifying what counts as "not self published".

Option 1 of this RFC doesn't clarify anything much even in the case that kicked this whole thing off.

The problem here is you're potentially creating a case where the most explicitly politically partisan material from organisations *whose entire purpose is to be politically partisan* becomes usable for BLPs. Do you not see how that's a recipe for disaster?

I think you should invite input from WP:BIOGRAPHY first. Void if removed (talk) 09:10, 8 November 2024 (UTC)Reply

if you think option 1 is magnificently explosive and will lead to insane policy, feel free to vote option 2, or propose another option.

i specifically made option 1 vague and made sure to include wording that not all lit from advocacy org is always nonSPS. i feel it addresses the crux of the issue without making an open ended question.

imho the other policies such as wp:independent, wp:due, and wp:npov already cover option 1 and the worst case scenario. Bluethricecreamman (talk) 14:02, 8 November 2024 (UTC)Reply

I also think this is a bad idea. I think you should not have "specifically made option 1 vague". Vagueness leads to different people having incompatible ideas. What I want is:

• Alice says it is self-published. Bob says it isn't self-published.
• We have an RFC.
• Alice and Bob can look at the result and agree on whether it is/isn't.

What you're most likely to accomplish is:

• Alice says it is self-published. Bob says it isn't self-published.
• We have an RFC.
• Alice and Bob look at the result and continue to disagree. The only difference is they start using the words "gray literature" instead of "self-published".

WhatamIdoing (talk) 15:36, 8 November 2024 (UTC)Reply

there is no RFC that will solve SPS for all cases. and there is no question that ends the debate without being so precise as to either not apply (folks will argue that GLAAD report doesn't count since it doesn't have criterion x, y, z), or will lead to folks pointing out logical inconsistencies for either side (i.e. news publishers website public comments, GLAAD ads on twitter).

I doubt any RFC ends the debate, especially when all folks are motivated by argumentum ad consequentiam, as WAID correctly pointed out. I am looking for the RFC question that lead to the least invocations of WP:IAR and answers the narrow question of if literature from advocacy orgs/similar orgs is always SPS or not. You can continue arguing if GLAAD is SPS after if option 1 wins, but I am looking to see if the SPS definition is one that is the literal definition that ends this debate entirely as per WAID and Void if removed, or if there are additional considerations the community does that we can base our next debate on. Bluethricecreamman (talk) 15:47, 8 November 2024 (UTC)Reply

That might be true, but I don't think that the RFC you are drafting will solve SPS for any cases. WhatamIdoing (talk) 16:02, 8 November 2024 (UTC)Reply

I agree with WhatamIdoing here. Be as specific as reasonably possible, and especially be specific enough to actually resolve the issue without another RFC. Loki (talk) 23:37, 8 November 2024 (UTC)Reply

@Bluethricecreamman, of course you're free to pose any RfC you want, but I disagree with the current one.

• Mostly importantly: I don't understand why you're proposing text for WP:IDSPS rather than for WP:SPS. More than one editor in this discussion has said something along the lines of "an essay isn't policy," and I think the goal here should be to generate consensus for the wording/meaning of WP:SPS as a policy (and then that would be used to modify WP:IDSPS as appropriate, but the WP:IDSPS text is not the focus here).
• Your proposed text is quite long, which increases the likelihood of people picking at lots of pieces of the text (e.g., I think it's a mistake to say things like "a good reputation for reliability," as that confuses SPS with RS). Right now, my own response to the RfC would be "neither."
• I don't understand why it's focused on literature from "advocacy groups and other similar orgs," when editors need to understand how to evaluate whether something is/isn't SPS for any source (e.g., material published by the government, a corporation, a university -- none of which I consider similar to an advocacy org).

Since I think the focus of the RfC should be WP:SPS, I'll say that a key problem in the current WP:SPS text is that the central paragraph does not define self-published, instead just listing some kinds of SPS. To the extent that "self-published" is defined, it's in a footnote: "Self-published material is characterized by the lack of independent reviewers (those without a conflict of interest) validating the reliability of the content." If that's the definition, it should be moved into the main text, and we need to be clearer about how to determine whether a COI exists, as WP:COI doesn't help us determine COI outside of WP (e.g., if simply working for the same organization creates a COI, as is the case for WP:COI, then a newspaper editor is not independent). That WP:SPS footnote also lists further examples, including "the material contained within company websites," and several people in this discussion disagree with that, as they believe that such material is generally reviewed by a sufficiently independent reviewer.

I guess I'd frame the RfC around adding the text "If there is no reviewer, the work is self-published. A given publisher might publish some material reviewed by an independent reviewer and other material that isn't. For example, a newspaper website might include both articles and comments from readers." Then I'd ask people to choose between something like the following in determining whether a reviewer is independent for the purposes of WP:SPS:

• A reviewer for a traditional publisher (e.g., newspapers, non-vanity-press book publishers, peer-reviewed journals, established record labels) is independent except when reviewing marketing material, and reviewers at other kinds of organizations are not sufficiently independent.
• A reviewer is usually independent, with limited exceptions (e.g, people who review marketing or campaign materials, AfC reviewers on WP).

I'd actually ask WhatamIdoing and Void if removed to come up with the specific wording for the first option, and those who take the second view (e.g., Alanscottwalker, ActivelyDisinterested) to come up with specific wording for the second option. Loki may want to add a third option.

Sorry that this is so wordy. FactOrOpinion (talk) 17:04, 8 November 2024 (UTC)Reply

Honestly, I think WP:USINGSPS is very clear.

My suggestion would be to incorporate the "Examples of self-published sources/Examples of non-self-published sources" into WP:SPS, verbatim. I'd maybe swap the order round, so you get the non-self-published sources first.

Non-self published: books, newspapers, journals. Virtually everything else is self-published, explicitly including company and charity websites.

But fundamentally this is my point: if you can't simply verbatim include that text, then you have a policy and an explanation that don't match. No amount of getting into the weeds with grey literature fixes that.

Either explicitly endorse WP:USINGSPS or come up with an actual definition of SPS.

So option 1 is move the examples from USINGSPS to SPS as is:

Examples of self-published sources

• Almost all websites except for those published by traditional publishers (such as news media organizations), including:

• Blogs
• Web forums
• Wikis
• Social networking sites like Facebook, Myspace, Google+, Twitter, and LinkedIn
• Sites with user-generated content, including YouTube and Find A Grave
• Business, charitable, and personal websites

• Books printed through a vanity press
• Advertisements, pamphlets, and press releases
• Newsletters published by organizations
• Patents (see Wikipedia:Reliable source examples#Are patents reliable sources?)

Examples of non-self-published sources

• The contents of magazines and newspapers, including editorials and op-ed pieces in newspapers (including online-only content of widely-circulated magazines and newspapers)
• Books published by established publishers (like Random House)
• Research published in peer-reviewed journals

Option 2 is change USINGSPS to this and then move it to SPS:

Examples of self-published sources

• Blogs
• Web forums
• Wikis
• Social networking sites like Facebook, Myspace, Google+, Twitter, and LinkedIn
• Sites with user-generated content, including YouTube and Find A Grave
• Personal websites
• Books printed through a vanity press
• Advertisements, pamphlets, and press releases
• Newsletters published by organizations
• Patents (see Wikipedia:Reliable source examples#Are patents reliable sources?)

Examples of non-self-published sources

• The contents of magazines and newspapers, including editorials and op-ed pieces in newspapers (including online-only content of widely-circulated magazines and newspapers)
• Books published by established publishers (like Random House)
• Research published in peer-reviewed journals
• Business and charitable websites

I think option 2 is ludicrous, but I think this is much clearer than messing about with "grey literature" vagueness. Void if removed (talk) 21:35, 8 November 2024 (UTC)Reply

Void, I think we'd also want to something about government reports, as that's a type of source that occasionally upsets people, whom we then find grasping at all straws for anything and everything that could be used to disqualify it. (For example: A government report concludes that a given place is no longer safe to live in, or it rejects someone's favorite tax-avoidance scheme as fraud, or that the disproportionate poverty among elderly women is primarily caused by being a stay-at-home mother, or whatever.)

It would be particularly ironic if we declared reports from reputable government agencies to be self-published by non-experts, but advocacy websites criticizing those same reports are non-self-published. The CDC says that horse dewormer doesn't treat COVID, but that's self-published, so we can't use it. That advocacy group claiming the opposite, however, is perfectly fine. WhatamIdoing (talk) 04:10, 9 November 2024 (UTC)Reply

We can always use SPS if its reliable. We just can't use it for WP:BLP, as per WP:BLPSPS Bluethricecreamman (talk) 02:39, 10 November 2024 (UTC)Reply

@Bluethricecreamman BLP material is a sizable portion of this site, to the point that it would be unusable on a great deal of the most relevant articles. It doesn't have to be a BLP article, but any material that is a statement about a living person, directly or indirectly. That is a very wide scope. PARAKANYAA (talk) 02:53, 10 November 2024 (UTC)Reply

Some people don't think that USINGSPS is entirely clear, as can be seen on the essay's Talk page. And as I noted earlier, an organization might have both non-self-published and self-published content (e.g., a NYT article vs. comments on that article, a government report vs. a government hearing transcript), so in determining whether a source is or isn't self-published, people should focus on the specific source and not the organization. That argues against some of those bullet points. FactOrOpinion (talk) 15:10, 9 November 2024 (UTC)Reply

I generally agree with FactOrOpinion here, and I especially agree that the proposed text should be as short as and as clear as possible. If you cut the collapse boxes entirely and proposed just adding the italic text directly (with the minor edit of replacing such literature with literature from advocacy groups, it would significantly improve your proposal.

I also agree that the focus on literature from advocacy groups may be too narrow, but that one I'm more torn on, since trying to pick out a general rule might be quite difficult. Loki (talk) 23:44, 8 November 2024 (UTC)Reply

• I think I tried to follow along and tried to think of framing a cogent definition of SPS, but cannot think of something succinct enough for an RFC; I also don't think I have enough experience to make a broad enough definition that covers all edge cases, so I'm keeping the question somewhat narrow on grey literature from advocacy orgs and other similar orgs.
• i went ahead and removed the collapsed text proposals. I also just went ahead and used the grey literature idea from Alpha to describe a broad class of docs we were discussing. I know we are gonna debate about if its too broad a category, or too vague... but any category will necessarily be too broad or too vague, or so constrained with so much words the RFC text will be too long to parse.
• I've gone ahead and just posted it to WP:RSN. If folks want another option to fill it out, feel free. I also think that if folks want to do an RFC on a fully constrained definition of WP:SPS, please feel free to. Bluethricecreamman (talk) 02:51, 10 November 2024 (UTC)Reply

Help me 2404:1C40:D7:BF1E:1:0:EC41:B98B (talk) 12:14, 5 November 2024 (UTC)Reply

I think your lost, this page is for discussing changes to Wikipedia verification policy. -- LCU ActivelyDisinterested «@» °∆t° 19:10, 5 November 2024 (UTC)Reply

Contact Telegram support, not Wikipedia. JacktheBrown (talk) 00:33, 13 November 2024 (UTC)Reply

I know we already talked a ton above about this, but may as well put the notice down here as well in its own section to notify folks who may have tuned out of the previous discussion. See here: RFC: Should grey literature from advocacy groups and other similar orgs always be considered WP:SPS and therefore subject to WP:BLPSPS? Bluethricecreamman (talk) 02:54, 10 November 2024 (UTC)Reply

Can somebody close the discussion linked above? This is already the correct venue for such discussion, or perhaps at the policy pump page. BusterD (talk) 16:14, 10 November 2024 (UTC)Reply

@BusterD i think ive seen folks move RFCs to their own page or to other more appropriate venues. cant i just do that?

which venue would you prefer? Bluethricecreamman (talk) 16:23, 10 November 2024 (UTC)Reply

I just did and moved it to centralized Wikipedia:Requests for comment/Grey Literature page as is common for larger discussions. Since we already have notices at all possible venues that may be interested in the discussion, no futher action should be needed. Raladic (talk) 16:35, 10 November 2024 (UTC)Reply

Thanks for doing that. A page named Wikipedia:Requests for comment/Grey Literature (or something similar) is one of the recommendations in WP:RFC for any RFC that is expected to draw a large number of comments. (Also, we wouldn't normally close an RFC over a location that is perceived as suboptimal by some; we'd normally either just move it, which is what @Raladic did for this one, or post a larger number of notices.) WhatamIdoing (talk) 05:37, 12 November 2024 (UTC)Reply

The birth place of Sam Mills is not Neptune City New Jersey because there is not a hospital with maternity facilities in Neptune City and his parents never lived there. Therefore change Neptune City ,N.J. to Neptune Township ,N.J. He was born at Jersey Shore Medical Center in Neptune New Jersey. Parts of Neptune City is across the street from that hospital. There are too many people confusing Neptune City Borough with Neptune Township 2601:45:400:2CC0:8818:726B:923B:C7C9 (talk) 00:31, 18 November 2024 (UTC)Reply

Is this about the Sam Mills article? You should ask at Talk:Sam Mills instead. (Also, not every baby gets born in a hospital, or in the place where their parents live.) WhatamIdoing (talk) 00:39, 18 November 2024 (UTC)Reply

I think I asked this question over a decade ago somewhere and it's been lost in the recesses of Wikipedia talk archives. If the owner/publisher of a reliable source writes pieces for their publication, are they a self-published source? Like, for example, if Ian Danzig writes an article for Exclaim! (which he owns and publishes), or HM's founder and publisher Doug Van Pelt or Jesus Freak Hideout's owner and publisher John DiBiase write articles for their respective websites, or A. G. Sulzberger writes a story for The New York Times, are those articles self-published sources only or are they considered reliable, independent published sources?--3family6 (Talk to me | See what I have done) 22:00, 19 November 2024 (UTC)Reply

… depends on the publication, publisher’s relationships with paper, and more. The discussions above as well as ongoing Wikipedia:Requests for comment/Grey Literature all suggest divergent opinions in community. Bluethricecreamman (talk) 22:27, 19 November 2024 (UTC)Reply

Your use of "secondary" in the heading is out of place. An article can be published by a reliable publisher but be primary. A paper in a journal published by a learned society reporting new experimental results is an example of that. And there is nothing stopping a self-published article or book from being based mostly, or entirely, on other sources, which makes it secondary. Jc3s5h (talk) 22:44, 19 November 2024 (UTC)Reply

Wikipedia:Reliable sources/Noticeboard/Archive 282#RfC: Quackwatch concluded that articles written by the owner/publisher of Quackwatch were self-published. WhatamIdoing (talk) 05:37, 26 November 2024 (UTC)Reply

@WhatamIdoing I'm now also trying to figure out if the commentary and factual statements accompanying the premiere of an artist's work is essentially a primary source. For instance, on my sandbox I'm working on an article, and I cite some BLP claims (a musician converted to Christianity and his band released to Christian metal albums) to a premiere of new music by that band. The author of the magazine post premiering the material is the owner of the magazine (thus it's self-published), but in this case I'm assuming that the commentary of that author could essentially be considered closely affiliated with the band, that is, essentially a press release. I'll note that the faith of the musician is not in question, that's been stated by the artist himself, but the timing of his conversion (mid-1990s) and that two albums were released afterwards I haven't seen expressly stated in interviews or other press releases.--3family6 (Talk to me | See what I have done) 16:07, 26 November 2024 (UTC)Reply

Of course it is a primary source. It is also non-independent ("Alice Artist tells you what Alice Artist thinks").

But depending on the structure, it might not be self-published; the most I could tell you is that the post could be argued to be self-published, and that it could be argued to not be self-published. The latter is because anyone who puts out a magazine is an established, traditional publisher, and some definitions exempt all traditional publishers. I would not worry overmuch about this one.

Remember that Wikipedia:Primary does not mean bad. WhatamIdoing (talk) 17:20, 26 November 2024 (UTC)Reply

@WhatamIdoing oh, primary is fine, my concern is that if it isn't primary. Because then it's a secondary source (the publisher) making a statement that's essentially self-published but used to support BLP claims.--3family6 (Talk to me | See what I have done) 17:43, 26 November 2024 (UTC)Reply

An Artist's statement (which is a case of "commentary and factual statements accompanying the premiere of an artist's work") is primary, non-independent, and often self-published (but not always, e.g., self-published on the author's website or a commercial gallery wall, but non-self-published if it is quoted in a scholarly article). It is also an excellent type of source to use (usually with WP:INTEXT attribution) for information about the art and the artist.

A magazine article that contains similar information is often primary (but not always, e.g., if it takes the form of a compare-and-contrast analysis). It is assumed independent unless there is some COI-type relationship between the artist/band/album and the author/magazine/magazine owner. It is usually not considered self-published. WhatamIdoing (talk) 22:59, 26 November 2024 (UTC)Reply

@WhatamIdoing Right. My concern is that as the magazine is independent of the artist, and the content is a summary put together by the magazine's publisher, it's technically an independent source but a self-published one given that the publisher is the author and thus not under normal editorial review. In terms of the spirit of what not using SPS for BLPs is about (defamation), it's fine, as the subject states his faith elsewhere, but technically if the source is an independent SPS it's a violation of the stated consensus to never use such sources for BLP statements.--3family6 (Talk to me | See what I have done) 12:27, 27 November 2024 (UTC)Reply

I think that "thus not under normal editorial review" is an assumption, and possibly a false one.

The usual practice seems to be to start off with "Well, is it obviously an ordinary newspaper/magazine/academic journal?" If so, then we're done: It's not self-published. WhatamIdoing (talk) 20:13, 27 November 2024 (UTC)Reply

@WhatamIdoing well, that's why I brought up this issue. And as you point out, Quackwatch, for example, is considered to be self-published if the content is from the website publisher. With the example I'm working with, we're talking about something that was a print magazine but now is fully online only.--3family6 (Talk to me | See what I have done) 20:26, 27 November 2024 (UTC)Reply

There are two ways to address this concern. I'll give you both:

1. If this magazine were still in print, would you still be worrying about whether this article is self-published? If not, then you shouldn't now. Being self-published has nothing to do with whether the publication is on paper or on a website.
2. What do you think is best for the article? If, using your own judgement as an editor, you think that citing this source makes that article better, then do it. If you think it makes the article worse – well, we wouldn't be having this conversation now, right? Because if you actually thought that, you'd have already rejected the source.

WhatamIdoing (talk) 21:33, 27 November 2024 (UTC)Reply

@WhatamIdoing on point 1, yes, I've treated them the same in the past, and if I'm going to presume one form is reliable or unreliable, I'm going to presume the other.--3family6 (Talk to me | See what I have done) 04:16, 28 November 2024 (UTC)Reply

Thank you for these replies, this was helpful and I'm already implementing some changes on pages I've created/worked on.--3family6 (Talk to me | See what I have done) 12:34, 26 November 2024 (UTC)Reply

@3family6, you have made 26,000+ edits over the last 16 years, including about 150 articles. At this point, the only rule I really want to see you closely following is this one: Wikipedia:If a rule prevents you from improving or maintaining Wikipedia, ignore it.

As you have noticed, I'm happy to blather on about the technical distinctions between these things and the awkward gray areas, but when you're working on articles, you need to do what's right by the article, even if "the rules" suggest hurting the article. Total compliance with the rules is not the goal. Let me trust you to do the right thing. If you give me decent articles, I can figure out ways to make the rules align with community practices. Articles first. WhatamIdoing (talk) 23:05, 26 November 2024 (UTC)Reply

@WhatamIdoing Wow, I really appreciate that comment, that is very kind. I guess my concern is your last part about community practices. I want to make sure that I'm actually going by community practice and not just unilaterally making content decisions. I was cleaning up one of my articles yesterday where I had include a source once way back that was not reliable. In general I think it's good for me to reflect on whether some of my editing practices are actually the best practice and actually improving the encyclopedia, or if they're bad habits that I fell into. I also just wanted to get some clarification on these gray areas not just for myself but for others. So there's a discussion I can point to.--3family6 (Talk to me | See what I have done) 12:31, 27 November 2024 (UTC)Reply

Years ago, an editor used to say that if you haven't made 50 mistakes yet, you aren't a real Wikipedian. Mistakes are okay. Fixing them is great.

In general, I think it's important and valuable to conform to community practices, but sometimes the best practice isn't the popular one. The community needs some editors to be able to see and speak to best practices, instead of simply following rules, as if our policies and guidelines were holy writ. So I need – we need – you and editors like you to be thinking about what's best, rather than what's officially endorsed by The Rules™. WhatamIdoing (talk) 21:30, 27 November 2024 (UTC)Reply

My take on the question is that most self-published sources are not reliable sources because we have no way of judging their reliability. We do make some exception; we accept blogs written by acknowledged experts when they are writing about the area of their expertise. In general, we accept sources as reliable when they (or their publishers) have a history of publication that has shown them to be generally reliable. It is not true that a self-published source can never be used. The point is that we should not use sources for which we cannot determine reliabilty, and we cannot make that determination for most self-published sources. If a self-published periodical (whatever that means) has enough history, however, we can make judgements about the reliability of that publication for topics it has published on. Donald Albury 22:20, 27 November 2024 (UTC)Reply

@Donald Albury Right. I would presume that music reviews, if nothing else, could be considered editorials, if not more. Where it matters would be BLP statements.--3family6 (Talk to me | See what I have done) 04:19, 28 November 2024 (UTC)Reply

Because I think there ought to be one. The present policy defends less accessible sources entirely but there ought to be a balance. If there is a more accessible source and a less accessible source for a given bit of information, and it's not desirable to cite both, the more accessible source should be preferred. If a piece of information with a citation from an inaccessible source is contradicted by a more accessible, equally reliable source, there should be a preference towards the verified piece of information unless the first source can be located. If I own the latest edition of a book on a topic and the text in question is the same as an older edition that is on archive.org, I should cite the older edition because editors can go look at the book themselves more easily. Doing otherwise might not be a problem in itself but might lead to issues down the line. Fangz (talk) 08:35, 4 December 2024 (UTC)Reply

When Wikipedia was young the better sources were almost exclusively off-line. While it is true that more reliable sources are now available on-line (almost all scholarly journals and many books), I would say that more than 95% percent of what is available on-line is still not usable as sources in Wikipedia. Most of the best sources on-line are still those that have been published on paper before or simultaneously with the on-line version. Unfortunately, many of the on-line sources are behind paywalls, but that does not mean that we should accept poor sources simply because they are free. The WikipediaLibrary has helped with that, providing access to paywall protected sources to editors who meet the requirements. Indeed, it has allowed me to drop my private subscription to JSTOR. We should always strive to use the best reliable sources to support content in articles. I would say that more of the highest quality sourcing is available on-line now than was the case 15 to 20 years ago, and so there is even less reason today to ease our sourcing standards than there was then. Donald Albury 16:15, 4 December 2024 (UTC)Reply

The nightmare case is editors misrepresenting sources, and doing it with obscure sources that are hard to check. I expect you can think of the particular editor I have in mind (I'll name them if requested but I don't know that it adds much). I don't have a real solution to this; such an editor would not really be stopped by a statement that high-availability sources are preferred. But it's something to keep in mind. --Trovatore (talk) 00:07, 5 December 2024 (UTC)Reply

In many cases it is a good idea to cite an off-line or subscription source with lots of detail, or especially authoritative, also citing one accessible online, perhaps with less detail or authority. Remembering also that google books previews are only available in different countries or times entirely as the publisher chooses. Johnbod (talk) 16:24, 4 December 2024 (UTC)Reply

I'm also seeing publishers or authors putting fairly recent but out-of-print books on-line for free, or as a free e-book. Not very common, yet, but I hope it is a growing trend. Donald Albury 17:55, 4 December 2024 (UTC)Reply

All else being equal, where there are multiple high-quality reliable sources available for a claim, we can lean towards more accessible sources. But often all else is not equal. For example, citing an older source over a newer one can give the impression that the claim may be dated and no longer reflective of the literature. And where equally reliable sources disagree, we definitely shouldn't disregard one perspective based solely on how accessible it is. Nikkimaria (talk) 00:21, 5 December 2024 (UTC)Reply

The text of WP:SPS currently says (1) Exercise caution when using such sources: if the information in question is suitable for inclusion, someone else will probably have published it in independent, reliable sources, and (2) Never use self-published sources as third-party sources about living people, even if the author is an expert, well-known professional researcher, or writer. Sentence (1) does not exist in the corresponding WP:RS/SPS text. Moreover, it implies that expert SPS cannot themselves be independent reliable sources, and it's not clear to me that the claim about probability is true. I propose deleting this sentence, or at least the portion of it that comes after the colon. Sentence (2) is a bit inconsistent with the corresponding text of WP:BLPSPS, which now includes the following exception: It does not refer to a reputable organisation publishing material about who it employs or to whom and why it grants awards, for example. WP:NPROF has similar exceptions in various places on that page; I'm not sure about other subject-specific notability guidelines for people. Should a corresponding sentence be added to WP:SPS? (FWIW, since editors disagree about what SPS does/doesn't encompass, I will likely start an RfC about that after the closure of the RfC on grey literature mentioned above. We could wait til that's over, as it may have broader implications for the wording of WP:SPS.) FactOrOpinion (talk) 22:19, 7 December 2024 (UTC)Reply

In regard to BLP, it seems like "It does not refer to a reputable organisation publishing material about who it employs or to whom and why it grants awards, for example" is not saying that it is ok to use an SPS on a BLP, just that this does not count as an SPS. If that is the case does it provide a counter example to "Never use self-published sources as third-party sources about living people, even if the author is an expert, well-known professional researcher, or writer"? - Bilby (talk) 23:22, 7 December 2024 (UTC)Reply

No, the recent clarification at WP:BLPSPS is saying that your employer is not a "third-party" to you. A press release is always self-published. A press release or a social media post saying "Bob's Big Business, Inc. is delighted to announce that they have hired Sam Sales as the new Vice President of Global Sales" is:

• non-independent (of the business; of the new VP of sales),
• self-published (written by the business, published by the business),
• primary (very close to the event, based on no prior publications),
• reliable (for sentences such as "Sam Sales was hired as Vice President of Global Sales in 2024"), and
• acceptable under BLPSPS.

WhatamIdoing (talk) 23:40, 7 December 2024 (UTC)Reply

Bilby, people have varied views on what is/isn't self-published (see the SPS definition discussion above). AFAICT, everyone agrees that (a) things like personal blogs, wikis, and social media are SPS and (b) things like traditional newspapers and book publishers are not self-published, but people disagree about (c) whether [Edited to add: and under what conditions] publications from advocacy organizations, universities, companies, think tanks, museums, learned societies, governments, etc. are SPS. The community needs to come to some consensus about (c), and that's why I'm thinking of creating an RfC about it. Depending on one's view about what is/isn't self-published, the BLPSPS sentence might mean that those publications are not self-published (and so don't fall under BLPSPS) or that they're self-published but are exempt from BLPSPS. FactOrOpinion (talk) 00:43, 8 December 2024 (UTC)Reply

No, "a press release is always self-published" is only true if the press release is published by a natural person, not an organization. It is, however, a primary source. Jc3s5h (talk) 03:03, 8 December 2024 (UTC)Reply

You made the argument above that a natural person (e.g., me) can self-publish something, but that when an organization (e.g., "WhatamIdoing, Inc.") does exactly the same thing, it's magically not self-published. I'm still not buying it. WhatamIdoing (talk) 07:17, 8 December 2024 (UTC)Reply

FactOrOpinion, rather than trying to keep everything perfectly in sync across multiple pages, maybe we should point to BLPSPS. That might involve adding works like "Per WP:BLPSPS," but would not necessarily involve removing any existing text.

About your (1): "Exercise caution when using such sources: if the information in question is suitable for inclusion, someone else will probably have published it in independent, reliable sources".

The change I'd suggest here is to change "published it in independent, reliable sources" to "published it in a non-self-published reliable source". The non-self-published part is the part that matters. That source will be independent 99% of the time, but it's technically not the problem we're talking about here.

About your (2): "Never use self-published sources as third-party sources about living people, even if the author is an expert, well-known professional researcher, or writer."

What we're trying to say is that even if the author is amazing, and even if the author has written a dozen non-self-published books on this exact subject, it's still 100% not okay to use their self-published works for BLPSPS purposes.

Imagine, e.g., that Gene Genealogist has written hundreds of articles and several books on genealogy. Gene has even written a whole book (HarperCollins, 2021, good reviews, decent sales) specifically on the subject of birthdates and is considered an expert in the field. Gene self-publishes this on social media: "Here's a fun fact: Paul Politician, Joe Film, and I all share exactly the same birthday. We were all born on the 32nd of Octember in 1960, making us all Baby Boomers. We are all 64 years old right now."

Using the list above, this post is:

• non-independent of self, but independent of the others,
• self-published,
• primary,
• reliable, and
• acceptable only for statements about Gene's own birthday/age/generation. It is unacceptable for statements about Paul or Joe.

The problem isn't that Gene isn't independent or reliable. The problem is that there was nobody to stop Gene if publishing this were a bad idea for some reason. And rather than say "Oh, it's never a bad idea" or "Do this only if it's a good idea" – and then have fights over whether this is a good or bad idea, then editors voluntarily placed a blanket restriction on ourselves: Don't use even independent expert SPS sources about other people. WhatamIdoing (talk) 00:01, 8 December 2024 (UTC)Reply

Re: (1), why do you believe that "if the information in question is suitable for inclusion, someone else will probably have published it in a non-self-published reliable source" is true? How would one even go about testing that?

Re: (2), you quoted a different sentence, one that I'm not questioning. I understand the BLPSPS condition. I also understand that there's significant disagreement about what is/isn't considered self-published. But my question was whether something like It does not refer to a reputable organisation publishing material about who it employs or to whom and why it grants awards, for example, which is currently in the text of WP:BLPSPS, should be added to the text of WP:SPS. Instead of your Gene Genealogist example, it's instead something like Notable Academic got a Notable Award from Learned Society, as published in Learned Society's newsletter. If one considers a newsletter to be self-published (and you do, though some others might not), then that sentence is saying that it's still OK to cite the newsletter for text on the academic's BLP re: the academic having gotten the award. And the BLPSPS text makes that clear, but the SPS text doesn't. FactOrOpinion (talk) 01:18, 8 December 2024 (UTC)Reply

(1) I don't think it's universally true, because "the information in question" could be basic information about the subject (e.g., all biographies should identify when and where the person lived; all articles about books should say what the book was about; all articles about species should say what kind of an organism it is), and sometimes part of that information might not be available anywhere else (e.g., a birthdate or birthplace; the author's explanation of what the book is "really" about; a scientific monograph from previous centuries).

However, outside of such basic information, if it isn't available in a non-self-published reliable source, it's unlikely to be WP:DUE for inclusion. Remember that this is an information statement rather than a rule (i.e., it does not tell you what to do), and it is not absolute, since it says this is only "probably" the case.

(2) My suggestion is that we solve the mismatch by having a single copy of the full BLP rules at BLPSPS (=not here) and that we indicate that WP:V does not have a complete copy of the full BLP rules. The alternative is that we have duplicated text, which will inevitably diverge over time. The available choices are:

• a single "official" text, and everything else points to it, or
• the maintenance hassle of resolving contradictions that arise between multiple copies.

The first approach is recommended in Wikipedia:Policies and guidelines#Content ("minimize redundancy"), but if you prefer ongoing maintenance hassles and the periodic need to de-conflict policies, then we could do that. WhatamIdoing (talk) 07:15, 8 December 2024 (UTC)Reply

I forgot to address your Learned Society example before I posted this. It's the same as the employer press release:

The Learned Society publishes a newsletter. The things we call newsletters are usually self-published, because they are usually being written and published by the organization; however, in some cases, the organization only sponsors the publication, and the publication's staff has editorial independence similar to a magazine (e.g., Newsletters on Stratigraphy). For simplicity, I'm going to stipulate that this is a self-published newsletter.

One of their self-published newsletters says that Prof. Notable Academic got a Notable Award from the Learned Society. This newsletter item is:

• non-independent (of Learned Society; of Notable Award; of Prof. Notable Academic [at least wrt this award]),
• self-published (written by the org, published by the org),
• primary (very close to the event, based on no prior publications),
• reliable (for sentences such as "Learned Society awarded the Notable Award to Notable Academic in 2024"), and
• acceptable under BLPSPS.

WhatamIdoing (talk) 07:29, 8 December 2024 (UTC)Reply

Re: (1), I still don't understand where you're getting the data from that allow you to conclude that "if the information in question is suitable for inclusion, someone else will probably have published it in a non-self-published reliable source." To me, it seems like a case where some editors believe this but don't have data to back it up and where it's unclear how one could gather such data. (Arguably, the vast majority of RS information isn't DUE, but that's a different issue, and is the case regardless of whether the source is SPS or non-SPS.)

Re: (2), your comment reminded me that there has been a similar conversation about the parallel texts in WP:SELFSOURCE, WP:ABOUTSELF, and WP:BLPSELFBUB. The people involved in that conversation decided that the best solution was "Remove the SELFSOURCE material in RS [and remove the BLPSELFPUB material in BLP]; retarget that shortcut to ABOUTSELF's location in V, and "advertise" the shortcut at that place instead; in RS [and BLP], summarize ABOUTSELF in a sentence and cross-reference it, without an unnecessary and potentially confusing devoted section."

Re: the Learned Society example, the only reason that it's acceptable under BLPSPS is because the exception is specified. It would not be acceptable under the current text of SPS. FactOrOpinion (talk) 17:09, 8 December 2024 (UTC)Reply

For (1), this is not a statement about collected data. This is a statement about editors' collective experience.

For (1), there are two classes of information that are suitable for inclusion. Those two kinds are:

1. Basic information about a subject, which belongs in an encyclopedia article because of the requirements of the genre. This includes, for example, writing boring information like "George IV (George Augustus Frederick; 12 August 1762 – 26 June 1830) was King of the United Kingdom of Great Britain and Ireland and King of Hanover from 29 January 1820 until his death in 1830" instead of jumping straight into the subjectively attention-catching content, e.g., "King George IV, nicknamed Prinny, was known for being obese, profligate, prodigal, and promiscuous."
2. Non-basic information about a subject, which belongs in an encyclopedia article because it is in desirable sources.

For the second category, the existence of suitable sources is definitional.

For the first category, the "probable" existence of suitable sources is the collective experience of editors. You might not be able to find every common detail (e.g., a complete birthdate may be unknown, or it may only be available in a self-published source) but if the subject actually qualifies for a Wikipedia:Separate, stand-alone article, you will be able to find enough non-self-published sources to meet the requirements of the encyclopedic genre, e.g., to place the subject in the encyclopedic context of time and place.

In re the Learned Society example, it was accepted in many thousands of articles before the exception was specified. The exception was written down to make the written rules more accurately reflect the community's actual practices. Per WP:NOTLAW, the written rules themselves do not set accepted practice. Rather, they document already-existing community consensus regarding what should be accepted and what should be rejected. The accepted practice was set by the community for more than a decade. Recently, we updated the written rules to document the already-existing community consensus that an employer is not a True™ third-party from its employees (for the purposes of this policy) and that an award giver is not a True™ third-party from its awardees (for the purposes of this policy) and that therefore self-published statements from these sources are not excluded by BLPSPS.

Perhaps this is the fundamental misconception. The written rules follow the community practice. The written rules document the community practice. The written rules are not supreme. The community is supreme. WhatamIdoing (talk) 18:20, 8 December 2024 (UTC)Reply

Thank you for having pointed out WP:NOTLAW. As much as I try to be familiar with (and abide by) the PAGs, I don't know that I've read all of them in their entirety, and I don't remember all that I've read.

Re: (1), I'd say that editors' experiences are a form of data, but it's unclear to me how the community determines what the collective experience is. There are ways to determine the consensus of a small number of editors (e.g., in an RfC), where those editors may make arguments based on what they believe about the behavior of editors not involved in the RfC, but that's still a tiny fraction of editors. I don't know how anyone could confirm in any reasonable timeframe that "In re the Learned Society example, it was accepted in many thousands of articles before the exception was specified." (You can't confirm it with any kind of straightforward search; instead, you'd have to individually look at the sources for that kind of info on many thousands of articles, and you also have no way of knowing how many editors left that kind of info out of yet other articles because they thought that the employer, awarder, ..., wasn't an acceptable source.)

It seems to me that in discussions, there's sometimes a tension between what PAGs say and what collective experience might be ("might" because I'm not sure how to determine "is"); that arose in the discussion about whether learned societies that are highly regarded within an academic field are wiki-notable, and is in play in the dispute about what is/isn't an SPS.

Re: your 1. and 2., I'd interpreted "if the information in question is suitable for inclusion, someone else will probably have published it in a non-self-published reliable source" as addressing both. And I've probably worked more on academic BLPs than other BLPs, which likely affects how I view all of this. FactOrOpinion (talk) 20:21, 8 December 2024 (UTC)Reply

Widely advertised discussions, including but not limited to RFCs, are assumed to represent the view of the whole community unless and until disproven, or at least seriously challenged.

Every decision is made by a tiny percentage of editors. Even if we had a thousand editors respond to a question, which almost never happens, that's still only 0.12% of last year's registered editors. But it's enough; we don't need large numbers of editors to make the right decision. If the first decision is wrong, we'll discover that over time and adjust.

In many cases, editors are learning by watching. They see that he cited the Oscars website, and didn't get reverted; she cited the Emmys website and didn't get reverted; they cited the Nobel website and didn't get reverted; and then rationally conclude that award websites must be okay. They will also see the Oscars website sometimes get replaced by a book, the Emmys website sometimes get replaced by a music magazine, and the Nobel website sometimes get replaced by a newspaper article. This means that when a single experienced editor shows up for a discussion, they're describing what they have seen get accepted (or rejected) by dozens or hundreds of editors across many articles.

Academic BLPs are challenging, because the accepted criteria give little consideration to why we require significant coverage in independent sources in the first place. I would expect it to be difficult to much past the stub stage for many academic BLPs without veering into OR territory. WhatamIdoing (talk) 21:07, 8 December 2024 (UTC)Reply

WhatamIdoing, thank you for your responses. You are helpful and extremely patient with all of my questions. FactOrOpinion (talk) 19:18, 9 December 2024 (UTC)Reply