Wikipedia talk:Contributor copyright investigations/20240901

How do we handle the copyright issues that were identified (and removed) during the previous (incomplete) investigation of this user, but which have not been revdeled (probably because RD1 was not available at the time)? Should those edits be revdeled now? Where the problematic material was removed at the time, such revdels would cause limited harm, only affecting edits made between 2006 and 2010. Renerpho (talk) 08:53, 2 September 2024 (UTC)[reply]

It depends? I know on other CCIs, where the material was removed pretty soon after it was added and I can prove it was copied, I tend to ask for a revdel just so nobody looking through the page history gets any bright ideas about re-instating the massive block of seemingly OK prose that somebody "unjustly" removed. But that's only if I can prove the material was copied from a certain place. If I can't, then the admins aren't allowed to revdel it so I just don't bother. GreenLipstickLesbian (talk) 21:09, 3 September 2024 (UTC)[reply]

@GreenLipstickLesbian: Most of GrahamBould's copyvios were so blatant that they're quite easy to prove. What I am concerned about are edits from 2009/10 in which copyvios had been removed. Such edits would clearly go along with an RD1 if they were made today; but because RD1 did not exist at the time, this did not happen. Should we go back into the article history and retroactively revdel those old edits from 2006-2010? The copyvios remain accessible in the article history until we do, so I believe that revdels are without any alternative in all such cases. But that's just my opinion. Is there a "procedure" for this?

The article that sparked this CCI, Jack Churchill, needed an RD1 that spanned over 1,200 edits and nearly 20 years, essentially wiping out the entire article history. Most of this collateral damage could have been avoided if the article had been dealt with in 2009 (even with the limited tools that were available at the time). I want to make sure that we only do this where it's absolutely necessary -- that is, in the cases that were not dealt with in 2009 and where the infringing material still remains in the article. Renerpho (talk) 23:23, 3 September 2024 (UTC)[reply]

@NotAGenious and GreenLipstickLesbian: This is a huge list, so just starting somewhere, like in 2009, may not be the best idea. I think an effective strategy would be to go through the list as remaining in 2009.[1] To see if this is the case, I have checked the first two articles (Fish anatomy and Anostostomatidae), and found copy-vios in both of them. The edits for #3 (Rhaphidophoridae) haven't been checked yet. #4 (List of schools in New Zealand) doesn't seem to be listed in the current CCI. It appears that this article was split into multiple lists since 2009, some of which contain remnants of edits made by GrahamBould, all of which may contain copyrighted material. Renerpho (talk) 11:22, 3 September 2024 (UTC)[reply]

Special:Diff/69202515 looks suspicious to me. This is #6 in the list from 2009. Like with #4 above, I cannot find that article listed in the current CCI. Their edit to article #9, with Special:Diff/89281707, seems to contain no copyrighted material, but many similar edits are listed in the CCI, and about 8% of the present day article was authored by GrahamBould. I don't understand why those articles aren't listed. Can someone explain this to me? Renerpho (talk) 11:51, 3 September 2024 (UTC)[reply]

I was hoping that articles tagged with a Y in the old investigation could be removed from the current one. The example Diaphanidae shows that, just because an article was tagged with a Y in the 2009 CCI, that does not mean that the copyrighted material was actually removed. Renerpho (talk) 13:16, 3 September 2024 (UTC)[reply]

@Renerpho Yeah, this is a pretty strange case in the sense that it's half-done but not quite. I'm afraid I don't have the experience or knowledge to put forth a good plan- I just decided to mark off a few edits that I saw somebody else had already dealt with. Which, until somebody comes up with something cleverer, is probably how I'm going to have to tackle it. But consulting the 2009 list, and working through the checked entries there seems like a great place to start! But if you're right about some articles being checked, but copyvio not being removed, then I can't think of a nice way to automate it. (Which sucks, I know).

And, concerning certain articles not showing up in the CCI- to create this page, @Mer-C used an offline script to print out every single non-minor edit over a certain byte size. They then culled it, using another script, to try to remove obviously OK edits. Stuff like adding wikilinks, changing links, ect. I'm not entirely sure how it works or what trips it, however, so sorry I can't provide a better explanation! But that's why those articles you pointed out aren't present in the current CCI. They seem to be in the first version, before the culling. And, as for other similar diffs still being present- yeah the autocull script isn't perfect, and it lets obviously OK edits through a lot. But those are, at least, easy to manually clear. GreenLipstickLesbian (talk) 21:06, 3 September 2024 (UTC)[reply]

What I am suggesting is to first work through the unchecked entries from 2009, as found here. The chance to find a copyright violation that wasn't identified during the first investigation is highest in those articles. If an RD1 is necessary, there's a chance it will cause a lot of collateral damage, and finding those cases as soon as possible seems prudent if we want to limit that damage.

As a second step, I think we should go through some of the entries that were checked towards the very end of the 2009 investigation (like Diaphanidae). I still hope that this is where people just abandoned the project, even leaving some already identified copyvios in place. I don't believe that the 2009 investigation was sloppy, but it was abandoned suddenly. Renerpho (talk) 23:33, 3 September 2024 (UTC)[reply]