Call for comments Edit
Hey folks, it would be very useful if some people would comment on this policy. We really need to have some sort of official (i.e. non-WIP) policy concerning when and how to delete copyright infringement. Please comment! -- Dan Carlson | Talk 22:59, Sep 2, 2004 (CEST)
- I'm quite happy with the current policy. The only thing I would suggest is an extension to the current "delete on sight" rule, to cover all copyvios discovered by admins. Either that, or reduce the length of time a page needs to remain on the PCI page. Normal Archivists, since they lack the deletion capabilities, can list them at the page, and an admin can delete them when they see the listing. It prevents the backlog which tends to happen - since we do seem to get a lot of copyvios popping up.
- And, where an article's content has been partly or completely replaced by copyvio material, expressly noting that admins can delete the copyvio version, and immediately recreate the article with the content prior to the introduction of the offending material. I've been doing this on several occasions. -- Michael Warren | Talk 23:12, Sep 2, 2004 (CEST)
I'm not sure I'm comfortable with the "delete on sight" practice, because there are a few cases where it's at least conceivable that the text is being contributed by the original author. (Of course this isn't likely to be the case, but I'd like to keep an "innocent until proven guilty" mentality about the process.) But I'll add a paragraph about the procedure for cases of copyvios replacing valid content. -- Dan Carlson | Talk 17:06, Sep 4, 2004 (CEST)
Copyvios in page histories Edit
Although the current policy for copyright violations in article histories is to delete the entire page, I am concerned that this policy may be too extreme, especially because it has a nasty side effect. If an article has part copyvio content and part valid content, plus it has an existing history (Galaxy-class is a perfect example), then the act of recreating the page and deleting the old version with the copyvio in the page history will actually wipe the record of the individual contributions to the article. Sure, the content may be preserved, but the attribution of who did what is not preserved.
After reading a Wikipedia discussion on the matter, I think that it should be considered acceptable to simply revert pages to previous non-copyvio versions and allow them to linger in the page history.
If worse comes to worst, there is a means by which we can delete individual revisions of an article by searching the old table and identifying the revision by the oldid variable (I think, anyway). If anyone complains about copyvios in the page history, then we can deal with it -- if the problem arises. In the mean time, I don't think it's fair to all the valid contributors to delete the record of what contributions they made. -- Dan Carlson | Talk 17:33, Sep 4, 2004 (CEST)
- I agree with you in part, Dan. However, cases such as Galaxy-class are extremely rare - most copyvios are directly posted, or added to articles which are fairly untouched (in fact, with the copyvio in question being an official Paramount publication, under the policy, I could have deleted the page immediately - such a copyvio should certainly not be allowed to stay in the history). Since this entire project hovers on the edge of acceptable fair use as it is, I think we need to be very strict on copyright violations. Wikipedia does not have these problems per se, so is more capable of following this idea.
- In cases similar to Galaxy-class, but where the copyvio is not Paramount originated, I would support a simple rollback. However, this case is not that. In that case, prior to deleting the article, I intend to copy the original history to a subpage, or even the Talk page, in order to preserve it. Where the copyvio is a new article, or replaces/adds to an article with little contribution, it should follow the standard procedure. -- Michael Warren | Talk 14:02, Sep 10, 2004 (CEST)
Okay, so let me see if I get you correctly: you think that plagiarized Paramount works must be deleted, but that other copyvios can be left in the page history unless (until?) someone complains. That sounds fair enough, though I'm not sure I agree about the danger. I still don't like the idea of deleting an entire page history set -- depending on how you interpret it, recreating the page could be considered a violation of the Creative Commons License, because attribution to previous contributors' contributions are no longer being acknowledged. I think it's important to be able to preserve as much of the page history as possible.
At the moment, I'm contacting Erik to see if I can figure out a way to delete individual versions of a page without causing more problems for the rest of the database. If that's the case, then I think that we can go your route and delete any Paramount-derived copyvios. If not, I think it's safe enough to toe the line and just roll back all copyvios without deleting anything. -- Dan Carlson | Talk 18:36, Sep 10, 2004 (CEST)
Update to centralize revisions with violations Edit
I think the procedure for dealing with the deletion of copyright violations should involve moving the revisions that are infringements to a centralized location so they aren't accidentally restored later at their original location, for example in a page merge. This has already been done a few times already, and this would just be codifying it, and possibly creating a better location than the sandbox, if I'm remembering correctly. - Archduk3 05:51, January 14, 2014 (UTC)
- I'm not sure that is necessary. The current version of the procedure states that simply removing/reverting the offending stuff is acceptable until a copyright owner explicitly complains about that, and I think this is still valid. If a copyright owner complains about stuff in a page history (which I can't remember happening in the last 10 years, although it might have been handled in private without me being aware of it), I consider that to be a problem between that copyright owner and Wikia, not a problem that we will have to handle.
- Also, I've just found a page on MA/de, where removing a copyvio by deleting revisions has actually led to a bigger problem. In that case, it now looks like user X wrote most of the article, although he just deleted all revisions after the addition of copyvio text and then manually re-added the non-copyvio text, thereby claiming the work of other users as his own.