User:Jclemens/CSD-RFC

"survived a prior deletion discussion" RfC
Should "survived a prior deletion discussion" in the lead of WP:Criteria for speedy deletion be altered? ~

"Administrators should take care not to speedy delete pages or media except in the most obvious cases. If a page has survived a prior deletion discussion, it should not be speedy deleted except for newly discovered copyright violations. [emphasis added to highlight disputed sentence]"

Text including the idea of XfD survival was originally inserted May 2007.

A strict reading dictates that the only possible exception is G12, Unambiguous copyright infringement. A reading that interprets the sentence as general guidance – not a rule – allows for exceptions, either unstated or clarified elsewhere. The proximate dispute involves G4, Recreation of a page that was deleted per a deletion discussion: whether one deletion discussion ending as non-delete makes a page ineligible for G4 going forward, even if its most recent deletion discussion closed as delete.


 * Scope of G4


 * 1) G4 should be applied only in cases where content is clearly recreated (perhaps include trivial changes) using XfD-deleted content
 * 2) G4 should be permitted in any situation where the evaluating administrator judges that the new content does not overcome the closing rationale given in the prior XfD where content on the same topic was previously deleted.


 * List of possible actions


 * 1) Text unchanged, endorse strict interpretation. Handle repeated recreation against clear consensus with WP:SALT create protection.
 * 2) Text unchanged, but reasonable interpretation of "should not be speedy deleted" – as opposed to must – allows G4 to be used. The disputed sentence is general guidance that specific criteria may override.
 * 3) Modify text or add a footnote.
 * 4) List all possible exceptions explicitly, adding any missing ones.
 * 5) Insert a qualification like "in general", "in most cases", or "exceptions specified in individual criteria".
 * 6) Change text to "If a page has survived its most recent deletion discussion".
 * 7) Remove the sentence entirely.

Previous discussion

 * Feel free to add relevant discussions. Please insert in chronological order and sign with ~.


 * WT:Criteria for speedy deletion, April–May 2007, around the original insertion
 * WT:Criteria for speedy deletion/Archive 18
 * WT:Criteria for speedy deletion/Archive 19
 * WT:Criteria for speedy deletion/Archive 39, September 2010
 * Deletion review, November 2010
 * WP:Deletion review/Log/2010 November 4
 * WT:Deletion review

Notifications

 * WP:Requests for comment (diff)
 * WP:Centralized discussion RfC: Discussion of "survived a prior deletion discussion" in lead of WP:Criteria for speedy deletion. (diff)
 * WT:Deletion review (diff)

Statement by Flatscan
The disputed sentence is reasonable guidance that probably prevents unnecessary tagging. I think that it is being interpreted too strictly, prohibiting legitimate tags under an ill-advised technicality.

If an article was deleted in a solid AfD and is recreated unchanged, the recreation should be deleted G4 – whether it has an old non-delete XfD carries zero weight. An AfD history containing a series of keeps, a borderline last AfD, and added sources may all be considered when choosing to delete or decline, but only the first has any connection to a previous non-delete.

Concerns that should be addressed at WP:Deletion review for specific cases and discussed here (WT:Criteria for speedy deletion) in general.
 * G4 is being intepreted too expansively
 * G4 ignores that WP:Consensus can change, since it only considers the page's state


 * List of possible exceptions


 * G4, a few examples given in previous discussions at DRV
 * G6 Technical deletions: A very new user creates a WP:Requests for adminship, but does not transclude it. The page is immediately MfD'd and is kept as the right of any user in good standing. The RfA is deleted G6, after a period when the user does not respond to questions about the page. (AN discussion, WT:RFA follow-up)
 * G8 Pages dependent on a non-existent or deleted page: After its AfD is closed merge (non-delete), an article is merged and redirected to a list article. The list article is later AfD'd and deleted, and the redirected article is deleted G8. (AfD 1, AfD 2)
 * G9 Office actions
 * U1 User request: User:Example's user page includes content that another user believes to be a violation of WP:User pages. That user nominates it at MfD, but the page is not found to be an egregious violation and is kept. Example later retires and tags his/her user page with U1.

Flatscan (talk) 04:56, 15 November 2010 (UTC)

Statement by Jclemens
The current problem is not that there is something wrong with the strict interpretation of CSD as it applies to G4, but that G4 has become routinely overused. G4, like all speedy deletion criteria, is to be reserved strictly for situations where no editor (acting in good faith and in agreement with our various policies, of course) would disagree with the deletion. That is the core of CSD: an abbreviated process should only apply when no one (same disclaimer as above) disputes the outcome.
 * If consensus can change, such that content which was kept at XfD on one occasion is later deleted, it is reasonable to believe that consensus can change back at some future point. Thus, when content is kept at XfD once, deleted after a subsequent XfD, and then recreated, it should immediately be sent back to XfD.  Only within an XfD can a discussion assess whether circumstances or consensus has indeed changed.  XfDs have had low tolerance for disruptive or apparently bad-faith renominations, so the effort invested in a pointless XfD is not costly.
 * G4 is for "sufficiently identical and unimproved" content. At DRV, we have seen articles on a topic rewritten from scratch deleted as G4's, and many editors in good standing endorse the deletion, opining that any recreation which doesn't adequately address the reason the article was deleted in the first case is fair game for a G4.  This level of discretion is unheard of in any other speedy deletion category.  Speedy deletion should never include this level of administrator discretion; the closest any other speedy deletion criterion comes is administrator understanding of what constitutes an "attack" for G10 purposes.
 * When an item has been repeatedly recreated against current consensus, G4 is not the only tool in the toolbox: protected redirects and create protection can each prevent disruption when it arises; in the absence of actual disruption, G4'ing a good-faith effort to create or re-create an article shortcuts the discussion that necessarily arises from a deletion discussion.

Flatscan has done a good job of listing obscure circumstances where reading the text literally will produce odd and unworkable results. I support a rewording to indicate that U1, G9, and uncontroversial G6 or G8 deletions are permissible. My primary concern has been, and remains, that G4 is being used to inappropriately shortcut good-faith discussion.

Every Wikipedian who's been around for a while knows that it is possible for XfD to reach inconsistent conclusions. Allowing G4 in cases where XfD has given differing conclusions on two separate occasions, regardless of which came first, is not the right answer. It is my hope that we will clarify community expectations of G4 and the number of G4 deletions brought to DRV will plummet.

Jclemens (talk) 08:03, 15 November 2010 (UTC)

Statement by VegaDark
I would propose altering the statement in question to read as If a page has survived its most recent deletion discussion, it should not be speedy deleted except for newly discovered copyright violations. (bolded portion being the change in question). The strict interpretation of the current wording has a number of problems, which I will go through in detail in the following examples.
 * Example 1: Very old CfD resulting in keep where the only contributor other than the nominator is the category's creator, with their rationale to keep is "Give it a chance to populate more." Admin closes as default to keep due to no consensus. Three months later, a second CfD is brought up with 10 users in unanimous agreement to delete, all with their own well reasoned arguments for deletion. Fast forward to today, the original creator of the category discovers after all these years that the category they created was deleted, and recreates it identical to the original. Under the strict interpretation of the current wording, this would require a brand new CfD instead of a G4 deletion.
 * Example 2: Article about an athlete is brought to AfD, before we had very good guidelines on the notability requirements for inclusion of athletes. The debate is closed as no consensus defaulting to keep. Shortly thereafter, WP:ATHLETE is created, setting out the guidelines that article x unambiguously no longer meets the notability requirements for inclusion, and a subsequent AfD of the article results in a unanimous delete now that the new guideline is out. Another user comes along and recreates the article (with no new claims to notability) 3 minutes after the article is deleted from the first XfD. Under the strict interpretation of the current wording, we would have to bring up another AfD for this article because of the prior no consensus keep, resulting in a minimum of 7 more days at AfD.
 * Example 3: A lengthy, hotly contested debate is going on, with reasonable arguments by both sides. Nobody would fault an admin closing the debate either way they close it. As it so happens, the content in question at one point in the past had an XfD resulting in renaming it to the current name, and although the original proposal was to delete, the discussion focus shifted to renaming. An admin comes in and closes the hotly-contested debate as delete. Shortly after, a user didn't like the close, so they immediately recreate the category, knowing it isn't G4-eligible due to the previous XfD resulting in rename (Under the current wording), hoping this time that although the arguments will be identical to the original discussion, the closing admin decides to keep it instead. The user's attempt to game the system pays off, as the new admin closes the debate as keep.
 * Example 4: Same scenario as example 2, except now the article is sent to AfD again, resulting in delete again. At this point the article has 2 deletions in its history, not enough to salt in my view. A user comes along and creates the page a third time, still substantially identical to the original. Once again, under the current wording, this would have to go a full 7 days at AfD. Presumably, at this point the closing admin would likely salt the page. However, admin B comes along, who happens to be from this athlete's hometown and is shocked that this high school star doesn't have an article. Admin B notices that the page has been deleted before and salted, but figures that the old versions of the page must have not talked about the high school stats the athlete had, so they recreate the page once more, still substantially identical to the original (although this fact is unbeknownst to Admin B, who presumably acted in good faith in creating the article). Salting admin goes to Admin B's talk page informing them of the situation asking them to delete the page, but Admin B is annoyed they didn't see the original deletion discussions so they refuse, demanding it be brought back to AfD (again!) so that they can participate in the debate. Under the current wording, this article would, yet again, be required to come to AfD. Not including the original discussion ending in no consensus and the subsequent deletion result, this scenario would create 3! additional trips to AfD, resulting in a minimum of 21 more days of clogging up deletion debates.

Under my proposal, none of these examples would be an issue. G4 would be allowable as soon as the most recent debate resulted in delete, which is what I always believed the criterion was meant to reflect, and is what I have followed since becoming an admin. Every issue with a G4 deletion can currently be brought up at deletion review, including a determination if consensus has changed, or if an article really met the G4 criterion or not. Salting is all well and good, but as I show in my example, I would submit that an admin should not salt until there is at least 3 deletions in the page history (which still allows for 14 days of unnecessary XfD debates), and would not prevent an admin from recreating the content in question (however rare that may be).

A strict interpretation of the current wording of G4 creates a double standard for how we treat deletions. Why should a page with an extremely old xfd resulting in keep be treated any differently than a page with no deletion history? Why should we be allowed to G4 one but not the other? The very nature of the G4 deletion criterion is so we don't have to repeat debates on XfD. It gives piece of mind to those of us who successfully argue for the deletion of something that someone wanting to bring it back is going to have to go to DRV first. A strict interpretation completely undermines this, which is why I believe the wording should be changed.

VegaDark (talk) 09:52, 15 November 2010 (UTC)