Wikipedia talk:Postponed deletion

What's the point of this thing? AFD isn't for cleanup, and an article isn't going to be (or at least, shouldn't be) deleted if there are correctable problems that don't make the whole article worthless. Can anyone please point to any afd discussion where an end result of 'postpone' would've actually bee a better outcome?

In the end, it gives the exactly same end result as saying 'keep', but adds another arcane detail that newcomers to the process won't be familiar with. - Bobet 18:35, 13 May 2008 (UTC)[reply]

Many users don't read the talk page, and tags are either swiftly removed without comment or added in such numbers as to be more like vandalism than help. As a result, the only way to improve an article at the moment is to AFD it. As Masem points out, this means that some articles that could have been cleaned up are deleted. Conversely, and I confess this is more aligned with my concerns, a lot of AFDs end up with a 'keep' result because the interested editors claim it could be improved, but the improvements never arrive, leaving us with a horrible mess of an article. In both cases, the right course of action is to keep the article for a time in order to ascertain whether the article can be improved, and to delete it if and only if it isn't improved. That's what postponed deletion would achieve. Percy Snoodle (talk) 18:45, 13 May 2008 (UTC)[reply]

Another way to improve the article would be to improve it (sorry). Like I said, AFD is for stuff when simple cleanup isn't possible. If you could point out to a specific case maybe I could see some value in this proposal, but I currently can't think of any scenario where it would be useful. - Bobet 18:52, 13 May 2008 (UTC)[reply]

AFD is supposed to be for that stuff, but until there's a way to get people to listen to requests for cleanup that doesn't involve threatening to delete the page, AFDs will be started for cleanup purposes. That said, I think "cleanup" may be being used here in a slightly unusual way; I wouldn't describe notability concerns as a cleanup issue. Regarding examples, you'll see a slew of them if you trace the links to WP:HEY, and I'm sure Masem can fill you in with ones that have gone the other way. Percy Snoodle (talk) 19:02, 13 May 2008 (UTC)[reply]

The only way to improve an article at the moment is to AFD it? Percy, you need go read the editing policy, and also WP:ATD from the deletion policy. And you may also find WP:WIP educational. --Pixelface (talk) 10:51, 17 May 2008 (UTC)[reply]

Thanks, Pixel. You know that's not what I mean. Percy Snoodle (talk) 11:13, 17 May 2008 (UTC)[reply]

I know this discussion may seem over, but I have revived the discussion. To for the reason, see: Wikipedia:AFDPP. —CodeHydro 19:49, 31 August 2010 (UTC)[reply]

Mechanism

My suggestions for mechanism.

Use a template as per the one on the article page
A page with the template subst'd is automaticaly added to [[Category:Postponed AfD debates]]
Arrange for User:DumbBOT or similar to produce an output on WP:PADSUM in a similar way to WP:PRODSUM
Expired PADs can be checked via PAD patrol or editor checks for either removal of the template (and cat), or relisting on AfD.

Hope this makes sense.Gazimoff ^Write _Read 21:36, 13 May 2008 (UTC)[reply]

This seems completely logical and well thought out. Two thoughts occur:

I think it would be a good idea to have a method of manually logging the addition of a page. In some situations (for example, if a maint template or AfD notice has been improperly removed previously), there may be reason to believe that the template will be removed before the bot can pick it up.
We may need to split the category into week of addition, for easy handling. But we can probably postpone (ahem) this for the time being, and deal with that problem if/when it arises.

Jakew (talk) 22:06, 13 May 2008 (UTC)[reply]

I agree. To elaborate:

The admin, when closing the AfD, closes it with reason postpone
The admin removes the AfD tag and replaces it with {{dated PAD}}. This would be a manual process, but by adding the tempate, you automatically add the article to the PAD categories.
The template follows the layout on the article page here, in a similar design to {{dated prod}}. This template can automatically include the article in a PAD category for the day/month/year the PAD started, to make handling easier with no bot required.
Either the category page or WP:PADSUM (the bot-generated summary for easy PAD patrol) can be transcluded onto this article page for a dynamically updating list of open PADs.

Hope this helps, Gazimoff ^Write _Read 22:30, 13 May 2008 (UTC)[reply]

Discussion

Oppose. Too easy to game the system. Nothing would ever get closed because every SPA and his brother would just stick a postponed request on the article. Corvus cornix_talk 19:01, 15 May 2008 (UTC)[reply]

We originally looked at this possibility, as it's a legitimate fear. The conclusion was that unless there have been major improvements to an article, you can only postpone once. Further, postponement is at the discretion of the closing admin, preventing abuse by SPAs and similar. You may ask for postponement, but you might not get it. This should eliminate opportunities to game the system, either through frivolous postponement requests, or through chain-postponement requests. Both these elements are on the project page already. Hope this helps, Gazimoff ^Write _Read 19:17, 15 May 2008 (UTC)[reply]

Reluctant support with the caveat that after this has been tried for a while, it can be stopped if it fails somehow. I like the idea that it is not automatically granted (to avoid obvious abuse). I suppose that misuse of this possibility (say, asking for postponement on twenty AfD's, and then not improving even one of them) can be considered disruption, if necessary? Just trying to avoid abuse of the system. Fram (talk) 08:31, 16 May 2008 (UTC)[reply]

I'd call that disruption, unless there was a very good explanation. It could happen once, but I think if an editor were to do that, (s)he would find it difficult to do it a second time, because of this: "Other editors may discuss the merits ... basing their comments on ... the reliability or dedication of the editor asking for the postponement". Jakew (talk) 11:00, 16 May 2008 (UTC)[reply]

Support with comment with the caveat that a full AfD may not be needed if the article remains unchanged. Really, "Postponed" means 'As it stands, this should be deleted, but someone claims it can be fixed within a month.' So, if it's not fixed within a month, why is there a need for a full, additional AfD process? Jclemens (talk) 18:28, 10 June 2008 (UTC)[reply]

Support as I am increasingly coming upon AfDs for which I can find sources, but it takes more than a paltry five days to do so (I like many others have other repsonsibilities beyond Wikipedia after all). We're here to build a comprehensive reference guide after all and if articles have realistic potential, it's far more important that we try to fulfill that potential then just remove it altogether and thereby diminish our coverage and value. Sincerely, --Le Grand Roi des Citrouilles^Tally-ho! 04:26, 17 June 2008 (UTC)[reply]

Oppose. This is a solution looking for a problem. We already have a traditional system that allows for all the time in the world to prove notability (barring BLP, soapbox or similar precluding issues). Ask an admin to undelete and move the article to userspace. Take all the time needed to find sufficient sources prior to moving the article back into mainspace. It really is that simple and avoids yet another unnecessary process. Vassyana (talk) 19:35, 2 July 2008 (UTC)[reply]

If the AFD process itself wasn't "broke" - in that many nominations have no previous discussion, tagging, or any other attempt at content dispute resolution before entering AFD - this step wouldn't be needed. While I agree that the option to userify a page should be encouraged more (I see very little of that going on or encouraged), the removal, even if temporarily, may not help other pages that link to it. Also, there is an implicit review control here. In the userify case, the fixing editor may not be able not really be able to fix the problem but puts the page back as it was, and then a whole new AFD process will likely come about, but maybe some long time after that is done. The postpone approach provides a guaranteed review, and thus (hopefully) preventing the article from ever entering AFD again. (Of course, we should have mechanism that allows userified pages to be reviewed before they are uploaded into place, that would effectively be the same thing). --MASEM 12:51, 4 July 2008 (UTC)[reply]

It would be easy enough to move protect userified pages to protect the kind of abuse you point out. Replying to the rest, I don't think the deficiency in using current processes will be helped in any way at all by adding more process. (Caveat: I am strongly opposed to the proliferation of rules and processes, as what we have is already a hulking mess and I've yet to see a situation where a case made for new rules/processes that can't be addressed by the current framework.) What would be helpful is raising a discussion on the village pump, or WT:DELETE, about your concerns. The problem would be developing language that would permit sysops to speedy close XfDs in the absence of prior reasonable and good faith efforts to find sources, clean up the article, merge the article, or so on. The reason it would be difficult is because wikilawyer encouraging phrasing would need to be avoided (i.e. previous good faith non-deletion efforts are not a hoop to jump through) and it would need to exclude cases where no amount of good faith efforts can fix the problem without making the exclusion an excuse to bypass the good faith effort requirement. Regardless, this is an issue that can be addressed in the current framework of policy and process without creating additional processes or rules. Vassyana (talk) 13:59, 4 July 2008 (UTC)[reply]

Oppose I warmly applaud the sentiment behind this idea, but in its particulars I am in agreement with Vassyana. (I suggest this guideline be marked as historical as it clearly will not be able to generate consensus.) Eusebeus (talk) 16:48, 4 July 2008 (UTC)[reply]
Oppose My arguments here are substantively unchanged from those at WT:Deletion_of_pages_under_construction. Basically, marginal notability can be established in 5 minutes, not five days. Adding unlimited debate is a motivation and a mechanism to game the system. Userification provides the same remedy without adding the problem of retaining an article that may not meet guidelines in mainspace. Protonk (talk) 17:33, 7 July 2008 (UTC)[reply]
Comment - I'm very surprised that none of the comments above seem to mention WP:ARS. If you look at this as a substitute to the {{resscue}} tag, it's almost the same thing. The difference is important though. Instead of only being used to signal editors that an article that may be on the verge of deletion needs some attention, this tag also delays the AfD. I do worry about delaying the AfD to the point where the AfD process will become very drawn out. While WP isn't on a deadline, I think it's obvious that there's a flow of new articles that need attention and editors can only give attention at a certain rate. If the rate at which attention is paid to articles slows, we'll eventually have a giant backlog of articles that need attention that WP won't be able to keep up with. If this does replace the ARS {{rescue}} tag, it would also take out the membership status of ARS (as ARS would't be need) and solve an allegedly ongoing problem with that project. I see this as a totally viable solution to the alleged problems with ARS. OlYeller^Talktome 15:01, 19 May 2009 (UTC)[reply]

Also, this is a discussion section, not a !vote section. This section is for a discussion on how to improve the guideline. If you think that it's not viable then a !vote later would be the place to say so. Otherwise, please give constructive criticism. OlYeller^Talktome 15:07, 19 May 2009 (UTC)[reply]

The biggest problems that I see occurring are that 1. An editor will add the postpone template to an article with no justification just as often as the {{hangon}} tag gets added and 2. That postponing an AfD for a month would create a backlog of articles that need attention that is so large, Wikipedia won't be able to keep up. My suggestion for number 1 is to state that it's only to be used to articles in AfD and that there must be evidence presented somewhere (the talk page) that the article could possibly be saved (Google News search for instance). This would keep articles that clearly cannot be saved from getting tagged and postponed. For problem number 2, I suggest shortening the postponement to an addition week (on top of the ~week it is to spend in AfD). OlYeller^Talktome 15:07, 19 May 2009 (UTC)[reply]

Wikipedia is not on a deadline

If an article is notable but lacks the references to show it, then it should exist for a very long time. If it's definitive that an article does not comply with wikipedia, then no amount of time will help it. Obviously, there is a gray area. But the idea that we can put everything on a simple 4 week deadline is inconsistent with the rules of the wikipedia, in my opinion. I think there has to be room for interpretation in what's an appropriate amount of time. "If, after a reasonable amount of time, these issues have not been adequately addressed from the ((OLDID/current version)), this article may be renominated for deletion." We don't want instruction creep, and we don't want the letter of certain guidelines to interfere with their spirit. If something has a strong probability of being notable but lacks references, we ought to be more lenient with the amount of time required to fix it. If something has a strong probability of being non-notable, we might be less lenient and expect more immediate efforts. Just my two cents. Randomran (talk) 16:12, 16 June 2008 (UTC)[reply]

While its true there is no deadline, WP also encourages editors to be bold, and extremely bold editors may simply wipe out content without waiting. We need a balance between these two extremes. We need more than five days, that's a given, and given that WP is a volunteer project, anything less than two weeks is unreasonable. However, waiting too long allows editors to lapse into complacency and forget they have to actually have to show it - something like 8 weeks or more seems awfully excessive. But maybe that would make sense is to allow the "closing" admin to set a time based on the comments: if the topic really appears non-notable, give the editors more time to find references, but not too long.

I will also point out that it is important that the effort is a good faith one. If someone finds a reference to a reference, the only accessible copy they can get is in a library 200 miles away, I'd give them the benefit of the doubt that it exists if they can't get there in 4 weeks as long as they report on it. --MASEM 16:56, 16 June 2008 (UTC)[reply]

Yeah, I think there should be wiggle room. Guidelines ought to delegate discretion to the community and the closing administrator. We want to be clear that this isn't about putting wikipedia on a deadline. It's giving someone a reasonable amount of time to prove that the article should be on wikipedia at all, in the face of evidence that it shouldn't. I might add to this guideline: "The postponed timeline is not a deadline to improve the article. It's a chance to challenge the existing evidence that the article does not belong on wikipedia. Often the best evidence is to improve the article itself." Randomran (talk) 17:10, 16 June 2008 (UTC)[reply]

Well, to make it more generic, I would replace "the article does not belong on wikipedia" with "for the article's proposed deletion", given that the former is just a tad harsh and also its not always the case that an article is at AFD for notability (though cleanup should not be a reason for deletion, some cases are extreme but necessary). I would still default to 4 weeks, but allow admins to consider more time for tougher cases. --MASEM 17:23, 16 June 2008 (UTC)[reply]

I think that's fair and means basically the same thing. Although, let's be honest: aren't AFDs for articles that don't belong on wikipedia? Anything else is just cleanup. Randomran (talk) 17:45, 16 June 2008 (UTC)[reply]

Practical problem

Let's just say there have been numerous "votes", with no consensus, or a consensus to delete. Then, at around the 10th comment, someone cites WP:POSTPONE with a reasonable justification. Overwhelmingly, there is no consensus for postpone. Postpone is almost invariably going to be a minority opinion, no? Something that occurs after it looks like there is enough of a consensus that the article is at risk of deletion. If there is never going to be a consensus for WP:POSTPONE, then under what circumstances do we postpone? Is it at the discretion of the administrator? I'm having trouble envisioning how this would work. Randomran (talk) 06:34, 17 June 2008 (UTC)[reply]

POSTPONE would be likely a minority !vote. The way I envision it, a user steps forward (at any time) and asked for postponing. Unless someone brings up good reason that postponing is not appropriate, the AFD admin should consider that !vote outweighing any delete !vote. That's the whole point of the fact that this basically puts the AFD on hold to give time to fix it up, and then reopens at a later date. --MASEM 12:26, 17 June 2008 (UTC)[reply]

So it's almost a veto. I think that's a little powerful. But I think admins and others will understand the difference between a fair request and merely gaming the system. I think this point about the strength of voting "postpone" and the last point about this not being a deadline should both be incorporated into the article, since those two points weren't clear. Randomran (talk) 14:59, 17 June 2008 (UTC)[reply]

It is powerful, but that's why it needs to be used with care. We should be able to judge whether the !vote is sufficiently by assessing the rationale. Ideally, it should be explain why postponement is reasonable and realistic, and giving a rough overview of how the editor intends to address the problems. For example, a strong postpone !vote might look like: "this topic was quite well-known in the late 1860s,[ref] so I'm going to look through archives of the New York Times from that period and rewrite the article based on what I find." If this is the case, then it is likely, I suspect, that a consensus will develop around the proposal. Jakew (talk) 20:54, 17 June 2008 (UTC)[reply]

I'm not sure about the consensus part. A lot of people do a hit-and-run on AFDs, and WP:POSTPONE is likely to be a compromise that comes along after a lot of other votes. So basically, we're gonna see a WP:POSTPONE vote with very few people having a chance to respond to it, let alone agree with it and build support for it. I'm okay with it being a veto, but then we need to explain just how strong this veto power is and the kinds of factors that administrators should take into account when closing the AFD. Randomran (talk) 21:14, 17 June 2008 (UTC)[reply]

Some thoughts

I think this is a good idea but do have a few thoughts and concerns. I do think four weeks seems to be quite a good length - not too long or too short.

Firstly I would want to make it clearer that the editor who requests the postponement must give a reason and some sort of explanation of how they are going to address the deletion concerns.

Also would hope this would not be used unless the closing admin felt there would be a consensus to delete otherwise. So it would not be used as a way of avoiding closing the AFD as a no consensus where the discussion merits it. Would prefer something to be added to this effect.

How would the list of articles be maintained? Hope it would be the job of the closing admin to add it to the list and not dependent on the template remaining on the article. Obviously if the list was dependent on the template remaining on the article this would be too easy for vandals, etc. to remove and for the article to be lost track of. This will only work if the closing is sure that the article will be reconsidered at the end of the period.

At the moment it seems to say it cannot apply to any biographies of living persons, would prefer this was amended to say "(such as BLP related policy violations)" to give a bit more leeway to the closing admin.

Lastly what happens at the end of the 4 weeks? I understand if editors feel the concerns have not been addressed it is renominated for AFD. How does the template get taken off the article if the problems have been resolved? Who will make that decision?

Apologies if some of these have already been addressed. As I said first I think this is a good idea but some clarifications are needed. Davewild (talk) 17:12, 21 June 2008 (UTC)[reply]

It would be expected that a user asking for a postpone would justify it and more importantly be the lead on it (thus avoiding drive-bys). The strength of this postpone reason is part of what is judged against the remaining consensus; if all but that !vote says "delete" but the editor insists to get time to find sources, we WP:AGF, and give four weeks; if nothing changes, its not going to change the likelihood of deletion. The list is suggested to work with a bot that non-template adds it to a list as to avoid POINT-y type template deletions, separating such in to day-by-day categories. Editors would patrol this list, when it gets past the four week (or what ever deadline is set, per above suggestion), and if they feel it fails to have improved per the postpone reason, it goes up to AFD again, with very low likelihood of another postpone being allowed (no indefinite "crying wolf" on these). If the editor feels it has been resolved, I think we want a {{ArticleHistory}} action that includes a brief reason why the editor felt it passed, just to log it, or something along those lines, so that there's accountability for the reviewing editor. --MASEM 19:15, 21 June 2008 (UTC)[reply]

I think you are ignoring the fact that deletion decisions are made by consensus, and that WP:POSTPONE effectively frustrates that consensus. I know that there is no timetable per se implicit in any guideline, but AfD is a process in which decisions are made at a certain point in time by arrving at a local consensus. In nominal terms, WP:POSTPONE splits the decsion making making process into two phases: pre- and post postponement. In reality, WP:POSTPONE effectively takes away the decision to delete away from a consenus of editors and gives the power to delay the decsion to a subset of the discussion's participants.--Gavin Collins (talk) 11:02, 4 July 2008 (UTC)[reply]

Why not userfy?

Why not WP:USERFY an article if it proposed for deletion? I think there is a presumtion underlying this discussion that editors are not bold enough to take their own iniative by moving an article to their own user space where it can be worked on at leisure. I don't think this proposal is appropriate, because there already exist many remedies to deletion that individuals can take without disrupting the AfD process.--Gavin Collins (talk) 11:02, 4 July 2008 (UTC)[reply]

This is actually a really good idea. Didn't think of that before. Randomran (talk) 15:04, 4 July 2008 (UTC)[reply]
Just as a quick follow up... I am concerned that WP:POSTPONE gets a little bureaucratic in that it relies too much on the discretion of an administrator, and the argument of a single editor. If other people share that concern, I would suggest revising WP:POSTPONE to incorporate the userfication process. The userfication process is a little less contentious and less suspect to abuse. In essence, someone putting forth a decent argument for WP:POSTPONE would be entitled to userfication of the page, giving them a reasonable amount of time to improve the article in their userspace. Randomran (talk) 18:38, 6 July 2008 (UTC)[reply]

Yes, this is the obvious and non-bureaucratic approach, indeed it is what we do already! It might be a good idea to make a template that people could add to the top of an AfD to request userfication, which would stop the admin closing the discussion missing a request and allow other interested editors to see what has happened to the deleted content. Tim Vickers (talk) 15:55, 20 May 2009 (UTC)[reply]

There's nothing wrong with userfication, save for a couple things:

Existing non-free images used in that article will become orphaned presuming they are only used in that article, and if the userification takes more than a week to complete, you will have people having to reupload media. Having a POSTPONE/rescue concept requires no change in images.
Userification without a followup check can lead to deleted content remaining in userspace for a long time. For example, there is a case of a non-notable documentary video called "The Money Masters" that has been created and deleted in mainspace many times due to lack of sources. Surprising, it has been userified at least four times [1]. Now, I appreciate there's a very gray line between preparing a page in one's sandbox, taking one's time to get it ready, and just holding deleted content indefinitely, but I would say that if an article hits AFD and that most of the !votes are towards delete but someone offers to fix the article, then we give them that time but check - whether its userification or through POSTPONE, and if no improvements are made, we still delete the article. Now, the only benefit about POSTPONE here is that the article stays in mainspace, potentially reaching more eyes to view and help it if it is marked appropriately.

Is POSTPONE a replacement for userification? I don't think so: I think userification is good for cases where the user may think it may take more work than either the 7 day AFD or 4 week POSTPONE period to improve an article, requiring deep specialized knowledge or access to references to do that. POSTPONE could be used for cases where sources seem to be apparent and there's no need to initiate any type of admin action until more time is given. --MASEM (t) 16:20, 20 May 2009 (UTC)[reply]

I'm not persuaded that the added bureaucracy and potential for misuse in this proposal outweigh the slight benefit of preserving non-free images from deleted articles (something of arguable benefit anyway). If the only good reason for this proposal is to reduce the amount of material in userspace, then our time would be better spent working on a bot proposal that would tag as speedies userfied content that hadn't been edited for a while. (This could be possible, but would need the closing admin to add a template to the userfied content) Tim Vickers (talk) 17:16, 20 May 2009 (UTC)[reply]

WT:Articles for deletion#Proposal #2 has a component relating to a maintenance template or category for userfied articles. Flatscan (talk) 04:58, 21 May 2009 (UTC)[reply]