Wikipedia talk:Articles for deletion/Relisting straw poll: Difference between revisions

Content deleted Content added

Inline

Revision as of 04:46, 29 August 2010

Due to several users' complaints about my AfD closes based on WP:NPASR, I opened a discussion at Wikipedia talk:Articles for deletion#AfDs with little or no discussion. The basic premises are: How many relists should be allowed for each discussion? When an AfD with insufficient participation reaches that limit, should the result be no consensus (WP:NPASR) or delete? We eventually found that it would be better to get the general public involved. It is hoped that this straw poll will allow us to see how people view this issue. (Note: A related discussion from a year ago can be found at Wikipedia talk:Articles for deletion/Archive 53#Proposal, treat AFDs with little or no discussion as "uncontested prods".)

Straw poll instructions

For each of the three questions, leave a !vote and a rationale under the subsection that best describes your views.

Relisting limit

How many times can an AfD be relisted? Currently WP:RELIST specifies a maximum of two.

None

The AfD will be closed after the one week time-span has elapsed.

7 days is more than plenty for compelling arugements for/against deletion. If no consensus is available after one week, close and default to keep. Lugnuts (talk) 14:19, 28 August 2010 (UTC)[reply]
absolutely not It is very difficult to get word out to knowledgeable parties. Relistings serve a crucial role. Kingturtle (talk) 04:33, 29 August 2010 (UTC)[reply]

One

The AfD will be relisted once if necessary, and closed after a maximum of two weeks.

Three weeks is way too long. Don't leave articles in limbo like that. If resolution isn't reached in 2 weeks, it should either be deleted (ala a prod) or closed as no-consensus. Admins should have more hutspa to what is necessary---don't keep relisting, make a decision and stand behind it.---Balloonman ^{NO! I'm Spartacus!} 13:45, 28 August 2010 (UTC)[reply]
I would support this for most AFDs with a second relist for AFDs on BLPs or if the closer thinks there is a good reason for a second relist. (such as the AFD tag being removed from the article) This is basically what I was doing until recently. --Ron Ritzman (talk) 13:58, 28 August 2010 (UTC)[reply]

Two

The AfD will be relisted up to two times if necessary, and closed after a maximum of three weeks.

If we haven't got a decision after three weeks, we're very unlikely to get one by stretching it out to four, five, six, or more weeks. However, I'd slightly loosen the existing rules to permit admins to use their best judgment if there are extraordinary circumstances ("normally no more than twice", rather than "never more than twice"). WhatamIdoing (talk) 03:57, 28 August 2010 (UTC)[reply]
This is essentially the current practice and I think it is a good one. Nsk92 (talk) 05:09, 28 August 2010 (UTC)[reply]
I find even relisting twice quite a stretch, but an AfD certainly should never be relisted any further than that. --Cybercobra (talk) 09:16, 28 August 2010 (UTC)[reply]
As per currently; one relist should be enough to resolve matters with two an absolute maximum in unusual cases. It's not reasonable to have a deletion tag on for weeks on end. Stifle (talk) 13:30, 28 August 2010 (UTC)[reply]
Also, per Iridescent (no limit, below), if an AfD gets no participation other than the nominator's after two relists, it should be closed as something along the lines of "Nobody cares," defaulting to delete. If even the article creators don't take part in the AfD after being properly notified that a discussion is taking place, then it shows that even they don't care, and the article amounts to nothing more than a tweet that is quickly forgotten. However, I would say that the admin who relists an article for the second time should (but should not be forced to) express his own opinion. -- Blanchardb -^{Me•MyEars•MyMouth}- timed 15:29, 28 August 2010 (UTC)[reply]
Two relists seems sufficient to get anyone involved who wants to become involved. After that, people just start tuning the listing out. Carrite (talk) 18:21, 28 August 2010 (UTC)[reply]
-- Cirt (talk) 00:29, 29 August 2010 (UTC)[reply]
Two relists should be the absolute maximum. I would disagree with Blanchardb, above, as to whether a lack of participation should default to delete. If someone nominates an article and it gets no response from any other AfD participant, that implies that the nominator didn't write a nomination compelling enough to even warrant a "per nom". In such a case, I think it should be a "no consensus keep", allowing a renomination at any time afterward. But that's a separate matter than the number of relists, which, as I said, should be limited to two. --Metropolitan90 (talk) 01:19, 29 August 2010 (UTC)[reply]
This is the optimal situation. And I agree with Metro regarding the default being a no consensus keep. Kingturtle (talk) 04:36, 29 August 2010 (UTC)[reply]

Three

The AfD will be relisted up to three times if necessary, and closed after a maximum of four weeks.

More than three, but finite

If you feel more relists will be helpful, but don't want them to continue indefinitely, !vote here and specify the precise number of allowable relists.

No limit

The AfD will be relisted indefinitely until sufficient discussion occurs.

"Nobody cares" isn't the same as "no consensus". If there's not a single person who cares about an article enough to keep it, if anything it ought to default to "delete". – iridescent 20:20, 27 August 2010 (UTC)[reply]
I agree with iridescent. I've currently been closing twice-relisted AfDs with one or no !votes as no consensus, but I agree with the opinion that, if someone decided to nominate an article for deletion, and it was not procedural, then it should default to delete like an uncontested prod. If someone has voiced opposition to deletion, then the discussion should continue until a discussion has developed into either keep, delete, or a "real" no consensus, with actual participation. —fetch ·comms 21:15, 27 August 2010 (UTC)[reply]
Numerical limits suck, and this is hardly a common enough problem that we need a rule on it anyway: more than one relist is uncommon, and more than 2 is very rare. Admins are expected to close AFDs using reason and common sense, just like everything else they do. Andrew Lenahan - Starblind 22:34, 27 August 2010 (UTC)[reply]
Articles can change for better or for worse so it could take many attempts to get a result.Sumsum2010 · Talk · Contributions 01:17, 28 August 2010 (UTC)[reply]
Why not? Both articles, and debates, can take time. TheGrappler (talk) 02:02, 28 August 2010 (UTC)[reply]
Bad idea. Two relists and three weeks is more than enough time for a debate to occur if anyone cares about the article. It is really bad practice to keep an AfD open indefinitely or for too long. It would add to backlog and also could lead to abuse when endless relisting is used as an obfuscating tactic. AfD decisions are not necessarily perfect but the process does need to move along and not be allowed to stagnate. Nsk92 (talk) 05:08, 28 August 2010 (UTC
I can't see it being used as a "tactic" because as far as I can see the problem is AFDs that languish in obscurity. The author isn't in control of other editors, and isn't stopping them from commenting. I would fully support doing something that helps resolve the obscurity problem. How about e.g. adding a special cleanup tag to prevent stagnation: something like "This AFD has been relisted multiple times and the opinions of other editors are urgently needed" with addition of e.g. Category:Wikipedia AFD debates requiring urgent attention, prominently linked from the main AFD page? The best way to stop multiple relistings is to find a way to draw editors to them not to reduce AFD to a glorified prod, I reckon. TheGrappler (talk) 14:00, 28 August 2010 (UTC)[reply]
I don't think that we should place a hard limit on the number of relists for a page; Instead i would argue that we would leave this to the users/admins who handle the AFD's and let them decide if it is sensible to relist it another time. Hence, a relist should only be done in cases where it is sensible to believe that this action will result in more, or more valuable input. Excirial^{(Contact me,Contribs)} 18:23, 28 August 2010 (UTC)[reply]
As per Excirial. I normally relist AfDs until sufficient discussion has taken place to determine either consensus or no consensus, or until it appears unlikely that another relisting will lead to more useful discussion. This is almost never more than twice, but I see no point in imposing an arbitrary limit. Sandstein 21:15, 28 August 2010 (UTC)[reply]
Largely agreed with Iridescent. Common sense is needed. If no one has cared enough to comment at all, and the nominator is arguing for deletion, silence has become consent and the article may be deleted. If there are arguments on both sides, leading to no clear consensus, but there's a reasonable chance that consensus could develop from additional discussion, the discussion should be relisted. If the discussion has obviously bogged down past the point of much hope of consensus, it should be closed as no consensus, to perhaps be revisited at a later date. Where that point is depends on the individual discussion, so like everything, we shouldn't set bureaucratic formalities, but should instead trust closing admins to exercise good sense. If you believe a closing admin made a bad call in a specific case, you can discuss it with them, and if you still disagree after that, take it to DRV. Seraphimblade ^{Talk to me} 00:22, 29 August 2010 (UTC)[reply]

Treatment of AfDs with no discussion

If there is no limit to relisting, then this section is moot. Otherwise, AfDs must be closed at some point: either as "no consensus" (no prejudice against speedy renomination) or "delete." Within "delete," there are two options: "soft deletion" and "hard deletion." "Soft deletion" is like WP:PROD; recreations will not be deleted per WP:CSD#G4, and undeletion will be granted upon request. "Hard deletion" is like other AfD debates.

No consensus

If an AfD has reached its relisting limit and still has no !votes other than the nominator, close as "no consensus," but anyone is free to renominate the article.

I agree with this. An AfD should not result in a deletion based on only one person calling for deletion. --Metropolitan90 (talk) 01:21, 29 August 2010 (UTC)[reply]

Admin discretion

The admin is free to use his/her discretion. The admin is free to treat the AfD as a Prod and can choose to A) Keep the article, B) delete the article, C) relist the article. Dictating the results based upon !votes is contrary to what Wikipedia's consensusology is all about.---Balloonman ^{NO! I'm Spartacus!} 13:51, 28 August 2010 (UTC)[reply]

Soft deletion

If an AfD has reached its relisting limit and still has no !votes other than the nominator, close as "delete," but anyone is free to recreate the article (i.e. treat it as an uncontested PROD).

Assuming "no limit" isn't accepted, per my comments above. – iridescent 20:21, 27 August 2010 (UTC)[reply]
Per my comment above as well. If there ends up being a limit, treat it like an uncontested prod. —fetch ·comms 21:16, 27 August 2010 (UTC)[reply]
Article could be recreated and be good.Sumsum2010 · Talk · Contributions 01:17, 28 August 2010 (UTC)[reply]
Assuming (a) that a deadline is in play, and (b) the closing admin believes that the provided deletion rationale is at valid a valid argument. I'm not sure what to do if the rationale is obviously poor, but the article is poor (yet unspeedyable) too, which is one reason for preferring no time limit. TheGrappler (talk) 02:07, 28 August 2010 (UTC)[reply]
WhatamIdoing (talk) 03:48, 28 August 2010 (UTC)[reply]
Sounds good to me. Nsk92 (talk) 04:54, 28 August 2010 (UTC)[reply]
Seems logical. This case seems fairly rare anyway. --Cybercobra (talk) 09:18, 28 August 2010 (UTC)[reply]
Yes, this seems perfectly sensible. It's conceptually a bit weird that someone who sends it to AFD should explicitly need agreement to delete, whilst had they just left a prod tag in place, it'd have gone easily. We shouldn't have significantly different outcomes depending on the process used. Shimgray | talk | 13:07, 28 August 2010 (UTC)[reply]
Seems like a good result. Stifle (talk) 13:31, 28 August 2010 (UTC)[reply]
Support a "soft delete" with the closing admin using his best judgment WRT the nomination rationale and/or the state of the article. As well as being recreatable, articles deleted under this provision should also be refundable unless there are WP:BLP or serious WP:V issues. In such cases the article may be userfied or incubated. --Ron Ritzman (talk) 13:47, 28 August 2010 (UTC)[reply]
An article that is forgotten even by its creator should default to delete with no prejudice against recreation. If a closing admin objects to such a deletion, he should refrain from closing the AfD and cast a !vote instead. -- Blanchardb -^{Me•MyEars•MyMouth}- timed 15:36, 28 August 2010 (UTC)[reply]
This seems a reasonable outcome. Carrite (talk) 18:22, 28 August 2010 (UTC)[reply]
If the closing admin feels that this outcome would be wrong, (s)he could always make an argument, which would prevent the discussion from being uncontested anyway (just as an admin who is evaluating an expired prod and believes it to be wrong can contest the prod). Otherwise, at some point, silence becomes consent. Seraphimblade ^{Talk to me} 00:24, 29 August 2010 (UTC)[reply]
-- Cirt (talk) 00:28, 29 August 2010 (UTC)[reply]

Hard deletion

If an AfD has reached its relisting limit and still has no !votes other than the nominator, close as "delete," and recreations will be deleted under G4.

Strength of argument is determinative

AfDs are not a vote and administrators are required to weigh the strength of arguments when closing AfDs (WP:DGFA). Therefore, in the unlikely event that an AfD has attracted no comments and must still be closed, the article should be deleted if the argument for deletion is compelling as a matter of policy (e.g., copyright violation or unverifiable content). If the argument is not compelling, is a matter of guidelines or requires community judgment calls (e.g., notability issues), the AfD should be closed as no consensus by default.

As proposer, Sandstein 21:29, 28 August 2010 (UTC)[reply]
I agree, although I think that's inferred in the above section. Any admin who would close an AfD without votes as delete, but nominated because of something like "this is a stupid page", is likely a stupid admin. —fetch ·comms 03:09, 29 August 2010 (UTC)[reply]

Treatment of AfDs with little discussion

Same as above, but in this case, one user other than the nominator has !voted "delete," and no one has !voted "keep."

No consensus

If an AfD has reached its relisting limit and has exactly one "delete" !vote other than the nominator (and no "keep" !votes), close as "no consensus," but anyone is free to renominate the article.

In (most?) circumstances. This is assuming (a) there is a relisting limit; (b) the comments are from people who can't quite bring themselves to agree to the the nominator, but the article clearly isn't a stand-out keep either. In debates with few participants, rather than count those contributors as "neutral" and thus remove them from the reckoning, we should really be concerned with the strength of the arguments and how many people they've convinced. If there are people left undecided, it suggests that neither set of arguments was convincing - in a sense, it's "no consensus inside the heads of the participants" rather than "no consensus within the hive mind". I'd particularly prefer it deemed no consensus if the other commentators are pointing out flaws in the rationale of the nominator. This seems the most likely scenario in which people comment but don't !vote: a lot of stuff at AFD is of poor quality, hard to put a defence together for without substantially reworking the article, and often it's tricky even to determine how such reworking should be done, but sometimes the rationales are weak too. In such circumstances it's common for a commentator to point out this weakness but neither be prepared to defend the article on its own (limited) merits, nor put forward an alternative rationale to delete. If the grounds for deletion are contested in this way, I feel it's best to treat it as no consensus. TheGrappler (talk) 02:16, 28 August 2010 (UTC)[reply]

Wait, above, you supported a soft delete for no !votes. I don't quite understand why you are supporting a no consensus if there is actually a delete vote and no opposition. Seems like it would make more sense the other way around. Although I do understand, it depends on the arguments presented and such. —fetch ·comms 03:56, 28 August 2010 (UTC)[reply]

Ah. you're correct - I misread this as only the nominator being behind the one !delete vote and the other comment being just a non-!vote. My mistake. TheGrappler (talk) 04:47, 28 August 2010 (UTC)[reply]

Soft deletion

If an AfD has reached its relisting limit and has exactly one "delete" !vote other than the nominator (and no "keep" !votes), close as "delete," but anyone is free to recreate the article (i.e. treat it as an uncontested PROD).

Again, per my comments above. Someone recreating the article is effectively a delayed "keep" vote, and thus one delete + one keep = no consensus. – iridescent 20:22, 27 August 2010 (UTC)[reply]
Per above, again. Treat as uncontested prod, but the new article can be re-prodded or speedied as well as sent back to AfD, as if it were about a completely new topic. —fetch ·comms 21:18, 27 August 2010 (UTC)[reply]
Same as above.Sumsum2010 · Talk · Contributions 01:17, 28 August 2010 (UTC)[reply]
100% agreement of the three editors who cared enough to voice an opinion (the nom, the !voter, and the closing admin) is consensus in my books. WhatamIdoing (talk) 03:50, 28 August 2010 (UTC)[reply]
If it is just one !delete voter in addition to the nominator, and nobody has left a "comment" which criticizes the rationale (even though it doesn't end in a "keep"), then this provides more evidence of consensus than an unopposed prod does. TheGrappler (talk) 04:49, 28 August 2010 (UTC)[reply]
I think more flexibility needs to be given to the closing admin in such cases. Sometimes, it may be OK to close an AfD as a hard delete if there was a single delete !vote (and no relistings) if that delete !vote is solid and well-argued. In other cases closing as a soft delete close may be more appropriate. Nsk92 (talk) 05:18, 28 August 2010 (UTC)[reply]
Agree on flexibility (it depends on reasons given) but clearly some form of deletion is called for in order that AFD and PROD get vaguely comparable results, and a presumption in favour of soft deletion is sensible. Shimgray | talk | 13:09, 28 August 2010 (UTC)[reply]
Ditto per my comment above. Should also be allowed for more then one delete !vote if the closer thinks they are weak. --Ron Ritzman (talk) 14:08, 28 August 2010 (UTC)[reply]
An article that is forgotten even by its creator should default to delete with no prejudice against recreation. If a closing admin objects to such a deletion, he should refrain from closing the AfD and cast a !vote instead. However, in this case, if the arguments for deletion are compelling enough, the AfD should be closed as a hard delete. -- Blanchardb -^{Me•MyEars•MyMouth}- timed 15:36, 28 August 2010 (UTC)[reply]
Two deletes and no keeps should be rendered as a soft delete, unless it's an obvious hoax or copyright violation or BLP violation, etc., situation. Carrite (talk) 18:24, 28 August 2010 (UTC)[reply]
-- Cirt (talk) 00:29, 29 August 2010 (UTC)[reply]

Hard deletion

If an AfD has reached its relisting limit and has exactly one "delete" !vote other than the nominator (and no "keep" !votes), close as "delete," and recreations will be deleted under G4.

Case by case treatment

I think AfDs with only one delete !vote and no other opinions (except for the nominator) really should be considered on a case- by-case basis. There are situations where a sole delete !vote offers quite a solid rationale and I think a hard delete is justified in such cases as an AfD outcome. I don't really see it as necessary to institute a rigid rule for handling such situations. Nsk92 (talk) 05:00, 28 August 2010 (UTC)[reply]
Per WhatamIdoing. --Cybercobra (talk) 09:24, 28 August 2010 (UTC)[reply]
I don't think this arises enough to make a hard rule about it. Stifle (talk) 13:32, 28 August 2010 (UTC)[reply]
An AFD is an AFD, just because it only has one !vote doesn't mean others hadnt looked at it. If it's gone through AFD, it shouldn't be treted as a prod for recreation purposes.---Balloonman ^{NO! I'm Spartacus!} 13:48, 28 August 2010 (UTC)[reply]
As above: AfDs are not a vote and administrators are required to weigh the strength of arguments when closing AfDs (WP:DGFA). Therefore, if an AfD has attracted no or few comments and must still be closed, the article should be deleted if the argument for deletion is compelling as a matter of policy (e.g., copyright violation or unverifiable content). If the argument is not compelling, is a matter of guidelines or requires community judgment calls (e.g., notability issues), the AfD should be closed as no consensus by default. Sandstein 21:30, 28 August 2010 (UTC)[reply]
It should be handled case by case, but AfD, being a public forum, isn't prod. Generally speaking, if the deletion rationale is reasonably sound, the article's been on AfD for at least seven days, at least one person has agreed with the nominator, and absolutely no one has bothered to object, we're at the point of hard deletion. Seraphimblade ^{Talk to me} 00:56, 29 August 2010 (UTC)[reply]
The closer should decide based on the soundness of the arguments presented. --Metropolitan90 (talk) 01:24, 29 August 2010 (UTC)[reply]

General discussion

Use this space for meta-discussion and suggestions not listed above. -- King of ♥ ♦ ♣ ♠ 20:03, 27 August 2010 (UTC)[reply]

None of the rationales stated here seem to assume the admin has no discretion or isn't going to read the AFD nomination and check it on merit. I'm not sure if the distinct "admin discretion" or "case by case" options are adding much value, or whether the format of the poll needs to be changed. I think everyone assumes that admin discretion will apply, and most people who post in other sections are merely giving opinion about the "default" position. TheGrappler (talk) 14:04, 28 August 2010 (UTC)[reply]

Non-administrative closures should be eliminated. Carrite (talk) 18:25, 28 August 2010 (UTC)[reply]

The unspoken problem here is lack of participation in AfD. I think we need to address that, as well. The ideal scenario is that we rarely or never have any need to deal with AfDs with little or no comment. Seraphimblade ^{Talk to me} 01:00, 29 August 2010 (UTC)[reply]
I would have thought it would be inferred that one should always use common sense, and rarely some IAR, on the issue of soft deleting uncontested AfDs that have horrible reasoning behind the delete vote/nomination. If there's a nom that is obviously not policy-based and no one comments, I'm fairly sure an admin would have the sense to not delete the article. The "case-by-case" thing will inevitably be built into any proposal like this, because not all AfDs are created equal and common sense should always be used. —fetch ·comms 03:14, 29 August 2010 (UTC)[reply]
Missing from this debate is the amount of time between an article's AfDs. I sometimes see an article survive an AfD, and then have to endure another AfD a few weeks or months later. I think if an article survives an AfD, it should be allowed a full year to develop before going through the ringer again. Kingturtle (talk) 04:29, 29 August 2010 (UTC)[reply]

@@ Line 10: / Line 10: @@
 The AfD will be closed after the one week time-span has elapsed.
 #7 days is more than plenty for compelling arugements for/against deletion. If no consensus is available after one week, close and default to keep. '''[[User:Lugnuts|<font color="002bb8">Lugnuts</font>]]''' ([[User talk:Lugnuts|talk]]) 14:19, 28 August 2010 (UTC)
-#'''absolutely not''' It is very difficult to get word out to knowledgeable parties. Relistings serve a crucial role. [[User:Kingturtle|Kingturtle]] ([[User talk:Kingturtle|talk]]) 04:33, 29 August 2010 (UTC)
+#:'''absolutely not''' It is very difficult to get word out to knowledgeable parties. Relistings serve a crucial role. [[User:Kingturtle|Kingturtle]] ([[User talk:Kingturtle|talk]]) 04:33, 29 August 2010 (UTC)
 === One ===