Wikipedia talk:Revotes on Vfd

When I wrote this I was thinking of a very mechanical, precisely defined, rules oriented way for revotes to happen. It is also possible for revotes to happen in a loose, guideline-oriented, "lets just do what makes sense" way. That is the preferred method.

This is when that method doesn't work.

Maybe this isn't necessary. Maybe the existing guidelines can be rewritten or simply enforced to ensure intelligent, sensible revoting when appropriate. I don't know. But this is here anyway as a think tank proposal.

I do think this is useful. Vfd pages can get cluttered up, especially when the article is rewritten over the course of the Vfd. And people tend to stick with their original impressions and their original opinions, thus evaluating the rewritten article in terms of it's old state in stead of it's new state. A fresh page gives the rewritten article the fresh start it deserves.

- Pioneer-12 20:13, 24 Apr 2005 (UTC)

This is a bad idea for several reasons. For one, the VfD notice explicitly states that the nominated article is not to be moved while the discussion is ongoing. Secondly, most VfD voters will return to change their votes if the article is changed substantially and if they feel the changes warrant a change in vote. The revote process would just make for m:instruction creep and overly complicate an already complicated process. android↔talk 15:17, Apr 24, 2005 (UTC)

Oppose. The VfD process is (properly) already weighted in favor of keeping articles whenever possible. The combination of requiring rough consensus, making the guideline that 1/3 keeps = no consensus, and allowing the acting sysop to exercise judgement makes it reasonable hard to delete articles unless they're really stinkers, and reasonably easy to keep them.

The proper thing to do when an article is improved is simply to make a conspicuous comment to that effect and urge people to change their votes. The five-day VfD discussion period allows plenty of time for this, not even accounting for the current backlog, and experience shows that many, many discussions conclude with a string of "keep as revised" or "keep improved version" and a judgement that there is no consensus to delete. Most VfD voters keep the discussion page on their watchlists, and if the page has been truly improved a) it's not at all hard to get a handful of changed votes, b) a handful is usually enough to get a total of 1/3 keeps, and c) acting sysops are sympathetic to improved articles.

We do not need to add a mechanism for parliamentary maneuvering. One "delete" vote on a VfD should stick; it should not be necessary to keep voting "delete" over and over again in successive VfD re-votes.

A particularly poisonous aspect is that there is unlikely to be any universal agreement on how much improvement is needed for a re-vote and this will introduce fresh arenas for controversy and fresh opportunity to game the system. Dpbsmith (talk) 22:56, 24 Apr 2005 (UTC)

Coping with loss: How to handle the VfD of a treasured article

If you believe that a substantial rewrite of an article has taken place without due notice of the community, feel free to politely contact individuals who have already contributed to the VfD discussion. They might change their votes. Don't do this very often; editors will get annoyed.

If an article is rewritten very close to the VfD deadline, it may be deleted anyway. Feel free to request its undeletion. That's what Votes for Undeletion is for, with its associated Undeletion policy. Please use sparingly.

We don't need the policy creep or gamesmanship that would come from this proposal. It's meant well, but would ultimately be harmful. --TenOfAllTrades (talk/contrib) 23:24, 24 Apr 2005 (UTC)

Thank you for your suggestions. Politely contacting individuals and asking them to reelavuate (not necessarily change) their vote in light of the revisions to a page makes sense. However, it's ludicrous that the best course of action is for a page to be deleted and then undeleted. That seems much more like procedure creep then this. A wrongly undeleted page should not have been deleted in the first place. If a page is wrongly deleted and has to be undeleted, then the Vfd system failed. - Pioneer-12 02:42, 25 Apr 2005 (UTC)

"Policy creep" has a specific meaning here with which you may not be fully familiar. It's what happens when people (usually with the best of intentions) keep writing new rules and policies to handle special and unusual cases that don't come up very often. The concern is that it may result in very complicated, Byzantine policies, and encourage rules lawyering and other un-wiki behaviour.

Votes for Undeletion arose precisely because it was acknowledged that the deletion process is not (and likely can't ever be) perfect. If you like, consider it to be the second-chance extension to VfD you were looking for. At worst, an article will be out of commission for five days while it is discussed on VfU. From what I can see of VfU, it's a pretty quiet place—which suggests to me that most deletions don't have serious problems.--TenOfAllTrades (talk/contrib) 03:17, 25 Apr 2005 (UTC)

I am familiar with the theory of policy creep aka instruction creep aka procedure creep. In fact, I wrote half of the current m:instruction creep page. To be frank: the theory of instruction creep is overrated. Policy creep refers to overcomplex, badly written, tedious to follow rules, not to rules in general. Having less rules is not inherently better or worse then more rules; it's the quality and the relevence and the ease of use of rules that matters. I think people have just had bad experiences with poorly written and poorly interpreted rules in the past and thus think it's better to have no rules. (Or only very simple, general rules.) That is an invitation to chaos. If a rule doesn't work properly, then you fix the rule! If there are exceptions to the rule, then you add the exception. Well written rules, when used with wisdom and in good faith, will work properly 99.9% of the time.

It is good that you mention rules lawyering. Rules lawyering is when someone tries to use the letter of the law to their advantage, ignoring the spirit of the law. Rules lawyering is not done with good faith, but for personal interests. Rules lawyering is only possible due to imperfectly or sloppily written rules. As the quality of rules increase, the potential for rules lawyering drops asymptotically toward zero.

I think the bias against clear rules under the banner of instruction creep is ridiculous and has severely impaired the evolution of an modern and comprehensive rule system on Wikipedia. Right now the Wikipedia policy system is stuck in the middle ages. Vfds pages are like banana republics... ignorance and pettiness are rampant, factions come and go at random, and the whole process is a mockery of consensus.

"Undeletion arose precisely because it was acknowledged that the deletion process is not (and likely can't ever be) perfect." Bah! Perfection must always be seeked, even if it cannot be obtained. The Vfd process, as it now stands, is broken. We might not be able to make it 100% foolproof perfect, but we can make it a hell of a lot better then it is now. This proposal won't completely fix it, but it is one way to improve it. - Pioneer-12 03:17, 26 Apr 2005 (UTC)

The problem is that the Vfd system is entirely based on user input, and user input can often be biased or even misused. Indeed there are some groups (like the "Deletionists" and "Inclusionists") who will add notices to their group page of things going on and the rest of the group will merrily march off and vote for their particular "wing" of belief. Not to say such groups are "evil", but their networking could be abused. So really the Vfd is only as flawed as the voters' motives are. Master Thief Garrett 04:06, 25 Apr 2005 (UTC)

We're kind of stuck with that, though. Until the Wikimedia Foundation's servers achieve consciousness, we have to use human input to make all of our decisions. For what it's worth, most self-identified inclusionists or deletionists use the label as a description of a personal bias one way or another, rather than a firm policy. The recent Schoolwatch mess is an unsavoury development, but I would say that it's uncharacteristic of most VfD participants. --TenOfAllTrades (talk/contrib) 05:07, 25 Apr 2005 (UTC)

Yes we are stuck with the current way. No, it's not firm policy with them at all; it was actually the Schoolwatch I was alluding to, but the name had slipped my mind for some reason. I think it's just so *wrong* if they think that each and every of the 50,000,000 schools worldwide deserve their own individual pages. I mean, that's more articles than we have now put together... Master Thief Garrett 06:21, 25 Apr 2005 (UTC)

I don't know ... there have been spamming problems before where Users have posted comments on several other Users' pages asking them to reconsider their votes, and the community has generally frowned on the spamming. It has even been used as evidence in an RfC. Rick K 05:10, Apr 25, 2005 (UTC)

Opinions have varied about that, I think it's fair to say. Would you agree, Rick, that there's a difference between personal messages, tailored to individuals, which ask them to reevaluate their vote, given that their specific concerns have been met, and simply messaging anyone you think will vote your way to delete or include an article?Grace Note 05:14, 25 Apr 2005 (UTC)

Yes, I would definitely agree that there is a difference there. Rick K 05:16, Apr 25, 2005 (UTC)

Re: Messaging people asking for their vote. In a democratic country, that's called Get out the vote, and it's a highly respected practice, as it encourages more people to vote on issues. Wikipedia is more consensus then democracy, but the same reasoning applies. A higher voter turnout is better--even if you are just appealing to a certain block of voters. It just doesn't make sense to go up to people who campaign in a strategic manner and tell them "Hey, stop that!" (I've seen admins campaign for votes, by the way.) A vote is a vote, and a campaign is a campaign. It's politics. Sometimes the side whose opinion "wins out" is the side who is most passionate about an issue. That is not necessarily a bad thing... passion of opinion should be a consideration during the voting process and during the consensus process.

In Wikipedia, bringing more people into the voting process means that the discussion will be participated by a larger portion of the community, each with their own views and concerns. This means that the ultimate consensus, if one is achieved, will have considered and accommodated a wider array of views and objections, and that the ultimate decision or compromise is more likely to be a wise and mutually acceptable one.

Unfortunately the vfd process is very fouled up since only a small percentage of the community chooses to vote on any one vfd. Which means, on any given vote, decisions which affect everyone are being decided by a tiny fraction of the community. Are their opinions really representative of the community at large? You can't say for sure. A page may come up for vfd one day and get almost total support. It may come up for vfd another day (or, more likely, a very similar page comes up for vfd), and, for whatever reason, the people who support the concept aren't around.... and suddenly the page gets almost no support. It's statistically improbable, but it can happen.

Granted, you aren't going to get 100% turnout in any vote, but when turnout is less then one percent, the results are unreliable at best.

Not to mention how the votes are cast and then objections are made and discussions joined in response to votes.... how utterly backward! It's this backwardness that necessitates this "revote clause", aka a "do-over clause".

The current vfd system is flawed all over the place. I'm looking into ways of improving it. This is just the first suggestion. Stay tuned.... and drop me a message if you have any suggestions of your own.

- Pioneer-12 06:16, 3 May 2005 (UTC)[reply]

from vfd

Wikipedia:Votes for deletion/Revotes on Vfd 01:19, 10 May 2005 (UTC)