User:Useight/RFA Subjects/Vote strength

Voting adjectives - question about/for b'crats (Archive 55)

Do b'crats take into account the adjectives people use to support/oppose? Obviously I don't mean the joke ones ("support on wheels" and so on) but "weak support", "weak oppose", "weakest oppose possible"? Or even "strong support" and "strong oppose"? It would probably be silly to be give people double their say just by letting them add "strong" to their vote, but in RfAs where a lot of people cast "very weak oppose" votes, do those get taken into account, or do they count as full votes anyway? --W.marsh 13:40, 9 May 2006 (UTC)

I would consider more the strength of the additional comments and reasoning than just the adjectives used, but if there are lots of strong supports and most of the opposes say weak oppose, that would be one thing that would tend to tip the nom towards promotion. Bare oppose votes without reasoning are harder to take into full account. Solid reasoning is much more helpful. - Taxman ^Talk 16:24, 9 May 2006 (UTC)

If it comes within our discretionary range, I would weigh in each oppose vote and its corresponding reason carefully. I believe that adminship is about judgement and oppose votes which are downright silly will not be given due weightage. For such opposes and those votes that say weak oppose I won't negate 4 support votes, I'll negate less. I've clearly outlined this during my RFB. =Nichalp «Talk»= 16:44, 9 May 2006 (UTC)

Crap, I should have opposed that, you're using numeric standards! *sigh* :-( (not your fault, but running anything on a wiki in that manner is known to Not Fly in the long term :-( ). Maybe the discussion below can bring further insights. Kim Bruning 02:34, 10 May 2006 (UTC)

See [1], Linuxbeak's breakdown of his promotion of Tawker; these were a factor. ~ PseudoSudo 00:43, 11 May 2006 (UTC)

Requiring diffs for !votes (Archive 68)

While we're on the train of thought of "now for something completely different" for RfA, I thought I'd share something that came to mind recently: require that every !vote provide (at least) three diffs supporting their case for supporting/opposing/etc. In my mind, this would accomplish a few things, including:

Force people to actually look through the user's contributions (odds are you'll have to click on at least 10 diffs to get 3 good/bad ones)
Prevent a user from being condemned for one bad diff (mistakes can be forgiven; a pattern of bad edits should be investigated)
Make arguments better supported and help keep it from becoming a vote (would help stop simple "Support" or "Oppose" votes with absolutely no other comment, and would stop opposes to which someone responds, "Can you provide an example of this?")

And probably a few other things generally relating to informed discussion. Not sure how practical this suggestion is, but I figured I'd at least throw it out there for discussion. It's also completely possible that I simply need some sleep, which I will try to do right after clicking "Save page" here. I look forward to everyone discussing and completely demolishing this idea. Cheers, EWS23 (Leave me a message!) 04:36, 3 September 2006 (UTC)

After some time, everyone would list the same thing. "Good user, good edit number, good talk page discussions." How could you enforce this, and how could you make it so everything couldn't be summed up together. On some RfAs, you may have one solid reason for opposing, in which case you may have difficulty coming up with several. Repeated bad behavior in one field is one such example. Michael 06:06, 3 September 2006 (UTC)

It's an interesting idea and I certainly understand where you're coming from, but I don't think it's in the Wikipedia spirit to force someone to participate a certain way. Discussing RfA's are purely voluntary and one should be able to participate however one wants, be it a singular word (i.e. support, oppose, neutral) or a three hundred word summation the candidate. hoopydink^{Conas tá tú?} 06:34, 3 September 2006 (UTC)

Otherwise, imagine making someone give three reasons for opposing - not one, but three. That would be awkward for both parties. Michael 06:39, 3 September 2006 (UTC)

For supporting, providing a specific diff seems unnecessary. The qualifications for adminship concern a body of work. e.g., someone who's a long-time recent changes patroller has done a lot of good work that bears well on an adminship. A diff (or three) of vandalism reversion, however, is meaningless. Christopher Parham (talk) 17:45, 3 September 2006 (UTC)

Qualified Adminship Voting (Archive 73)

I've tried to come up with an RFA reform idea that does not:

Drown us in process
Make genuine opposition impossible
Make RFA intolerably long

The result was the rather badly named 'qualified adminship voting'. Basically, RFA would be a vote with a 70% threshold, but all the less meritous stuff (e.g. editcountitis, editsummaryitis, namespaceitis, timeitis) and accusations of misconduct without a supporting diff would be automatically discounted. That solves (what I think is) the main objection to having RFA as a vote - that a Support vote on the basis that the candidate had written 3 featured articles would be given 1/4 of the weight that a 'Only 99.5% edit summary usage' oppose votes gets. I'm not sure if anyone else thinks this is a good idea, so I haven't put it in WPspace yet - it's available at my sandbox and I'll move it to WPspace if people think it's worth discussing. Cynical 11:09, 19 October 2006 (UTC)

The fundamental problem with this approach is that it evaluates some people's metrics for adminship as being improper while ignoring others. Slippery slope; where do we stop adding in new, additional metrics that won't count as votes if people use them? Further, for every category of vote that you state as being improperly based, people will come up with another way of saying the same thing that circumvents the improperly based vote ban. For example, if you won't let John Doe vote on on the basis of editcountitis, namespaceitis, timeitis, editsummaryitis...John Doe could just as well say "Oppose: Not enough experience yet." You can't insist a diff be provided to support that; it's too inspecific. If you block the vote based on being too inspecific, you've just added another reason to the list of reasons to ban votes. Where do you stop? --Durin 12:52, 19 October 2006 (UTC)
- Not enough experience would come under timeitis, wouldn't it? I think I specifically mention 'length of time...' in the text anyway. I don't think there is much risk of a 'slippery slope' here as all of the 'excluded' vote types under this would be numeric-type votes. In other words, anything which is along the lines of 'this user only has X number of [measurement A] and my criteria say X+1' would be excluded, but anything else would be fine even if some or many people might question its relevance (e.g. lack of participation in a particular Patrol).Cynical 14:37, 19 October 2006 (UTC)
  - - What it sounds like you're saying is that inexperience is not a reason to oppose someone for adminship? Am I correct in interpreting you? --Durin 15:10, 19 October 2006 (UTC)
  - (edit conflict) There may be a worse problem with supressing votes of types you don't like. Clearly 'only has 9000 edits, too new' should be disregarded, but 'only has 50 edits, too new' is usually a reason to prevent a user becoming an admin (I remember two recent cases where a user with less than 500 edits has had over 50% support for a reasonable length of time; one was an admin on another project (I can't remember who it was) and the other was TawkerbotTorA, who had a good excuse). For very new users who haven't done anything wrong, it's hard to find grounds other than inexperience to oppose, but many people would consider the inexperience to be valid opposition grounds. --ais523 14:44, 19 October 2006 (U T C)

- - - Just FYI, you're probably thinking of either Walter or Amgine, though Walter was the one with less than 500 edits here on the English Wikipedia at the time of his RfA. Thanks! Flcelloguy (A note?) 03:07, 21 October 2006 (UTC)

A problem with that is that people with editcountitic tendencies might still vote in the way that they do, but not tell us (just go "oppose per <someone else>" instead of "oppose - lack of template edits"). You cannot objectively determine who has or has not a valid reason to support. >Radiant< 14:53, 19 October 2006 (UTC)
Forgot to mention, the proposal does have basic editcount and length of time requirements built in - it would require candidates to have 1000 edits and 3 months experience prior to applying. This avoids the stupid requirements we have now, but without the sort of problems suggested by Durin. Note: I've moved this to namespace to give it a better opportunity for discussion. You can find the actual text of the proposal at Wikipedia:Qualified adminship voting or WP:QAV. Cynical 18:35, 19 October 2006 (UTC)
- What is stupid about the requirements we have now? Durin's "...in review" pages suggested that edit count 'requirements' had not substantially risen; even if they have gone up slightly over time this is likely merely a function of the proliferation of tools that reduce the "effot per edit" quotient. Since it is effectively total "effort" people are seeking to evaluate in some way by editcountiting, this is an entirely benign phenomenon.
- But you do fall into Durin's trap: you consider them to be "stupid". This does not make them so, and nor does carving them on forbidden stone.
- This proposal, most importantly, is in the "we all hate the opposers" category. It ups the ante against those who would oppose an RfA while allowing supporters to continue with "zOMGBBQ he isn't one since bananas were square?" type !votes which, frankly, should receive significantly less weight than 0.25 of an oppose that says "has too few edits to demonstrate depth of knowledge". -Splash - tk 19:51, 19 October 2006 (UTC)

I would reiterate my perenial rant: some votes may indeed be "stupid", but but restricting how people vote is a bad idea and will create many more problems than it will solve. Besides, the threshold for promotion which is now 75%, allows an ample margin of error for "stupid" votes (both oppose and support). Oleg Alexandrov (talk) 03:09, 20 October 2006 (UTC)

- I tend towards opposition in RFAs so I don't hate opposers as such. I just think that (given the shortage of admins) some of the requirements (did I hear 4000 edits the other day? cant remember whose RFA it was) are getting over the top, and we ought to get rid of them so that legitimate concerns over a person's suitability for adminship (e.g. civility, newbie biting, lack of contribution to community projects) can be given more prominence than they currently get. Perhaps something like this would get more support if we had a two-pronged approach: bring in the restrictions of the sort that are in QAV, but also make adminship easier to lose (one thing that I thought about yesterday was the possibility of putting all new admins on administrative 1RR for 3 months) so that the odd bad candidate promoted in this way would be swiftly deadminned Cynical 08:26, 20 October 2006 (UTC)

Opposing per a low edit count, like 300, is not like calling the sky pink. You don't discount votes that at least make an effort to make sense or don't seem obviously contradricted (ie "oppose, 9000 edits too low") by the current situation. This proposal is incredibly subjective, and not feasible by any means. Aside from that, people can always just go "oppose per X" when they really mean "9000 edits too low". Votes are not really that bad either, its idiots and irrational people that pollute them. No governments let people vote on every issue, thats why we have parliaments/houses and committees usually. Letting any random user, new or not, experienced or not, vote on RfA is half the problem. The other half is that people don't agree on what an admin is, and sometimes people get opposed for things they would not expect or that do not relate to adminship. Both of these are tough cookies to fix. Mote Bureuacrat discretion could help the former, and a referendum discussion on what and "admin" is and is required of one should be had.Voice-of-All 15:19, 20 October 2006 (UTC)

What we're all forgetting about is chimpanzees and dartboards. Put the two together and we have an easy solution to all of our RfA worries. Marskell 11:52, 20 October 2006 (UTC)