User talk:Abigor/Vote proposal

From Wikimedia Commons, the free media repository
Jump to navigation Jump to search

Comments

[edit]

I'm really against sort of arbitrary limit, but if I had to choose one, it would be no greater than 10. FWIW. –Juliancolton | Talk 00:34, 9 September 2009 (UTC)[reply]

On reflection - I think my only real concern would be that we do not disenfranchise any of the smaller language communities from getting admins in their language. When was the last time there was a non en RfA.....
We lnow en wp can impose its wishes on us but Commons is multi lingual & we must take care not to alienate non en folk. --Herby talk thyme 08:36, 9 September 2009 (UTC)[reply]
Have we had a lot of opposition purely on the basis of non en speakers? I don't think we have but if so, that's bad. I'l make it a point to support non en speakers if someone else who knows them can present the info we need to make informed decisions. ++Lar: t/c 13:04, 9 September 2009 (UTC)[reply]
Notice that in order to start a RfA the contributor must know English language at least at en-2 or en-3 level, otherwise he/she will not be able to answer the comments placed by other contributors, so there is an inherent discrimination but I do not know how to solve it. Best regards, Alpertron (talk) 13:55, 9 September 2009 (UTC)[reply]
See that's the problem. That's how candidates feel too. No, you can use any language. I welcome candidates to not use English at all. People know how to use online translators. Other users can help. So can bookmarklets like
javascript:location='http://translate.google.com/translate?u='%20+%20encodeURIComponent(location);. Rocket000 (talk) 14:12, 9 September 2009 (UTC)[reply]
So the translations of Commons:Administrators will need to be changed to include the information you wrote above. I will start with the Spanish translation. Best regards, Alpertron (talk) 14:34, 9 September 2009 (UTC)[reply]
I would welcome users that are non English, we need to remember at all times that Commons isn't a English project and I am pretty sure people would be willing to translate comments or nominations in English and the other way around when needed.
The English Wikipedia is kind of a big project so the English Wikipedia can make a user sysop here by coming to vote, but I'm sure a user with good motivation from a small wiki without a big community would pass here also. Huib talk 19:04, 9 September 2009 (UTC)[reply]
@Lar, I would count a neutral vote as vote to make the ten votes, but not as a vote for the end %. Huib talk 19:04, 9 September 2009 (UTC)[reply]
Hmm... see above, and can you work an example please? I don't want 3 supports and 7 neutrals to pass, I don't think... ++Lar: t/c 16:11, 10 September 2009 (UTC)[reply]
In these cases I assume that the votes will come slower because of the communication overhead, so I think that if in the first 7 days less than 10 votes are counted, but there is still 75% of positive votes, another 7 days should be given to the RfA. Best regards, Alpertron (talk) 20:56, 9 September 2009 (UTC)[reply]
Something like that makes sense to me... if it's not obviously failing, a "relist to gain better consensus" move being more automatic seems goodness. ++Lar: t/c 16:11, 10 September 2009 (UTC)[reply]
Lar, I didn't think of it in that way, when we should do that we could just keep it the way it is because it is even easy when you have 1 support and 9 comments.
Maybe it would be better to change the text in:
Promotion normally requires at least 75% in favour, with a minimum of 10 votes including support, oppose. Votes from unregistered users are not counted. However, the closing bureaucrat has discretion in judging community consensus, and the decision will not necessarily be based on the raw numbers.
I think it would be a never ending story when we start with saying a neutral counts for 0.5 support, or 0.25 support or a oppose, it would be more easy to just not count them right? Huib talk 17:15, 10 September 2009 (UTC)[reply]
We never came to a resolution about how to count them when I asked before... not fair to count them as supports, or as opposes. Maybe don't count them at all, but "promise to consider the input from the comments attached to them carefully in cases where consensus is borderline" ?? ++Lar: t/c 20:23, 10 September 2009 (UTC)[reply]

What next... discussion seems to have petered out? Put this forward at the pump(s)? ++Lar: t/c 03:19, 15 September 2009 (UTC)[reply]

I placed it in the village pump, should it go to other language villagepumps also? Huib talk 03:48, 15 September 2009 (UTC)[reply]
yes I think so. ++Lar: t/c 04:14, 15 September 2009 (UTC)[reply]
done, I left a note on all village pumps, bars, bistros (you get the picture :) ) Huib talk 16:40, 15 September 2009 (UTC)[reply]
Please remember the disquiet caused by the debate over formalising suffarage criteria. Please ensure that there is a consensus and this gets put to a vote before implementing, I think the two weeks you've given yourself to get this through is way to short.KTo288 (talk) 14:06, 16 September 2009 (UTC)[reply]
Hi thank you for bringing this up,
When I read it like you are saying I realize that I should have used better words for this to avoid this little misunderstanding.
I mend to say that in the best case scenario we wouldn't face any objections and we could implant it in two weeks, when there are comments like now we need to take them into notice and change adjust or talk about stuff a little bit longer, I'm not in a hurry, it doesn't make a difference for me when we get it done in 2 weeks or 2 years, the only real thing that matters to me is to get in done in a good way with enough support from our Community and the other communities that need Commons for hosting there files. Huib talk 17:00, 16 September 2009 (UTC)[reply]

I don't care about the number change, but I'm against putting RfX numbers on my watchlist. If I wanted numbers on how many RfX there currently are, I'd find them myself. But, if this is forced on us, at least give us an option in preferences to turn it off if we want.--Rockfang (talk) 15:52, 16 September 2009 (UTC)[reply]

  • You know, I like this idea, but I would rather say 10 support votes and 75% instead of just 10 normal votes and 75%. Admins should have trust from the community, and 8-2 is not very much. I'm however also open for the 10 usual votes option. Furthermore, I would not say: "there are 3 admin requests", but rather list for which person there's a request, similarly to this template (the part "Adminkandidaturen" is the part that lists current nominations for adminship). Numbers give the people nothing, unlike names that they might know. So, that's my two cents on this matter. Btw, I find it very good that you took this in the sitenotice. I had already seen it in the village pump, but such an important thing ought to be in the sitenotice. --The Evil IP address (talk) 16:11, 16 September 2009 (UTC)[reply]

Is this proposal fixing a real problem?

[edit]

Has there been real problems from lack of scrutiny of admin candidates during the past year? Have the admin getting less than 10 votes been problematic admins? In a RFA, 7 or 8 supports with no opposes seems to be a problems with voter turn out rather than with the person. I think that the best idea is to advertise the RFAs more to get more input on all of them. FloNight♥♥♥ 20:54, 16 September 2009 (UTC)[reply]

  • I think it will make this place more democratic. so yes it is a problem now. tgkprog
I'm partially with FloNight. I don't think we should advertise RfAs individually, rather the system, if their is a need for it. — Kanonkas // talk // e-mail // 12:37, 17 September 2009 (UTC)[reply]
Imagine a country where elections would not be advertised to the general population. People know they can vote of course (the "system" is known), but they would find out about a particular election only if they happen to walk by a vote office, or if friends of the candidate tell them about it. Would you consider this more democratic than if billboards and TV announcements were displayed? –Tryphon☂ 12:55, 17 September 2009 (UTC)[reply]

I searched around and found the incident that triggered this proposal. I continue to think that this new requirement is not addressing the real problem. Commons needs more people to participate in its local work instead of uploading images and returning to their home wikis. I don't think that making it harder for people who are actually volunteering to be more active on Commons is a good way to address the problem. Instead, Commons needs to find ways to draw more people in and encourage them to share the work load for the stuff that keeps it running. FloNight♥♥♥ 13:49, 17 September 2009 (UTC)[reply]

I agree with FloNight again.
As to Tryphon, their is a difference. RfA isn't a popularity contest, it's a discussion. Commons and other wikis aren't a democracy nor a political system, AFAIK. Its primary purpose is to determinate consensus through discussion, not voting. If it were a vote, that would make it more like a bean counting system. While RfAs seem to give the idea/impression of it being just a straw poll, that's wrong. Polls usually discourage consensus, but nonetheless, straw polls such as in RfAs, are more of an attempt to see the community's consensus in reasoning (discussion) than votes. One part of the bureaucrat's aspects, is to find the community's consensus, with regards to how strong votes are. An oppose without any reason will most likely be counted less than an oppose with an explanation (It's hard to explain this any further without a real example). Polls or surveys sometimes impede rather than assist discussion. Polling is not a substitute for discussion, thus I don't support the view of advertising individual RfAs. Plus, I think this proposal encourages people to vote for the sake of voting, which is not a system I agree with. — Kanonkas // talk // e-mail // 16:57, 17 September 2009 (UTC)[reply]
I agree completely with Tryphon. Far too few people even realise that they can be involved in the Rf* process. More effort must be made to be inclusive. --Herby talk thyme 17:10, 17 September 2009 (UTC)[reply]
You are talking about something completely different, Kanonkas. We are talking about participation, not how the votes should be interpreted in the end. And I don't see how this (counting votes based on arguments instead of raw numbers) would be incompatible with advertising RfAs more widely. I would even say it's quite the opposite; how are you supposed to gauge community consensus, when only a very small subset of it is participating in the discussion? –Tryphon☂ 13:52, 18 September 2009 (UTC)[reply]
It's related during the process. This proposal, with the idea to "raise" awareness of individual RfAs, seems to promote the view of RfA being solely a straw poll and a democracy, which it isn't. We do want a part of the community to voice their opinion, though. As to your question: that's how it's been all the time. You don't get the whole community to voice their opinions on stuff either. You get a part of it, not the whole deal. As I said above, if it's necessary, we may want to consider making people more aware of the current RfA process. — Kanonkas // talk // e-mail // 22:12, 18 September 2009 (UTC)[reply]

I agree with FloNight. The change might be reasonable considering the user numbers alone but it is not necessary. The only promising measure is to increase attention to the votes, not a change in the voting policy. The latter will be more vulnerable to "campaigns" and might as easily lead to a less democratic procedure as an improvement. --chris 論 12:51, 18 September 2009 (UTC)[reply]

Hum - so not telling people what is going on is a "good thing"?
In practice "campaigning" takes place anyway - this might go towards ensuring campaigning did not work in that everyone would be informed anyway instead of just those who were approached? --Herby talk thyme 12:55, 18 September 2009 (UTC)[reply]
Totally. The more people know about an RfA, the more likely it will be a statistically representative sample of the community. –Tryphon☂ 13:52, 18 September 2009 (UTC)[reply]
To be clear. I'm in favor of making more people aware of all RFAs, including reminding other wikis of the ongoing need for admins and the need for more people participate in Commons community discussions (including FPC, Xfd, and Rfa and discussions like this one :-). Also, people might vote in their first RFA when a local wiki friend runs. Hopefully, some of these people will take an interest and stay. This is one way people get drawn in.
My main concern is that we will not consistently have better turn out for voting. Until, we know that is going to happen, I'm reluctant to raise the threshold for promotion. I briefly looked at the admins that were given sysop with less than 10 votes last year. I didn't notice anything worry at first glance. The ones I checked were actively using their tools. So, I'm not certain that we should be doing anything that would make it harder to promote someone with a similar number of votes. Automatically re-listing everyone with less than 10 votes seems like too much process if they have strong support and no opposes. And what happens if at the end of the second week they still are below 10? Not promoting someone with 8 or 9 supports after 2 weeks and no opposes seems wrong to me. FloNight♥♥♥ 15:09, 18 September 2009 (UTC)[reply]
Jumping in a bit, but what about making RfA's a monthly process (if there are any that came upin the previous month), to be announced for a few days each time? Ingolfson (talk) 10:35, 19 September 2009 (UTC)[reply]
An idea, not sure if it makes sense. In the French wikipedia, admins-to-be are asked to send a note on the wiki-fr mailing list to allow more people (those who don't keep an eye on the village pump, or the admin noticeboard, or the RFA, or wherever people advertise otherwise) to see the RFA. It might be something we want to do here as well? Moreover, commons-l is a rather interesting and useful list, so having more people there might also be an advantage in trying to strengthen the "community-life"? notafish }<';> 20:22, 21 September 2009 (UTC)[reply]
Good idea - we need a wider understanding & appreciation of Rf* process to encourage greater involvement particularly among non en members. Sadly I do not find the commons-l riveting though :(. --Herby talk thyme 08:28, 22 September 2009 (UTC)[reply]

Yes, this proposal is fixing a real problem. It is reasonable to set a threshold for the number of votes. It is also reasonable to require that the voters actually have some contributions here. Commons is too important to every other wiki to be lackadaisical about who can be an admin here. If someone can't muster 10 votes they are not well enough known to be an admin here. ++Lar: t/c 04:42, 26 September 2009 (UTC)[reply]

Up until now it seemed to work pretty well using 4/75% so I don't think that your statement is wholly true. :-)
Have the people that have received less than 10 votes and gained admin on Commons been more problematic as admins than the people that gained admin with more than 10 votes? And have the people that gained admins on Commons with less that 10 votes been less active as a group than the people that gained admin with more than 10 votes? If not then the change is not solving a real problem. If I saw information showing either of these was true then I would agree that there is the need for a change.
At the exact moment that Wikimedia is holding discussions about ways to increase diversity (part of the Stategic Planning), it seems like a rather bad time to increase the requirements for admin on Commons without any firm plan in place to increase the diversity of the admins here. If Commons is suddenly swaped with loads of people asking for adminship, then I think the criteria would need to be reviewed. AS well, up until now the number of people volunteering has not be huge so I don't think it is too hard to do a through evaluation of each candidate.
Per the other comments in the below thread, I hope that we're not going to use 10 votes for the quorum which would give the odd possibility that 8-2 would be promoted but 9-0 would not. FloNight♥♥♥ 10:57, 26 September 2009 (UTC)[reply]
If you want to carry out an analysis of admin actions with a view to deciding who's more problematic by all means feel free, it would be an interesting study (although publishing the results might result in fireworks). But, with all due respect, when I review this discussion, I find myself far more in agreement with Rocket and Herby, et al, (both users with multiple permissions here at one time or another and major contributors to the Commons "way of doing things"), than I do with you. Commons is not other wikis. Commons is Serious Biznis, and we don't want carpetbaggers. 4 votes is far too low, times have changed. The one RfA we had recently that didn't pass the 4 vote threshold initially was a fluke, bad PR or whatever, and it got sorted, the candidate ended up getting far more votes than needed (and also exhibited the sort of grace and consensus seeking that we want in our admins... as soon as that candidate realised there was a problem, they immediately pledged not to act, and asked that we extend the time so a true consensus could be determined. 100% a class act, we need more like that...). We are not going to have trouble getting 10 votes. ++Lar: t/c 16:23, 26 September 2009 (UTC)[reply]

Watchlist

[edit]

It might good to add these to the watchlist, but aren't there other fields that could benefit from more community input, e.g. CfD. Maybe these should be added too. -- User:Docu at 06:51, 10 September 2009 (UTC)[reply]

Set minimum support=8 not total votes

[edit]

20-Sep-2009: As noted under "Comments" (at top), setting the total votes to 10, would require 8 support + 2 oppose (as 80% > 75%) to qualify. Instead, just double the current support=4 to be reworded as support=8, for the following reasons:

Bahasa Indonesia: 3

- Reason R1: setting total votes=10 can foster hunting for negatives to get 2 oppose +8 support,

Bahasa Indonesia: (missing text)

as warping priorities ("Could 2 people please vote against our 8 votes?")

Bahasa Indonesia: 3

- Reason R2: by setting minimum as merely support=8, then the wording will compare easily,

Bahasa Indonesia: (missing text)

as just double: the new requirement would be clearly double compared with former wording.

Perhaps people who devised the former wording (as support=4) realized that setting a minimum total votes (such as 10) could warp priorities by coaxing negative votes to reach the minimum total. I like setting support=8 because then it could be compared, easily, that the requirement was doubled, if people were to ask, "How much more restrictive did admin-voting become in 2009?"
However, I'm not completely sure if omitting total=10 is the best plan, because perhaps setting a quorum of 10 people would be more objective, than just trying to get 8 supporting votes and rush to close the voting. Please discuss. -Wikid77 (talk) 12:52, 20 September 2009 (UTC)[reply]

Strangely I was just moving to a similar view :)
It would seem odd that 8 support & no oppose would fail while 8 support & 2 oppose would succeed. --Herby talk thyme 12:57, 20 September 2009 (UTC)[reply]
Each RFA is open for a week in order to get exposure. During the week, people have the opportunity to look through the contributions, and vote one way or the other. If the person does not get oppose votes, then I assume that they are okay to have the tools. If the process is broken, and no one is evaluating the candidates, then the part that needs to be fixed is getting more people to participate in the RFA discussions. My preference is that we emphasis that aspect of the situation rather than putting in rigid numbers increases that are arbitrary and might need to be overridden by a 'crat if a good candidate has been overlooked by the people that evaluate the RFA noms. But if we are going to use rigid numbers, then let's do it in a way that doesn't "promote a 8-2, while not promoting a 9-0". FloNight♥♥♥ 17:18, 20 September 2009 (UTC)[reply]
I think minimum 8 supports is sound. Minimum 4 is probably inadequate given the size of the project. But it seems illogical to say 8 supports + 2 oppose demonstrates more consensus than just 8 supports. Superm401 - Talk 21:25, 21 September 2009 (UTC)[reply]
I agree for all the reasons cited above. Walter Siegmund (talk) 05:01, 23 September 2009 (UTC)[reply]
In an odd way, 8:2 can be a more reliable indicator than 8:0 or 9:0. If a vote is 9:0 it may just mean that nobody has thought about the possible negatives. If the vote is 8:2 then 10 people have explored at least some of the positives and negatives, and there is 4:1 agreement that the positives outweigh the negatives. But I also agree with FloNight's earlier comment. Is this solving a real problem? Regards, Ben Aveling 00:22, 26 September 2009 (UTC)[reply]
Yes. The current threshold is way too low. ++Lar: t/c 04:37, 26 September 2009 (UTC)[reply]
Ben Aveling, could not the same be said for 10-0? What is needed is for people to regularly look into the candidates contribution history, and not go by name recognition of the editor. Otherwise, editors that are well known on a large wiki have the advantage of name recognition bringing in loads of votes, while someone coming from a smaller wiki has a disadvantage. Some regular editors on Commons look into the history of candidates, but they are not here every week for every RFA so sometimes I think name recognition plays more of a role than it should in the RFAs. FloNight♥♥♥ 20:22, 27 September 2009 (UTC)[reply]

On second thought, instead of 10 votes total, another alternative is to require 10 supports minimum and at least 75% positive votes, for passage. (this still doesn't sort how to count neutrals though) ++Lar: t/c 16:25, 26 September 2009 (UTC)[reply]

I agree that a minimum support level is better than a minimum support and oppose. If an RFA is going 9 1 it is silly to let the one opposer derail the RFA by striking their oppose. WereSpielChequers (talk) 20:11, 27 September 2009 (UTC)[reply]
Lar, I'm not sure that is any better. 9 supports with 2 opposes(81.81%), would not pass even though it is over 80%? That would be 11 people participating in the discussion. If it is going to be raised, then I think requiring 8 votes and 75% makes the most sense (but I still think it is not necessarily the correct solution to the problem since their is no indication that admins with few votes are more problematic admins. .) FloNight♥♥♥ 20:21, 27 September 2009 (UTC)[reply]
I think 10 supports (and 75% or better) is reasonable, not at all a hard thing to get, but I'd take 8 supports (and 75% or better), for now. I just don't want to stay where we are, it's too low. ++Lar: t/c 18:32, 5 October 2009 (UTC)[reply]

Shouldn't be making it harder

[edit]

I've failed RFA about 300 times. We shouldn't be making it harder to perform simple remedial tasks, like deleting images or (when the feature is turned on properly) renaming them. I don't think that there is the same scope for damage that there is at other wikis here either. Noodle snacks (talk) 22:26, 22 September 2009 (UTC)[reply]

In all fairness, Commons likely has the highest potential for admin abuse of all projects, since it could affect hundreds of other wikis. –Juliancolton | Talk 22:33, 22 September 2009 (UTC)[reply]
Harder? No offense, but we set the bar pretty low. Compare with en.wiki. As Julian said, Commons admins are entrusted with quite a lot actually. We can pretty much vandalize every wiki all at once by overwriting some commonly used image with goatse or something (including many protected main pages and other highly visible pages since cascading doesn't cross wiki borders and many projects don't or can't locally upload their own version, they also tend to forget the little things and just protect the POTD). In fact, I've run into a surprising amount of images here that are being used in the interface on various projects (even when there's an exact copy in the MediaWiki skins directory, the one that isn't editable). This power grows as Wikimedia as a whole grows. The ability to view deleted files and edit history globally is coming (hopefully) soon too. Anyway, it's not like this is a proposal to make the minimum approval 80 or 85%, we just want to ensure that we get a decent size representation of the community, whatever that may be. If we have to extend the voting time, so be it, but looking at the stats linked above, there's only one less-than-10-vote RfA that stands out as that should have passed anyway (and did), but that was somewhat of a special case (returning admin, long time inactive, newer crowd didn't know him). A few came close to the cutoff due to poor voter turn out perhaps, but for the most part good candidates have no trouble reaching the mark. Rocket000 (talk) 13:16, 23 September 2009 (UTC)[reply]
We need more admins, we need admins who actually use the tools, however we must have competent admins & I think we need a higher standard than en for the reasons above. I'm afraid "vanilla" en wp ones may not be quite there on the automatic basis that is sometimes assumed (I have to state some en wp admins are highly competent here). --Herby talk thyme 13:37, 23 September 2009 (UTC)[reply]

If you've failed "like 300 times" maybe there's a reason for that. Per Herby, if anything, our standards should be higher than other wikis. Rocket explains it well, this wiki is too important to take adminship selection lightly. ++Lar: t/c 04:44, 26 September 2009 (UTC)[reply]

[edit]

Please add links to relevant pages. For example, the reader needs to verify where the policy you are quoting is written. And to read the WHOLE policy page, not only the small tiny part you are wanting to change.

Please don't use acronyms nobody understands like RfA (In French RFA means Federal Republic of Germany) is it what you are talking about ? I guess not. Teofilo (talk) 07:20, 23 September 2009 (UTC)[reply]

It's one of the most well-known wiki acronyms there is. It's almost up there with NPOV. Clicking the link would have directed you the page that it was referring to. As for the policy, how about searching for it. Anyway, if we're changing it, what difference does it make? Rocket000 (talk) 13:34, 23 September 2009 (UTC)[reply]
The policy change proposal should include a link to Commons:Administrators so that the reader can see that Abigor's quote is not a faithful quote. A faithful quote would have been :
Adminship nominations usually remain open for seven days, for votes and comments. The period starts when you posted your nomination (if a self-nom) or when you posted your acceptance (if somebody else nominated you). At the end of that period the nomination will be closed by a bureaucrat who will either grant or decline promotion. The nomination may be left open for more than seven days, or a longer period specified, should a bureaucrat consider that desirable.

As a guideline, promotion normally requires at least 75% in favour, with a minimum of 4 support votes. Votes from unregistered users are not counted. However, the closing bureaucrat has discretion in judging community consensus, and the decision will not necessarily be based on the raw numbers. Among other things, the closing bureaucrat may take into account the strength of any arguments presented and the experience and knowledge of the commenting users. For example, the comments and votes of users who have zero or few contributions on Commons may at the bureaucrat's discretion be discounted.
Also the vote duration is important. If we raise the quorum from 4 to 10 people, the vote duration should be lengthened to 2 or 3 weeks instead of one. Teofilo (talk) 07:08, 24 September 2009 (UTC)[reply]
Disagree. 1 week is fine for adminship (I've proposed we raise 'cratship to 2 weeks though). If after 1 week there aren't sufficient votes, we can extend by a fixed amount, say exactly one week more. If after that, still there are not enough votes to exceed the threshold, that's a fail. ++Lar: t/c 04:46, 26 September 2009 (UTC)[reply]

Parkinson's law

[edit]

Factor I.—An official wants to multiply subordinates, not rivals; Source: Parkinson's law, 1955

Parkinson's law applies here. Abigor is an admin. Thus he wants to diminish the number of rivals. Thus he wants to makes things harder for people to become admins.

But the problem of Commons today is mainly that we don't have enough admins to perform admin tasks. See the Requests for deletion backlog on Category:Deletion requests April 2009. This is a 6 month old backlog.

Making things more difficult for people to become admins is not a good idea.

I appreciate the idea that adminship should be given only to people who handle the admin tools seriously. But the best way to achieve this is not to make the entrance into adminship more difficult. The best way to do this is to make admins accountable. That means restrict the adminship term to only one year. And make confirmation compulsory, like stewards on meta.

The present trend with an unaccountable small club of admins making things more difficult for other people with differing views to enter the club does not enjoy me very much. Teofilo (talk) 07:53, 23 September 2009 (UTC)[reply]

Are admins rivals?
Is WMF a business or a governmet department?
Does Parkinson's law apply to WMF?
Sv1xv (talk) 08:04, 23 September 2009 (UTC)[reply]

I forgot to say something: after the backlog problem is solved, when users become confident that requests for deletion are processed efficiently enough, the number of requests for deletion is going to grow even more and we will need even more admins. Teofilo (talk) 08:12, 23 September 2009 (UTC)[reply]

I guess you're wrong Teofilo. It's not more difficult to become a admin, but bureaucrats/admins are forced to listen to more - standard - users. Sumurai8-X (talk) 10:21, 23 September 2009 (UTC)[reply]
after the backlog problem is solved LOL. Rocket000 (talk) 13:36, 23 September 2009 (UTC)[reply]

What?

[edit]

Commons is a mature project with large amounts of files and activity. 4 votes sounds antiquated, a candidate on en.wikipedia or de.wikipedia could get that in the first day. Since we aren't at the size where number of votes isn't an issue (like it is at enpedia), but 10 votes and 75% makes sense to check that enough people have !voted for adequate consensus to exist.--Ipatrol (talk) 02:21, 26 September 2009 (UTC)[reply]

Yep. Candidates here often get 10 votes the first day as well. ++Lar: t/c 16:26, 26 September 2009 (UTC)[reply]

Math equations that will help us solve this problem

[edit]

I'v almost completed a math equation, really math equations that will help us solve this problem, if however my theroy is correct.I will not be able to explain how it works for the 3 or 4 days because I have a big test coming up. Heres what i can explain so far...

What if we put the power of the voters votes, in terms of a single persons power to make a decisions. We can add y intercepts in specific instances later to avoid raw number issues.

If we can conceptualize power as a quantitative value then we can relate "power" to other factors, such as dates, group size, reliability ratios and even rates of change between these and other variables. I hypothesis that if we graph these rates of change(s). And observe cretan plotted points, We might make some serious head way in solving this problem. Cozzycovers (talk) 06:39, 29 September 2009 (UTC) (Cozzycovers comments moved here from main page by Tommy Kronkvist (talk) 13:31, 29 September 2009 (UTC)[reply]

This may be too complex. ++Lar: t/c 18:30, 5 October 2009 (UTC)[reply]

Recap and suggestion for moving forward

[edit]
  • Many Commons contributors have expressed the view that the current policy for promotion at Requests for adminship needs to be updated because Commons is no longer a small wiki. The current guideline for promotion by 'crats is a minimum 4 support votes and 75% in favor.
  • A proposal to update to the current policy was started on September 8, 2009 by Abigor. The proposed wording reads:
"Promotion normally requires at least 75% in favour, with a minimum of 10 votes including support, oppose. Votes from unregistered users are not counted. However, the closing bureaucrat has discretion in judging community consensus, and the decision will not necessarily be based on the raw numbers."
  • Discussion during the last month indicates agreement from a significant number of contributors participating in the discussion that the 4 vote minimum seems outdated and a larger number of people supporting a RFA would be better. Reservations about the need for change were expressed by a minority of the contributors to the discussion. More people voiced different opinions about the precise wording for the change. In all, there is support for an update to the policy with a growing consensus that 8 supports and 75% in favor would be an agreeable change. Proposed new wording:
"Promotion normally requires at least 75% in favour, with a minimum of 8 support votes. Votes from unregistered users are not counted. However, the closing bureaucrat has discretion in judging community consensus, and the decision will not necessarily be based on the raw numbers."

I still think 10 and 75% is better but I'm not going to want to see us stuck at 4 so 8 is better than nothing. I note we still have not decided what to do about neutrals (how to count them). ++Lar: t/c 02:28, 12 October 2009 (UTC)[reply]

If we use 8 supports instead of counting a minimum 10 votes, it will solve most of the potential problems with odd vote counts that could happen and avoid the problem of how to count neutrals. No 8-2-0 or 7-0-3 being promoted, with 9-0 not passing. So neutrals are read purely as comments for the 'crat to use if they need more help deciding whether to promote, not promote, or extend. FloNight♥♥♥ 10:59, 12 October 2009 (UTC)[reply]
I think most are now on board with X supports and 75% support (I think X should be 10, some think 8) rather than Y votes total. I do not think it avoids thinking about how to deal with neutrals (remember, neutrals are NOT comments, we have both) though. Let's work an example... suppose we have a vote with 12 supports, (I chose 12 as it is above both 8 and 10 :) and because it makes the math easy ) 3 opposes and 5 neutrals. Is that a pass or a fail? 12s / (12s + 3o) == 80% but 12s / (12s + 3o + 5n ) == 60% (if you count the neutrals as "not supports"). If we want numerics, which way to go? If we want "bureaucrat discretion" ... I'm going to end up looking really hard at the supports and opposes and probably, unless the opposes are weak and the supports really strong, call it a fail. Is that what people want? I keep harping on "how do we consider neutrals" for a reason. ++Lar: t/c 16:28, 12 October 2009 (UTC)[reply]
I think someone made the point earlier that voting neutral has the same effect as voting oppose if neutrals are counted in the vote total. I don't think that is the intent of someone casting a neutral vote, nor does it make sense to interpret it in that manner. I would treat neutrals as synonymous with comments. Walter Siegmund (talk) 19:55, 12 October 2009 (UTC)[reply]
I agree and support the new (second) wording in boldface above. Walter Siegmund (talk) 16:17, 12 October 2009 (UTC)[reply]
Lar, how do you interprete the neutrals now? From watching the 'crats close RFAs, I thought that they were not figured as either as support or oppose for the purpose of a hard percentage count, but were used in the close RFAs to help inform the 'crat based on the wording of the neutral comment. I agree that using neutrals as you suggest would not be be significantly different than an oppose; and as Wsiegmund says, is not the obvious that most voters intend to use a neutral vote as an oppose. FloNight♥♥♥ 03:55, 13 October 2009 (UTC)[reply]
I can't speak for other 'crats but what I do NOW is count noses and calculate percentages, using neutrals as wild cards. If it is very obvious one way (96% or whatever even counting neutrals as opposes) or the other (64% or whatever even counting neutrals as supports), which most of these tend to be, and there's not anything odd about the noses I just counted, that's that, i announce the percentage leaving the neutrals out. It's the 'crat discretion area where it's less clear. In that case I look more closely at the votes and the discussion and the reasoning given, and most of the time I then involve the rest of the crats and we decide jointly what consensus appears to be. Evaluating neutral comments is an important (but not the only) part of that process. I think EZ might be a bit more mechanical than I am. A reason I want higher thresholds is that it makes it easier to determine what the will of the community is when more voices are heard. AND to make it more predictable about when an extension is called for. Missed the thresholds and it's been a week? Automatic extension. There's no rush after all, the wiki won't blow up because someone waited a bit longer.... missed the thresholds again after the extension? Probably a candidate that we just don't know enough about to promote comfortably. ++Lar: t/c 23:39, 13 October 2009 (UTC)[reply]
Retrieved from "https://commons.wikimedia.org/w/index.php?title=User_talk:Abigor/Vote_proposal&oldid=439743532"