Commons talk:2008 Election suffrage poll

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


Comment[edit]

(Since this is a talk page...) I don't like the per month thing; can we have an option that just requires a certain number of edits (or even 100 in the last year, etc., if you like) as well please? —Giggy 14:32, 8 August 2008 (UTC)

Second this request by Giggy (talk · contribs). Cirt (talk) 14:43, 9 August 2008 (UTC)
Done and done! rootology (T) 16:59, 9 August 2008 (UTC)
Hm, looking at this, this is nice broad range of numbers now. We can go from either very to utterly lenient (20 month/2 months only) to lightly exclusive (50 month/2 months only + 300 total edits in the preceding year). This seems like a very broad but simple range for people to choose from. rootology (T) 17:03, 9 August 2008 (UTC)

A "no change" option is necessary here to confirm that the community does not want the status quo - certainly some people want no change. If done as a third sub-vote independent of the other two (something like a yes/no vote on "Is some suffrage requirement necessary?") it should not have an affect on the numerical part of the poll; which it would if "0 edits" was included as an option on the two above.--Nilfanion (talk) 18:07, 9 August 2008 (UTC)

Done. rootology (T) 18:14, 9 August 2008 (UTC)
  • What is the edit count required for suffrage on the ... poll on suffrage? (serious question). Naerii (talk) 18:39, 9 August 2008 (UTC)
  • Well, this is what I put in the header here:
This poll is intended to gauge what normal, regular, day-to-day Commons users want for this aspect of the local project. Votes by non-regular Commons users are welcome here (i.e., a user who spends 95%-99% of their time on English Wikipedia or Polish Wikiquote may participate on this poll). However, precedence of weight is given to normal/routine Commons users that are more familiar with this project, when this poll is closed and evaluated by our local 'crats. This is an internal, local project matter, and precedent, circumstances, or views from other Wikimedia projects may carry reduced to no weight on this one (and vice versa).
  • Anyone can vote since we have no suffrage requirements today, I should think, but to use a "hypothetical", if the bulk of someone's Commons contributions is to have shown up once to just oppose a previous vote on Commons, their voice on this poll may be treated as of less or invalid weight when our local 'crats close this vote. rootology (T) 18:43, 9 August 2008 (UTC)
  • I'm also concerned about how this may affect people with a few thousand edits who have been inactive for some time but see a request that they would like to vote on. I could support any one of these options with the proviso that after some cut off mark (say, 1000 edits) activity is no longer a requirement for suffrage. I think after a certain amount of time/edits/experience/involvement then you have a right to have your vote count as a member of the community regardless of your recent activity level. Naerii (talk) 18:44, 9 August 2008 (UTC)
  • I covered that above--it's still 'crat discretion, since it always will be. This just sets the baseline that they operate from, for when the next inevitable contentious vote or external conflict is disruptively brought here from whatever project, as we saw on Johnny's election that was canvassed versus. If someone has 2000+ edits but vanishes from five months, I'm sure the 'crats would take his/her history into account, and that's written in the poll above--reread wording on Option #1. Basically, as I see it (someone tell me if this is a stupid view) the people who work on Commons should decide this, the same as English Wikipedia users decide what happens on RFA or RFAR votes there--Commons users don't have as much weight/clout on English, ditto going the other way. Same as a German WP group of editors who suddenly arrives on Wikinews has no special authority or value just because there may be a lot of them mysteriously appearing en masse. rootology (T) 18:48, 9 August 2008 (UTC)
  • Ahahah, just goes to show how poor my reading skills are. Thanks for the clarification. Naerii (talk) 18:48, 9 August 2008 (UTC)
  • I agree with Naerii (talk · contribs) here about this example with someone with thousands of edits but less recent activity who shows up at RFA/RFB etc. In that case they should probably not be discounted, unless they have been inactive for a very long period of time. Cirt (talk) 19:25, 9 August 2008 (UTC)

Poll length and advertising[edit]

Since this is a rare and pretty high-impact matter with site-wide scope for all users, and people may not check their Commons accounts regularly if they are more active elsewhere, perhaps one month even for length, and stick it on MediaWiki:Watchlist-details as we did Picture of the Year previously? No one who logs in within the month then can conceivably say they missed it... rootology (T) 18:59, 9 August 2008 (UTC)

Good idea. Cirt (talk) 19:25, 9 August 2008 (UTC)
Someone will still say they missed it. Someone always does. --ShakataGaNai ^_^ 07:21, 10 August 2008 (UTC)
Heheh, too true. Cirt (talk) 08:45, 10 August 2008 (UTC)

Too many options?[edit]

There seems to be an awful lot of options now... so what happens if after a month there's no clear favorite? Maybe rather than making a concrete (and thus very gamable) system, just make this a simple expression on what the 'crats are supposed to weigh:

  1. At least some non-userspace edits (the current system in de facto terms)
  2. Recent contributor (20-30+ contribs per month over 2 months, but flexibility for long-term commons contributors who have not been active for a few months)
  3. Established commons contributor (200-300+ contribs)
  4. Established SUL contributor (extensive contributions on one or more wikimedia wikis)

I guess in theory people could just support all options they find acceptable, and oppose all others, allowing 'crats to weigh a number of things in any particular case. Of course, that could be a bit burdensome if there are 100's of votes on something, but even then it will probably only be a relatively small number of votes to entertain doubts about counting. --SB_Johnny talk 22:16, 9 August 2008 (UTC)

Excellent. I was just thinking there were too many options. Phrasing it in a way that makes the poll less set in stone and more of a way to suggest options to the 'crats is a very good way to go. Cirt (talk) 22:58, 9 August 2008 (UTC)
Those four points are really things up to their discretion already--these are like Lar mentioned elsewhere to hash out the expected baselines for suffrage, and the 'crats of course could and would adjust on a case by case basis. If we add a lot more this will turn into the last poll, with many many options, and not get a lot done, unfortunately. The "per year" is a hybrid of what everyone seemed to want in some form last time, combined with Giggy's suggestion that it be in the past year as the baseline. The "two months" thing was my idea, to reflect on people who were aware of the present status quo at any given time. Anything is gameable, unless suffrage is based just on "Well, I'm a 'crat, and I know this guy," but that doesn't help with people coming cross-project like we had in the RFCU to try to toss around their "other project" clout. There was another almost-incident on English Wikiquote just in the past 48 hours where elements of mischief from English Wikipedia moved over there to sway an issue over a username for an SUL issue. Nothing is absolutely game-proof and never will be, but these options seem to be the best ones to make sure that if push comes to shove on any contentious vote, that real "Commons" people--not people that just wandered over via SUL and have no history, usage, or stake in Commons--decide based on other project factors. Someone could have 40,000 edits on Romanian, English, or Italian Wikipedia, be an arbiter, and checkuser there, but that doesn't grant insight into whats happening on a project they don't legitimately take an active part in. We can add more stuff to the poll, sure... but if we don't have at least some simple starting point like it is now, then nothing'll happen. Specific to the SUL stuff, though, I think we shouldn't even worry about that. Its just a neat technie feature that shouldn't matter for voting in a positive manner. Local voters should decide local elections. :) rootology (T) 05:06, 10 August 2008 (UTC)
Oh, and what you're proposing is good, as an end result, but this is just to hash out "what numbers" people prefer. The results of this poll is what would form that (hopefully without SUL being a factor, IMHO). rootology (T) 05:29, 10 August 2008 (UTC)
Well, I'm not certain about this, but I think some people might want the "SUL editcount" to be factored in. We generally do that on Wikiversity, but then again we use different selection processes for sysops. --SB_Johnny talk 12:30, 10 August 2008 (UTC)
I strongly oppose that. Commons is big enough and has a vibrant enough community (unlike smaller wikis where it's hard to scrape up 25 votes in a CU election) that we really don't need people with no contributions here at all coming in to tell us how to do things. Sorry to be so blunt but that's how I feel, and I think it's the consensus viewpoint as well. ++Lar: t/c 13:28, 10 August 2008 (UTC)
Right, but it's never a bad thing to have some "proof" of that consensus. We had some very odd "silent majority"-type discussions on Wikibooks a couple years ago regarding "video game guides", and found (eventually) that having consensus on record was generally useful when faced with a minority view that insists that it's actually a majority view. --SB_Johnny talk 13:39, 10 August 2008 (UTC)

We need to use approval voting... vote for all the options you can accept. (which is how it's written) The one with the most votes gets enacted. The last poll, we analysed the total different number of voters and no proposal got a majority of all votes cast. This time I think with less options maybe that won't happen. But I think we need to take the one with the most votes, even if that's not a clear majority, because we have to move forward on this. As I said, I see this as guidance to the closers anyway, not a mandatory mechanical line. ++Lar: t/c 13:28, 10 August 2008 (UTC)

Agree with this last comment by Lar (talk · contribs). Cirt (talk) 13:30, 10 August 2008 (UTC)

we really don't need people with no contributions here at all coming in to tell us how to do things. I'm a user of the french WP and I really can't agree with that. If I really understand (but I'm not sure I do) this poll concerns only elections like RfA. In that case I could eventually agree with you. But the problem is that with the exact same argument you could forbid WP/WV/... users taking part in RfDs and that would not be acceptable since Commons is where other projects share their ressources. If the users of the other projects have no right to contribute in decision making in Commons, then they also should stop collaborating with your project at all (i.e. keeping their resources locally). I assure you that is not what I want but you need to keep in mind that Commons is not an independant project but a centralized resource library. Polletfa (talk) 12:33, 12 September 2008 (UTC)

Go live?[edit]

Well, there were previous polls: February 2008 and March 2008, which for Commons with the level of support there seemed a clear preference for the regulars here to have some form of suffrage. Shall we put this live tomorrow, as suggested above, one month duration, to give the 'crats the guidance they requested? rootology (T) 03:59, 12 August 2008 (UTC)

Works for me. Cheers --Herby talk thyme 07:16, 12 August 2008 (UTC)
Sounds good. Cirt (talk) 12:24, 12 August 2008 (UTC)
OK, set. rootology (T) 13:17, 12 August 2008 (UTC)

Proportional Vote Weighting[edit]

If you count voters with 10 or more edits as being fully counted votes, why not proportionally weight those with fewer than 10 votes with weight n/10? That way, someone with nine edits would have their vote weighted at 90%, and someone with one edit would have their vote weighted at 10%. This kind of proportional weighting would reduce the temptation for gaming the system by disingenuosly piling up a record of otherwise pointless edits, just to get one's baseline edit count up for future votes.

More generally, you can construct a weighting function that counts recent edits more than ones dating back many months. But if the basic threshhold is N edits over some recent interval, the weighting function could be piecewise linear for each epoch.

Moulton (talk) 18:26, 12 August 2008 (UTC)

What is the administrative burden?[edit]

I was wondering: If requirements for a number of edits per month and/or last year is the outcome of the poll, what will be the administrative burden for the 'crats closing the polls? Do the 'crats have some good tools for retrieving the edit counts the last month/year of a given user?

I am asking because I am a little concerned about the cost/benefit balance in introducing such requirements. I followed the RfC of SB Johnny closely, and although there was a lot of debate, I did not at any time doubt the outcome nor the 'crats ability to make the right decision. The 'crats have been selected for their competences at executing sometimes courageous/difficult decisions. I trust them and respect them, and I think they should trust themselves that they are capable of making the right decision.

If a vote is close, there is always a danger of executing the "wrong" decision. Let's say a user who would actually be a good CU/admin whatever, does not make the cut. Well, too bad, but not a catastrophe, and the user can apply again later. Anyway if the vote was close there was probably something the user could work on in order to get even better. There is also a danger of granting priviledges to users not quite up to it. The user may start out making some mistakes, the user may improve or be demoted again. Not a catastrophe, I think.

I have a slight feeling that exact requirements are a YAGNI (You Ain't Gonna Need It).

The rules leads to new exceptions and special cases, like the list of relatively inactive admins, who would not pass the edit count requirements.

How about users (like me), who makes an edit and then four minor edits afterwards to fix all the typos, because they never use the preview button (lol). They have an excessive high edit count as compared to their real participation, and we can go on and on about reliable metrics for evaluating activity...

How about the number of edits in talk name spaces. Are they not an indicator of involvement and knowing the business, or is it just users who like to be communicative?

I prefer in this case some more qualitative statements about activity, involvement, which is then interpreted in practise by the 'crats.

-- Slaunger (talk) 19:27, 12 August 2008 (UTC)

Some statistics[edit]

I don't really care what happens with the proposal, because I rarely comment on RfAs here. However, I reviewed a few stats, and found pretty shocking numbers with the X edits per month proposal. I reviewed 100 admins (the first 100 admins alphabetically, ignoring Elcobbola, who was not an active editor at the beginning of the numbers I took). Of those hundred, the following number would be able to vote in RfAs under the various proposals:

  • 50 edits per year: 92/100
  • 150 edits per year: 82/100
  • 300 edits per year: 74/100
  • 20 edits per month, for two months: 52/100
  • 30 edits per month, for two months: 43/100
  • 50 edits per month, for two months: 33/100

Moreover, I looked back to January 2008 and beyond, for users who had not met the edit requirement in at least one of the seven months. Missing the requirement in any one month means not being able to vote in RfA for the next two months. The following number met the requirement in each of the last seven months:

  • 20 edits per month, in every month this year: 35/100
  • 30 edits per month, in every month this year: 26/100
  • 50 edits per month, in every month this year: 18/100

Looking at the 30 edits per month criterion, which seems to have a decent amount of support, that's saying that 57% of current administrators would be unable to vote in RfA right now, and 74% would have been locked out of RfA at least once from March 2008 to present. I understand that bureaucrats will have discretion to count the votes of those who have special reasons for not meeting the criteria, but given that so many seem to be affected by the X per month proposal, that doesn't seem like a simple task.

Food for thought, anyway. Ral315 (talk) 00:55, 13 August 2008 (UTC)

How are so many inactive? Wow. Did you factor in admin actions and uploads as edits? Those count too. rootology (T) 01:06, 13 August 2008 (UTC)
I counted everything that showed up in Special:Contributions, which included uploads and page moves, but not admin actions. My understanding was that admin actions likely wouldn't be counted as they don't show up as an edit; if they were included, the stats would change somewhat.
As far as why so many are inactive, my guess is that it's vacations, short breaks, and the like. Most of the people who didn't make the edits per month number made it in every month but one or two this year, and had a period where they didn't edit. Bdk, for example, was largely inactive from January 26 through March 20, and while he/she made 50 edits in January and March, Bdk made only 14 edits in February. Finnrind was largely inactive from October 27, 2007 through February 16, 2008 (but had made a few thousand edits before then, and has made a few thousand since). Some I saw had breaks as short as two or three weeks, but the timing of those breaks left them without enough edits to make one or more of the minimums. Ral315 (talk) 01:40, 13 August 2008 (UTC)

Thank you for the stats Ral. This confirms (with greater detail and statistical support) what I've been saying. It's absurd that such criteria as being proposed in this poll would eliminate so many administrators from being able to vote. The culture at Commons is different than on the various language pedias. This is typically not the primary project of a participant here. I certainly hope somebody puts this poll out of its misery soon. It's poorly constructed, poorly thought out and is going to create substantially more problems than it could ever possibly solve. Even the notional problems it would solve are virtually non-existent and can easily be solved by bureaucrat discretion without creating cumbersome bureaucracy and red tape.

I'd also like to point out that though the stats were run for administrators, it equally applies to non-administrators. We have plenty of solid contributors here who would never meet the criteria suggested by this poll. Edit counting is just about the worst metric one can use to determine the experience and capability of an editor. When those enamored with this poll come up with some way of measuring how much editors read, how much they think about and how much effort they spend per edit, then maybe there will be something to work with. Relying on edit counts is akin to counting grains of sand on the beach in order to find out how enjoyable the beach experience is for a given editor.

It is shameful and repulsive that we should rely on such a method simply because there's supposedly nothing better. There is something better. It's called bureaucrat discretion. You know, those wonderful volunteers who have already been vetted for their ability to do the job? Maybe we should trust them to do their job than force entirely arbitrary metrics down their throats. If you're going to enforce any change suggestion in this poll, you might as well write a bot to promote administrators. --Durin (talk) 14:53, 15 August 2008 (UTC)

Edits=uploads?[edit]

Kinda says it all - to me edits=uploads. Looking at the page there seems to be confusion? Cheers --Herby talk thyme 08:37, 13 August 2008 (UTC)

I agree that uploads count as edits. —Giggy 09:37, 13 August 2008 (UTC)
And the poll explicitly says it, too. Edits/uploads/logged stuff/admin actions. Basically, do you leave some Commons paper trail? rootology (T) 13:14, 13 August 2008 (UTC)
Fair enough - that was how I took it but one or two comments seem to suggest others understood it differently. Cheers --Herby talk thyme 13:23, 13 August 2008 (UTC)
Yep. And I frankly can't think of a better way to gauge activity and participation in a site... than participating in it. I don't know why to some that makes no sense... rootology (T) 13:26, 13 August 2008 (UTC)
You & me both - nothing weirder than "other" people :) ( for the benefit of those who read this "weirder" is humour - ok). --Herby talk thyme 13:33, 13 August 2008 (UTC)

@Herby: That was before I added the upload part. :) Rocket000(talk) 00:04, 14 August 2008 (UTC)

I've been driving the CommonsHelper bot on a regular basis; these uploads don't appear in my upload log, they appear here. These count, too, and possibly this sort of 'count' should be mentioned over on yon poll. Cheers, Jack Merridew 04:43, 14 August 2008 (UTC)

This poll is procedurally inconsistent[edit]

We are told that "If you oppose any change from the current status quo of 100% bureaucrat discretion, do not edit this section, and note this in the Option #3 "No Change" section below." Yet it leaves people who voted in favor on some of the proposals free to vote in opposition to Option #3 "No change". That's a double standard which biases the poll in favor of the proposed change--a democratic relative of "have you stopped beating your wife yet?" --Itub (talk) 00:42, 17 August 2008 (UTC)

  • Indeed. Can we please kill this poll now? --Durin (talk) 13:55, 18 August 2008 (UTC)
  • The crats indicated they would be against that, and there didn't seem to be any general support for such a thing, either... rootology (T) 14:12, 18 August 2008 (UTC)
  • There is no way this poll will achieve consensus. Further, the structure of it creates a highly problematic result, even if some concept of consensus were achieved. Any poll which would disenfranchise large segments of editors here is not going to have a binding outcome. --Durin (talk) 15:53, 18 August 2008 (UTC)
  • You can't know if it will or won't get consensus and neither can I--a month is a long time. If you're so against suffrage for disenfranchising voters, two questions, and I would this time prefer direct answers to each, rather than bluster. ;) I am genuinely curious as to how and where your views are coming from.
  • 1. Are you as equally against the existing suffrage requirements on en.wiki's arbitration elections as you are these proposed suffrage requirements? If not, home come?
  • 2. You honestly have no problem with people who have zero existing stake or any contribution history being free to have their votes count on another project? If 30 German WP editors all come to an en.wiki RFA to nuke with Opposes an opponent, and they had 1-10 contributions ever on en.wiki, you'd be for their voices to count? rootology (T) 18:54, 18 August 2008 (UTC)
Ignoring your personal insult (bluster), I've already have in effect answered these questions. I've pointed out that the culture here is entirely different, which makes these suffrage requirements meaningless. Yes, I have no problem with any random passerby making a vote at an RFA here. Why? Because I believe that bureaucrats are well capable of and have the brains for discernment of drive by votes without having to put in ridiculous bureaucratic red tape to force them to discount votes based on entirely arbitrary and meaningless metrics. I fail to see some mass problem this poll is meant to address. Instead, what I see is "Hey, other projects have suffrage requirements, why not this one?" In case you didn't know, en.wiki doesn't have suffrage requirements at RfA. Bureaucrats are expected to sift through and discount such votes in close calls. --Durin (talk) 21:30, 18 August 2008 (UTC)
En.wiki is its own project. Commons really exists for the other projects. People upload stuff here to use elsewhere. Most people here edit elsewhere too, unlike enwiki. It is, to many, a second home to another project where they upload their images. Commons' nature is that it gets a lot of visitors, who will want to, and should be allowed to take part in elections and such as they please. Majorly talk 22:11, 18 August 2008 (UTC)
Although not directed at me, I'll answer those questions, too. (1) en.wiki, as has been mentioned, is in a league of its own. It is huge, well publicized, and an 800 pound gorilla of sorts. While I am not familiar with the arbitration elections suffrage requirements, I am sure that they are based on some of the unique circumstances found at en.wiki. Trying to justify anything from there with any other project is a tenuous argument at best, and especially true when looking at matters involving Commons. To many (most?) Wikipedians, Commons is just an extension of Wikipedia. It is the multimedia repository that is used to store all the photos for Wikipedia. As a result, Wikipedians very likely assume that any rights and priviledges that apply to them at en.wiki apply here, not the least of which is suffrage. So, naturally, if they wander over here and see a poll under way, even if it's something they only vaguely understand, they will expect to be able to comment on it as they would on en.wiki. I know I did when I first started visiting here regularly. (2) Looking at it from the point of view of a former drive-by voter, YES, you bet that I would expect my vote to count equally, even if my vote was my very first edit. If I take the time to vote on something, I expect that my opinion will at the very least be considered. Looking at it from an experienced Commoner, I don't particularly like the fact that someone could be a rabble-rouser on another project and send a swarm of drive-by voters over here to "fix" an election by dumping a slew of (probably well-meaning, even if manipulated) votes on the poll. However, I have faith in the process, and I feel that the 'crats are smart enough to recognize blatant manipulation of the system like that, and would regard the votes accordingly. Moreover, it has always been the well-considered comments that go along with votes that win me over (and usually the 'crats). There's no hard and fast number (1 excellent argument = 5 or 10 comment-less votes in support or opposition), but certainly votes that really probe the issue and come up with a sound argument (or at least indicate agreement with a sound argument) are more impressive than a long list of sheep-votes baa-ing a support or oppose. To me, anyone who comes up with such an argument deserves to have their vote count fully, regardless of any silly suffrage requirements. If there was a way to test for it, the only suffrage requirement that I would insist on is that people have a brain and use it a bit before they vote. :-) --Willscrlt (Talk) 09:49, 19 August 2008 (UTC)

Stoppt die Abstimmung[edit]

Den Text nur in einer Sprache zu veröffentlichen ist falsch! Es gibt sehr viele nicht englisch sprechende die auf commonce zugreifen. weingstens die häufigsten Sprachen müssen in einem solchen Falle zur verfügung stehen! ich habe auch nichts gegen eine kurterklärung. Mein englisch reicht nicht zu verstehen um was es überhaupt geht. Grüße aus der Eifel Caronna (talk) 12:50, 21 August 2008 (UTC)

My rough translation of the statement above:
Cancel the poll
It is wrong to provide a text in one language only! There are quite a number of people not speaking English who access Commons. At least the most common languages must be available! I wouldn't mind a short explanation. My English is not sufficient to understand what this is all about. Greetings from the Eifel. Caronna (talk) 12:50, 21 August 2008 (UTC)
--AFBorchert (talk) 15:03, 21 August 2008 (UTC)
Hallo Caronna, eine kurze Zusammenfassung über diese Abstimmung in deutscher Sprache findest Du hier. Ich beantworte dort oder auch hier gerne weitere Fragen dazu. Viele Grüße, AFBorchert (talk) 15:03, 21 August 2008 (UTC)

Site notice upgrade[edit]

Giggy upgraded the notification on this from watchlist-details to full site notice (I cleared it from the watchlist in turn). A good idea, and there's still 3+ weeks to go give or take. rootology (T) 03:08, 22 August 2008 (UTC)

10 edits per month[edit]

"50 in the past year" is one option that is a nice low bar, however it puts people who have only recently come to Commons at a disadvantage. Say they have four months of activity of 5, 10, 15 and 20 edits. They are well on their way towards 50 edits per year, but fall short of both the lower thresholds (20pm & 50py). I would prefer a "10 edits per month for last two months" as more users would meet that threshold, I'm guessing. Too late to add it? John Vandenberg (chat) 15:04, 24 August 2008 (UTC)

Probably too late to add it now... maybe if the voting and opinions work out a certain way there will be a run off election/poll consisting of extremely even lower numbers and an additional "no change" option, since those seem to be the leaders so far by far. Or they could just split the difference based on the poll and precedence (SB Johnny situation) and go with something like you suggest or a 5/10/20 a year one...? rootology (T) 15:23, 24 August 2008 (UTC)

Bad promo link[edit]

The link appearing at the top of every page for this: http://commons.wikimedia.org/wikipedia/commons/wiki/Commons:2008_Election_suffrage_poll And, it returns the lovely 404 error file not found before forwarding you onto the standard does not exist/create page.--Pro-Lick (talk) 19:23, 24 August 2008 (UTC)

Very weird. It seems to be inconsistent on my end. I asked a couple people to take a gander on IRC, I have no idea why its doing that, but I got the 404 as well. rootology (T) 22:22, 24 August 2008 (UTC)

Bias and Number of Edits by Users versus Admin[edit]

I am not well versed in code, and I do not have fluency in any language aside from English. However, I feel that giving Admin or users with large technical programming skill preference over a simpleton such as myself would be unfair. On the other hand, I would like to be able to trust the data that is posted in any language on any wiki project as I am studying linguistics in concert with modern culture and media. Fact checking and priority of world-view should be balanced by the average user, as opposed to those with the time and skill to make over 500 edits in two months. What this poll essential comes down to is lay wiki users ability not to only trust the information they are accessing but to have some input given their own perspective which is an inalienable right. However, making irresponsible edits and contributions should be overseen by Admin cooperatively in approximate real time depending on log-in overlap. Bias reflected in first-learned-languages of users or Admin should be retained and left to those who feel they can accurately represent trends of the culture of their languages. To exclude data based on language would be wrong, but to make cross-cultural comparisons inaccessible to lay users would defeat the purpose of any wiki project (which, I hope, is to allow learning, intellectual progress, and cross-cultural understanding to develop). I do not feel knowledgeable enough to cast a vote, but I hope this comment is regarded with utmost seriousness, as I have a firm belief that integration of language, fact, literature, art, history, and modern culture from many nations will lead to a more peaceful and productive global community. Thank you for reading. -VitaminN

Poll results[edit]

Some raw results (just vote counting; the final results can be slightly different if suspicious activity such as blatant canvassing or sockpuppetry is spotted):

Option 1
Two months of activity:
  • 20 per month: Symbol support vote.svg Support: 22; Symbol oppose vote.svg Oppose: 62
  • 30 per month: Symbol support vote.svg Support: 9; Symbol oppose vote.svg Oppose: 57
  • 50 per month: Symbol support vote.svg Support: 3; Symbol oppose vote.svg Oppose: 63
Option 2
Edits in the past year:
  • 50 in the past year: Symbol support vote.svg Support: 83; Symbol oppose vote.svg Oppose: 36
  • 150 in the past year: Symbol support vote.svg Support: 50; Symbol oppose vote.svg Oppose: 46
  • 300 in the past year: Symbol support vote.svg Support: 5; Symbol oppose vote.svg Oppose: 58
Option 3
No change, no firm suffrage guidance:
  • Symbol support vote.svg Support: 122; Symbol oppose vote.svg Oppose: 24; Symbol neutral vote.svg Neutral: 1

Some considerations from several comments:

  • The poll should have been translated to some main languages and properly announced on Village pumps prior to opening.
  • The structure of the poll is flawed because it is hard to assess if people wanting some sort of suffrage eligibility are more or less than the ones opposing such criteria due to distribution of their votes through several options.

Some sugestions from several comments:

  • Scratch this poll and start anew with a proper structure, well anounced and translated;
  • Scratch it altogether and live with whichever results we conclude from it.

My personal evaluation:

  • The only options that had more support than oppose votes were "50 in the past year", "150 in the past year", and "No change, no firm suffrage guidance". Of these the 50 vs 150 edits in the past year question is clearly favorable towards "50 in the past year". In terms of total difference between support and oppose votes, the clear winner is "No change, no firm suffrage guidance". besides, several people voted in different options, which introduces a bias hard to balance properly.
  • I don't think we can extract any good conclusions, and it clearly seems that the existence of suffrage eligibility criteria is not consensual. I'd say scratch the idea altogether and follow the "No change, no firm suffrage guidance" option, but since I voted for this, I'd like to listen to other points of view. Patrícia msg 11:58, 13 September 2008 (UTC)

Thanks for kicking this off, Patricia... I think we actually can draw conclusions:

  • The community clearly spoke in favour of the status quo as far as firm guidelines vs. 'crat discretion. By a wide margin, in fact. Now, some analysis of the oppose reasons might help... I did that, and I think most of them weren't saying they did not want any suffrage at all, but rather that they wanted to leave it to 'crat discretion. (that's clearly not true for a significant fraction, who by their comments indicated they want ANYONE, even someone who signed up after the vote started, and made no contributions at all, to be able to vote, but they are by far not the majority of the opposes)
  • So then... the community wants 'crat discretion. We asked for guidance about exactly where, and the input here is less clear.
    • All the monthly quota options had far less support than evaluating contribs over a longer term. (even the lowest bar, 20/month for 2 months, which is numerically LESS than 50/year, was defeated by a wide margin)
    • However, ff the various numbers/times the one that had the most support (Numerically AND percentage wise) was "50 in the past year"... this is actually a more stringent criteria than the "10 ever" we used in the recent CU election that drove some of us to want to seek guidance. We don't know if that is because this is the lowest one on offer, or if 10 was too low and more people favour it.
  • The conclusion I am drawing here is that as a 'crat, in close cases, I'm going to apply a standard, but use judgement about it. I'll discuss it with fellow 'crats and arrive at a consensus on what it is, and then apply it. It is likely to be something that relates to sustained contribution over a longer period of time rather than very recent contributions. But in close cases, I am going to continue to not favour allowing just anyone to vote. That is how other communities skew results and that is not what I want. It's also, based on my reading of all the votes (and especially their comments) not what a significant majority of users want either. Some want carpet baggers, but they are in the minority, and a rather small one at that. ++Lar: t/c 13:42, 13 September 2008 (UTC)
  • I agree with Lar. "No change" could be used in regular cases, but stricter counting needed in biased votings (we need to filter "just this voting" accounts and puppets). --EugeneZelenko (talk) 15:33, 13 September 2008 (UTC)
Then I think we were diverging in the raw reading but completely agreeing in the interpretation of the results. I too feel that the outcome of the poll favors giving margin to bureaucrats to apply 'crat discretion where it makes a difference; that 'crat discretion is to be used and votes appearing from canvassing can then be ignored; and that when there are doubts, there should be some discussion between bureaucrats and a consensual reading of the final voting. What I think is that we're not going to be discarding votes based on a defined number of contributions based on any result from this poll, but rather that we may do that within reasonable limits. Patrícia msg 17:55, 13 September 2008 (UTC)
  • I think the essential statement being made here is "we chose you for your judgement, use it". I'm fine with that, and I'll go with that. So yeah, in close cases, like in all consensus-y things, those who have a bit more experience, a bit more "know how" in the relevant areas (in this case the relevant area is the existence of Commons), will have more weight than those who are just driving by, but we're not going to be striking votes because they don't meet any specific criteria (rather, we do mental strikings (etc.) in getting a consensus). My stance on this. Giggy (talk) 23:47, 13 September 2008 (UTC)
  • Based on the raw results, I consider Option 3 the most supported one.--Jusjih (talk) 01:11, 14 September 2008 (UTC)
  • No change/bureaucrat discretion seems to be the favorite, but I find that a bit of a conflict we're using bureaucratic discretion to determine this :) (only a small one) Bastique demandez 02:23, 14 September 2008 (UTC)
  • I find the result to be not inconsistent with choosing 10 edits ever as an earlier poll was alluded to that was started with bureaucrat discretion and had such standards. Maybe I am reading my own bias too far into that interpretation. I see 100/2month (~600/year), 60/2month (~360/year), 40/2month (~240/year), & 150/year as clearly opposed, 150/year as slightly supported (nearly ambivalent), 50/year & no change as clearly supported. Ideally the next potential poll would focus on equivalents between 0/year and 150/year and, possibly following a w:Poisson distribution, would have a peak support at some usage level with support tailing lower on either flank. Such a result would be more ideologically secure than going forward with a poll where the lowest usage levels were chosen.
Even though people complain that it was flawed, I'm quite happy overall with its running. For one, it showed many people that some basic features like vote handling in Commons was not heavily structured and it gave those users a chance to participate in that construction. The question of different language backgrounds was raised. The idea of anticipating such a poll with a note on the village pump of many languages was used. Machine translation was addressed. Some people spoke about which languages should be covered. The importance of artificial languages was raised. I'd like to have some definitive minimal list of languages to translate into produced. Some Commons statistics were shown that told us what kind of usage users use or are used to (^_^). One other item was that this is a volunteer-based program and people that contribute heavily want something additional to show for their efforts. It seems they would like to have more of a chance to guide the evolution of Commons or, at least, to minimize sock-puppetry or meat-puppetry. We even had a multi-language discussion! So even if we can't use much of the results in terms of their intended purpose, we can still use the history of the results in future polls or votes. :)--Thecurran (talk) 16:53, 14 September 2008 (UTC)

It really is too bad there was no xx edits total option without the year or month requirement. (Unless option 3 falls under that one). That one might have had a chance passing, I know for myself I don't mind small edit requirements that much but the time limits stated really put me off those options. Garion96 (talk) 18:00, 14 September 2008 (UTC)

  • BTW, if we ever have another such vote on procedure, I would like it be considered that wiktionarians be as notified as wikipedians. I started on en-wiki but by the time SUL came around, I had found many more places to edit in en-wikt and I thus am marked as an en-wikt user possibly forever even though my edits in en-wiki now surpass my en-wikt tally. The only reason I raise this is I keep hearing 'pedia this and 'pedia that. We need to remember that Commons is part of each 'tionary and other projects like WikiSpecies, so I would like to see invites that are extended to the village pumps also go to the beer parlours. :)--Thecurran (talk) 01:56, 15 September 2008 (UTC)

I think the bureaucrat's comments above endorse the only sane reading of this poll (I don't fathom why this much drama was needed to tell us the obvious). Just please don't start another Vote over a minor issue. Surely the 'crats have enough guidance here now!--Nilfanion (talk) 02:12, 15 September 2008 (UTC)

The poll was seriously flawed but it gives us a lot of info. I don't have much time right now but here's my analysis:

  • We can't conclude anything from this poll based on the vote counts (I wouldn't even know which way to count some of them).
  • The way the poll is structured makes us seem more polarized then we really are.
  • I would say we want something but we can't agree or don't know what that something is. Most of the blame again goes to the poll itself—a lack of options or the wrong ones.
  • We should definitely not have another poll nor close this as non-conclusive.
  • I think everyone agrees, at least somewhat, that crat discretion should always play a roll (both ways) so a strict limit that can never ever have exceptions is out of the question.

Basically, I think we should have something written down somewhere (guideline, not hardcore policy) that 'crats can go by when needed. This is basically for those times when things get the way they get occasionally—the whole reason we had this poll in the first place. I don't think it's necessary to count up every single voter's edits in every RfA just to scratch out a vote or two, and even if counted, exceptions can always be made. So I say let's go with the 50edits/year guideline when needed otherwise pretty much no change. In this way, the question isn't "is 45 close enough in this case?" but "is now a time where such a policy is needed to unskew the results?" Rocket000(talk) 11:02, 15 September 2008 (UTC)