User talk:Talrias/Adminship reform

What are you trying to solve?
What exactly are you trying to solve? This seems to me like a solution desperately searching for its problem. I have seen only one or two cut and dry cases of admin abuse. One (Aevor... I can't spell his name) got speedily demoted, another (Guanco) had his case investigated by the ArbComm. 172 ended up having his adminship temporarily then indefinitely suspended, though I am less clear about his being in the wrong. The truth is that very few admins do anything out of order, and those that do get sorted out by existing systems. Which brings me round to my point. The current RfA procedure ends up giving fine admins. Why change it? smoddy 18:15, 24 Jun 2005 (UTC) (BTW, mentoring is a good idea, but I would prefer to assign it to new users, at their request.  That is a proposal I have been sitting on for some time, and may make formally soon)


 * The problem I am trying to solve is the voting procedure, not the issue of cut and dry admin abuse. I think that the voting procedure is just a bad way to appoint new admins. It gives fine admins, fantastic - that's not the problem. The problem is that we have an unclear method of solving borderline nominations, and that admins take action when there is no clarity in what should be done in the situation. My proposal offers a way to eliminate the first problem and provide clarity for the second one. Talrias (t | e | c) 18:21, 24 Jun 2005 (UTC)


 * No, it is not unclear. The bureaucrats are specifically charged with deciding community consensus by any appropriate means. How much clearer would you like it? I still see a failure to define specifically any problems. - Taxman Talk 20:31, Jun 24, 2005 (UTC)


 * The point I am making is that the current system does not facilitate their doing that, hence wat I see is a need for a new system. Talrias (t | e | c) 20:57, 24 Jun 2005 (UTC)


 * Of course it does. In 100% of the cases there has been a decision made. We can armchair quarterback them 6 ways to Sunday, but in the end, someone has to make a hard decision. Relying on numbers alone is far more problematic, so it has been decided that someone, specifically bureaucrats, should make a decision based on what they judge as consensus. In your summary of problems, a few have nothing to do with the admin nomination process, and the rest, I fail to see how those are a problem relating to the process. Those are problems related to how people view the process. Those people are wrong, not the process. The final one, that manual reversion is too tedius is spurious. For one, it is certainly not required for every admin nominee, for another it's not that hard to do a few, which is all that would be required anyway. So while I applaud you for making an effort to improve things, I agree with others that this is a proposal in search of a problem. - Taxman Talk 21:40, Jun 24, 2005 (UTC)

One problem you're creating is one of time management. Our time is valuable and should be spent improving the encyclopedia, rather than bureaucracy. Most admins have better things to do than mentoring a wannabe. Your solution won't necessarily work for appointing new admins, because there's the issue of assigning candidates to mentors, which could lead to clique-forming. The solution for your second issue (unclarity) is WP:AN. Radiant_ &gt;|&lt; 19:06, Jun 24, 2005 (UTC)


 * I think I said in the proposal that the person wishing to become an admin finds a mentor, not the adminee goes into a pool for selection. Regarding the time system, it would be a matter for debate whether the time cost involved at the beginning to promote an admin is worth it in the long run with having to spend less time handling admin issues. I don't think mentoring a 'wannabe' would necessarily be that time-consuming. Talrias (t | e | c) 19:36, 24 Jun 2005 (UTC)


 * really? Take a peak at some admin logs. An active admin on RC patrol can rack up dozens of actions a day. It would be a high intensity task and we are already short of higly active admins.Geni 23:19, 24 Jun 2005 (UTC)


 * Of course it wouldn't just be the mentor looking at the adminee's actions. There are many other people doing RC patrol, and they would be able to spot any "bad" actions the adminee is taking. Talrias (t | e | c) 11:10, 25 Jun 2005 (UTC)

What nonsense. Of course we have plenty of bad admins. &mdash; Chameleon 19:10, 24 Jun 2005 (UTC)


 * What is this in reply to? Talrias (t | e | c) 19:36, 24 Jun 2005 (UTC)


 * If you believe that to be the case, please gather consensus for your view so that you can be sure you are not the one that is off base, and carry through the appropriate procedure to have their adminship revoked or have them censured. RFC and Arbcom are your tools. - Taxman Talk 20:31, Jun 24, 2005 (UTC)

Talrias, I'm sorry to say this, but I have to agree with smoddy; this looks like a solution in search of a problem. Jayjg (talk) 19:13, 24 Jun 2005 (UTC)


 * I believe I highlighted some problems in the first half of the proposal. Do you not agree these are problems? Talrias (t | e | c) 19:36, 24 Jun 2005 (UTC)


 * Actually, it wasn't that clear to me what the specific problems were; was it about rolling back vandals and page protection? Perhaps you could give a one sentence summary of each existing problem as you see it, maybe in bullet form, before the more lengthy discussion. Jayjg (talk)  20:11, 24 Jun 2005 (UTC)


 * How's that? :) Talrias (t | e | c) 20:27, 24 Jun 2005 (UTC)


 * O.K., I don't understand how the first three are problems with adminship - I can't make heads or tails of the first one in particular. As for the last two, they appear to be problems with people using bad reasons for their voting, rather than bad admins being created.  That said, most people seem to use bad reasons for voting about everything (particularly political leaders), but I don't see a way around that in a democratic process. Jayjg (talk)  20:36, 24 Jun 2005 (UTC)


 * The first one is better put forward in the paragraph describing it, I had difficulty summarising it, and if you've got better wording please suggest it.


 * I do not think that bad admins are being created. I think that many potentially good admins are not being created, due to the system we use to create them. If admins were appointed through the system I have proposed, I think more people would be likely to put themselves forward for adminship, either on principle (like me) or because they think that there is not enough trust or faith in the current system.


 * do you have any evidence for this?Geni 23:19, 24 Jun 2005 (UTC)


 * The simple solution is to not vote on it, but rather have discussion about it. Wikipedia is fundamentally not a democracy, so why do we elect our admins? A discussion about creating admins, either as I have suggested it or in a different manner, would not only be a better guide of consensus but would also be more transparent. Talrias (t | e | c) 20:52, 24 Jun 2005 (UTC)


 * We don't elect admins. We make votes, and a bureaucrat judges consensus.  I think we should allow our sagacious bureaucrats to do their job – that was what they were selected for.  smoddy 09:19, 25 Jun 2005 (UTC)


 * An election is selecting by voting - your description sounds like electing in all but name. Talrias (t | e | c) 11:29, 25 Jun 2005 (UTC)


 * It is sort of an election, but not wholly an election, as it is judged on more than that.  It is similar to VfD for users.  I still don't see what your problem is.  There is discussion, and the bureaucrat uses their judgement whether to promote or not.  smoddy 13:44, 25 Jun 2005 (UTC)

I support this proposal entirely. I don't think I have anything else to say about it at the moment. -- Phroziac (talk) 22:03, 24 Jun 2005 (UTC)

Good points, odd solution
Talrias, you make some good points. Previous vandal-fighting isn't a strong requirement for being made an admin, but it does play some part in the decision ; perhaps access to 'revert' links should be distributed more readily than other admin powers.


 * This is a feature I am looking forward to in MediaWiki 1.5 - and while some people are saying that the adminship procedure should not be changed with the seperation of powers I think we would have a lot less controversy if rollback was handed out alone. Talrias (t | e | c) 11:36, 25 Jun 2005 (UTC)

People voting on RfA are more likely to think adminship is a big deal than the bulk of active users -- because those who feel it is a big deal will vote on RfA every week; whereas those who recognize it as not being a big deal will only stop and vote from time to time. It seems to me the community as a whole feels adminship is no big deal and should be granted to any reasonable user; but never has occasion or reason to take a stand about that.

Voting is excellent for those times when opposition turns up. Who is opposing, and why? Those are the key issues that determine a successful bid. Some aspects of opposition (the # of other users who philosophically don't like a user, despite that user's good will) grow in proportion to how active the user is... and so do votes of support. For this reason it makes sense to look at the % of votes which are in opposition.

Mentorship seems a poor substitute; relying on an extra single point of failure (the mentor) and compressing the longish three-month timeframe for curent votes (you must have edited for 3 months) into a short 3-week timeframe. Perhaps you would be satisfied with the addition of a mentorship option for users who want to become admins, but don't have a clear consensus of support?

+sj +  02:49, 25 Jun 2005 (UTC)


 * That sounds like an interesting alteration. However I don't think your summary of the proposal is quite accurate - it still relies on the bureaucrat at the end to make the final decision on whether to make the person a permanent admin or return them to "normal user" status. Also, the admin-mentor must accept the user as a mentee, so while the prerequisites to become a mentor might vary, I wouldn't expect them to change dramatically. In fact, through this system the time taken to become a proper admin is 2 weeks longer (voting on RfA is 1 week, mentoring would last for 3). Thanks for your comments :) Talrias (t | e | c) 11:36, 25 Jun 2005 (UTC)