On Thu, 2004-02-19 at 09:32, cprompt wrote:
On Tue, 2004-02-17 at 08:47, Michael Barnathan wrote:
Edits != Trustworthiness. Edits don't even accurately represent how active a contributor you are, as you can see from any long standing revert war, someone with thousands of edits but only 1 or 2% of them new articles, or even an article like Peerage, which has some 200 (valid) edits by the same user). Edits just represent how many times you type some text in and click a button, and I think basing adminship, a symbol of the community's trust in a user, solely or even automatically on how many times you click that button is absurd. It would help to lessen sock puppetry, though I don't agree that it will eliminate it. I think that monitoring the quality of edits would help eliminate sock puppetry more effectively; not too many people will wait months while making valid contributions to the encyclopedia on a sock puppet and keeping their original account active at the same time.
(Accidentally hit a key shortcut which I know now is "send". Continuing message...) Agreed. And aside from that, I've rarely seen some sort of automatic mechanism for advancement that was at all successful for any reason. There will always be people who will try to abuse the system. On forums, some people will bump threads and make "me too" posts to advance their rating. In roleplaying games, if a repeated action will lead to advancement, some people will spend an entire day doing that one action.
And it isn't even limited to people trying to abuse the system. Consider a new user who wants to be a sysop, and knows they need 500 edits be automatically promoted. They'd stop making large edits. They'd change a word here or there, add punctuation, and waste a lot of time, and clog the edit history, just to get their edit count up.
I think it's important to leave the process to the humans, and I also think it's fairly important that humans stop using the edit count as the /only/ metric in voting on adminship.