Timwi wrote:
Luna wrote:
There is no reason to believe that Wikipedia has now reached a perfect state and requires no further changes to the way it works.
True, but that doesn't even come close to implying that this particular crazy idea should be implemented.
You are right; however, it *does* refute the previous argument that tried to justify that it "should not" be implemented.
The reason it should be implemented is because nobody can currently know whether it will work better than the current situation or not because no-one's tried it before. If it doesn't work out, we can still go back to the old system, but then we'll at least *know* that it didn't work and can address whatever problems came up. I have given several ideas why I think it might work, and why the proposal addresses existing problems.
That's why I suggested a random sample of 100 - subject only to some fairly low level conditions that indicate a willingness to stick around, and no obvious signs of dementia. Track that 100 for three months to see what happens. Also track the 100 most recent RfA appointments and compare the statistics for the two groups. Why should anybody object to controlled results?
What's the line of defense when somebody lines up 20-30 admin socks, blocking and desysopping everyone in sight, resysopping each other and retaliating when anyone tries to stop them?
Surely this is the same old argument that also says that wikis can't work because anyone could rack up 20-30 socks and keep reverting their favourite articles to their favourite biased version.
If you let more people be admins, there *will be* enough admins around to block 20-30 misbehaving accounts. Notice that if one of them gets blocked (desysopped) for good reason, one of the socks resysopping him would also qualify as misbehaving; eventually all of the socks will be blocked (desysopped). Vandals will soon realise that it is not worth it, and after a short initial rage, people will stop even attempting to create 30 sockpuppet accounts.
The idea that there could be 30 sockpuppets in a random sample of 100 users is out of touch with reality.
Ec