On Mon, May 3, 2010 at 1:51 PM, Gregory Maxwell gmaxwell@gmail.com wrote:
On Mon, May 3, 2010 at 4:18 PM, phoebe ayers phoebe.wiki@gmail.com wrote: [snip]
Absolutely. But I would expect and hope that the median review time is minutes, for typical and uncontroversial changes. If it's not— then we need to improve the review process, because we're managing median vandalism revert times less than that.
This is a good point. Definitely and absolutely we ought to be able to do it in minutes if the subset is just semi/full protected articles. Flagging ability seems unclear if it's the whole 3M article set, but let's worry about that later...
<snip>
I don't think the goal of "More accurate™" is in conflict with "Maximally inclusive in whom is allowed to edit". Through the power of the default-view and the power of transparency we can have _both_, and I think the community has demanded a system which provides as much.
The challenge here is that the initial impression for "review" and "More accurate" is a secretive, restrictive, controlled, and slowly moving system... largely because this what traditional mediums provide. While "inclusive" is viewed as a crazy anything-goes anarchy (which was never really applicable to Wikipedia, even before protection). I think the message we need to express is that we're trying to combine the qualities of both extremes into a moderate composite which is even closer to the radical openness of the early Wikipedia while simultaneously being more accurate than the current system.
I don't know how to craft a PR message around this because to an outsider it sounds impossible for exactly the same reason that the whole idea of Wikipedia sounds impossible. People are very quick to jump to the 'restrictive' understanding because it makes Wikipedia finally make sense: "See! radical openness really doesn't work!"
Right. Even though I phrased it like that, I think we should probably stay away from "Wikipedia: now with more accuracy!" interpretations in the PR, because it's not true. What we're really doing is putting in a better vandalism & subtle-vandalism detection tool, the latest in a long line of improvements that started with patrolling RC by hand and that currently features our new bot overlords. The overall philosophy that anyone can edit is unchanged; there has always been a little asterisk appended to that phrase that says "but we don't have to keep your edit" and we are perhaps just now highlighting that in the software, for better or worse.
(The biggest problems seem cultural -- how much editorial judgment is used in flagging, especially on the high-profile articles).
Re: flagged/unflagged view, seeing all articles vs only the one you've edited, and whether to display a message -- can we learn from de.wikipedia & their experience? Is there a good usability-based way to do testing for these questions? (Has it been done, or discussed somewhere?) All I've got to go on is gut feelings one way or another.
-- Phoebe