On 9/9/07, Bryan Derksen bryan.derksen@shaw.ca wrote:
Anthony wrote:
On 9/9/07, Bryan Derksen bryan.derksen@shaw.ca wrote:
Stable versions doesn't have to do anything visible if the default is for people to see the most recent version rather than the one marked stable. Enable it, let people noodle around figuring out the procedures for what to mark, and if after a while the resulting version marking looks good maybe then make it the default anon view.
The thing is, if stable versions don't have to do anything visible, then the developers don't have to implement anything in the first place. People can just stick a note on the talk page for an article saying "I declare [http://en.wikipedia.org/w/index.php?title=Hey_Ya%21&oldid=156732871 version 156732871] to be stable" and other people can say "*I agree" or "*what are you high?" or whatever.
But once this has been tried, assuming enough people actually follow this procedure to give meaningful results, what then?
Then convince the devs to accept a patch making stable versions the default non-logged-in page.
The software doesn't "understand" what those links in the talk pages mean so there's nowhere to go in terms of expanding Wikipedia's functionality based on it. You need software to be able to do things like have a "show stable version first" as a user preference, or to create a stable-version-only database dump.
If the information was at all organized (some sort of template system, for instance), it'd be quite easy to use it to create a stable-version-only database dump. "Enable it, let people noodle around figuring out the procedures for what to mark, and if after a while the resulting version marking looks good", then work on importing the information into the database. I imagine choosing a stable version is going to take some time. If 1000 stable versions can be chosen over the next month and it takes 10 seconds to import each result into the database, that's less than 3 hours of work importing the information into the db.
And besides, the basic point of my complaint is that after years of people talking about stable versions absolutely _nothing_ has been implemented
This doesn't seem accurate. See for instance this thread: http://www.nabble.com/Stable-article-versions-t3656072.html
"The developers working on this feature [stable versions] have it implemented on a test server right now."
At the same time, I can't figure out just what it is that they're talking about. Erik Möller and P.Birken are listed as the people to contact for more info.
http://www.mediawiki.org/wiki/Extension:Review
Maybe that?
And as you pointed out, Featured Articles get linked. And so do Good Articles. So combine the 1591 Featured Articles with the 2857 Good Articles, and we've got a good candidate for the stable version of 4448 articles.