Keith and Fred are both suggesting things that make Wiki not Wiki, but would work for a free online encyclopedia in the long run. Do we want anonymous editors spamming their URLs all over, or writing "poop" or "Andrew is gay" in major pages? It's not wiki to lock them out, but we're writing an encyclopedia.
I have recently seen numerous instances where anonymous editors have removed vandalism, often within 30 minutes (or even quicker) on pages that are not especially popular. This means several things. This means anons can contribute very constructively, but it also means that our traffic has reached the point that pages are served so often. I wouldn't expect every visitor to correct the vandalism (figure of 1:6 comes to mind). Therefore, every epidosde of vandalism is witnessed by countless visitors until a kind soul picks it off, either through RC patrol, a Wikipedian with the article on his/her watchlist, or said anonymous saint.
Fred is suggesting a para-wiki where arcticles are fact-checked before they go live. I would like to propose that as soon as an article reaches featured article status, it is semi-protected. Only registered users (or even registered users with a good track record) can edit those pages. Changes can be proposed on talk by anons and then be effectuated by registered users. This will put an editing funnel on our most prized content.
Jfdwolff
J.F. de Wolff wrote:
Fred is suggesting a para-wiki where arcticles are fact-checked before they go live. I would like to propose that as soon as an article reaches featured article status, it is semi-protected. Only registered users (or even registered users with a good track record) can edit those pages. Changes can be proposed on talk by anons and then be effectuated by registered users. This will put an editing funnel on our most prized content.
I agree that the most obvious place to start is by in a sense protecting the articles that are deemed to be pretty good and therefore less in need of edits---these are the ones where the ratio of destructive to constructive edits tends to be higher. I would prefer a system that fell short of *actually* protecting the pages by limiting edits to a certain group of people, though. If the issue is whether people will see vandalism before it's reverted, this can be accomplished through the proposed rating system in a number of ways. For example, articles that have a version rated above some threshold may be default display that version to non-logged-in users, rather than more recent versions. If changes are made that do indeed improve the article, there would be a lag until the new version is recognized as also high-quality and displayed; if they were destructive changes, the new version would not get displayed in the first place.
There are a number of issues for how to work this out, both in terms of community and logistics, but I do think a way to go is to introduce a display lag of some sort on articles deemed "already very good", but *without* actually restricting.edits---instead merely delaying their general visibility.
-Mark