Roy T. Fielding wrote: > I am confused as to why we need a wiki. Labs is supposed to be for > internal projects by committers, whereas a wiki is best for contributions > from the general public.
I would say, ease of content management. Writing in a textarea some wikitext feels more efficient than writing HTML on your text editor (or structure-messing HTML editors) and then svn commit it in and then log in and svn update it there and then wait. I don't think it's about public access at all, in fact, I think none here would be opposed to a publicly readable but only committer-writeable wiki, sort of a content-management companion to svn. That said, a wiki is only one way of solving the CM problem. Maybe there are others (protected web page using ajax/web_dav over svn autoversioning mounted repo?) > Of all the wikis I've used, MediaWiki seems to be the best overall > if you don't mind the content being stored in MySQL. Agreed. It's not only about the software itself, its about the community around it: as long as wikipedia is around, mediawiki will be actively maintained and improved. Of all people, we should be the ones understanding how important that is in the long run. And maybe we can create a lab to have mediawiki store its content in subversion instead ;-) > However, the > DekiWiki fork made for an impressive demo at last week's > Gilbane content technologies conference. > > http://www.opengarden.org/dekiwiki I'll take a look. > As far as infrastructure is concerned, our current main wiki is MoinMoin > and it is "barely maintained". To install a new one would probably > require agreement to upgrade from MoinMoin, which is a lot of work. I'm not suggesting to infra@ to change (even if maintaining mediawiki is sure a safer bet, on the long run, than moinmoin, if only because of wikipedia), but I think that labs should be allowed to experiment to stuff that is also close to infrastructural needs, because it allows to show alternatives and gather momentum around options that might also be useful to tunnel thru existing impasses. That said, I'm not an infra member therefore I will gladly let them make their decisions because they know better. I just would like the opportunities to show alternatives and the potential benefits that they might bring to the table. Infra is, operationally, the diametrical opposite of labs, so it's natural that some impedance mismatch happens. Remember, in '97 I was the one advocating installing java on our apache server against a Brian that didn't trust it. We have come a long way from then, gradually and without disruption. So I'm perfectly willing to apply a little pressure and then wait patiently for it to act. At the same time, this very project was established as a way to counterbalance the perceived 'ossification' of our structures and I would like it to be allowed to experiment in all directions, including the services that we provide to our users (of course, without requiring an excessive infra tax to pay or guaranteeing a long-term persistence of such service). -- Stefano. --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
