On Sun, 2008-02-24 at 10:54 +0100, Dave Page wrote:
> On Sun, Feb 24, 2008 at 10:24 AM, Magnus Hagander <magnus@hagander.net> wrote:
> > On Sun, 2008-02-24 at 01:49 +0100, Dave Page wrote:
> > > So I was thinking some more about the lack of real use that techdocs
> > > is getting coupled with the lack of any kind of spam etc. that we've
> > > picked up on it. I think we should reconsider having an official
> > > PostgreSQL wiki for users, probably at wiki.postgresql.org.
> > >
> > > Thoughts?
> >
> > I guess I'll have to concede. It's quite clear that the techdocs we have
> > now doesn't work the way we intended. I'm not entirely convinced that a
> > wiki will work much better, but since others are, let's give it a shot.
>
> Hey, that was easier than I expected.
Well, you're in the same city as me right now, so you might just beat me
up if I'm not careful.
BTW, let's discuss this over beer later tonight ;-) I assume you're
buying?
> > AFAIK, Stefan has some experience with the options there - what wiki
> > software should we be using? Do we still have one VM empty on the boxes
> > in France we could use?
>
> I just spoke to Stefan - we do have a spare clean VM pretty much ready
> to go, originally provisioned some time ago to become a new developer
> VM for the webteam, but I never got round to moving everything from
> developer.pgadmin.org. We can put that to use.
Sounds like a good idea. We do have a working developer.pgadmin.org
right now, so finishing that migration isn't critical. Though we'll
certainly want to do that eventually, but there's no rush.
> The other thing we need to consider is tying this into the existing
> community logon system. I believe mediawiki now supports OpenID, so we
> should just see about hooking a provider into the existing database.
Yeah, I've been chatting with Stefan a couple of weeks back about using
OpenID to fix the login issues with his prototype bugzilla install.
Let's bite the bullet and look at implementing that. We could use it for
the developer wiki as well, as well as other potential future services.
I've looked a bit at OpenID for a different project at work, and it
seems it should be fairly easy for us to implement. There are good
libraries to build off - we'll certainly need to change it a bit to work
for us, but it's doable.
But I fully agree - tying into the existing login system is a
requirement, not just a "nice to have".
//Magnus
