next level of

Dave Cridland [Home] dave at
Thu Jul 17 13:21:02 EEST 2003

On Thu, 2003-07-17 at 10:20, Mark McLoughlin wrote:
> 	The point made about separating the specifications work from the
> implementation work is a good one. Maybe just a separate list for
> discussing specifications would suffice, though.

I'd seperate them entirely and completely.

Apart from anything else, the project hosting work may prove to have
different hosting requirements - it's only the project hosting that's
really pushing the server move, isn't it?

I'd also prefer it if the aims were modified in the standards area to
attempting to make de jure standards out of the de facto ones.

So I'd have: - Front page. - Current and working standards. - Reference implementations and other
core projects.

> On Thu, 2003-07-17 at 03:36, Havoc Pennington wrote:
> > Hi,
> > 
> > I'd like to see what people think about making some changes to
> > to create a stronger "center of gravity" for
> > X/Linux/UNIX desktop development shared between the desktop projects,
> > toolkits, and applications such as Mozilla and
> > 
> > Concretely, I'm proposing the following steps:
> > 
> > 1. We welcome desktop-related development projects on an
> >    indiscriminate basis; if it's desktop-related and open source, you
> >    can use hosting.
> 	I'm not sure about "indiscriminate" ... I don't think
> should host a whole bunch of dead-end projects. Maybe there should be
> some sort of a consensus on list that "this looks promising" before
> hosting it ... hard to make decisions like that, though.

Consensus on the list works for me. The definition of "consensus" can,
of course, be the same as used at the IETF working groups - Consensus is
reached when the chair(s) says it has been. Currently the chair seems to
be held by Havoc, and I'm happy with that - Havoc is reasonable even
when I'm in complete disagreement with him. (Yes, this is high praise

Additional chairs can be appointed by consensus, too.

> > 2. We move to better hosting facilities.

Certainly needed if project hosting is to be encouraged.

On the subject of CVS vs SVN, I'd go for Subversion every time. I've
hosted my internal stuff on it for ages, and I found it much, much
easier to use than CVS. I particularly like to concept of allowing
Apache to give authorization only to a particular branch, too. Makes the
"committers" much less like an elite cabal.

At the very least, a choice might be nice.

> > 3. We investigate the idea of making a versioned "desktop platform
> >    release" that would be a distribution with multiple modules, much
> >    like a GNOME or KDE release. It would contain a snapshot of stable
> >    tarballs for various desktop platform components.
> 	I'm not sure whether "defining a platform" would work too well in
> practice. You're always going to have bits some people don't want and
> not have other bits that some people do want.

Well, you're sort of talking about a reference platform, I think.

Which essentially means that people can use it, or write their own - if
they choose to write their own, and it fits with the standard, then
great. If it fits completely, then it's a second reference

However, this also means that shared *data* - say, common Desktop files
or whatever - needs to be completely seperated out, since this is
standards stuff not software stuff. I think this needs to happen anyway.

> 	So, yeah it sounds like a sensible, natural progression. I don't think
> you'll have any trouble finding people to help out ... but have you
> investigated funding yet ?

I've got 53p, a lollipop, and a half-eaten biscuit you can have.


More information about the xdg mailing list