On Tue, 2004-01-20 at 13:23, Donnie Berkholz wrote:
> I didn't particularly want to lead the installer project, I was picking up
> the slack until someone else was interested in leading it because nobody
> stepped forward. Brandon asked to stress it at the meeting because he
> couldn't be there, I did, and I got given the responsibility.

Yeah i read the logs on that right now.

Tseng sees the real desktop issues as secondary, while the installer is
not a desktop thing as such (you never see it beyond your install). I
personally think such a vast project (it is a lot) is really beyond the
scope of this team and at least is not a good way to start defining what
this team is supposed to do. You yourself imply that any installer
release up to at least 2004.1 is very unlikely, I'm afraid it's gonna
suck resources from places where _the desktop_ needs them.

On the other hand i do see the need for concentrating installer efforts
(there have been quite a few) and try to get at least somewhere with
them. Either this project is too big to be completed by a few
individuals alone (as former failed installers might indicate) or there
just isn't enough interest to do it (yes i know it gets _asked_ about a
lot, but OSS isn't about asking, it's about doing).

So the quest for a GUI installer i see a bit as a Holy Grail thing for
now, I'd say we could use our round table for more real goals and at
least not put too much emphasis on the installer project.

On a side note, the one 'decided on' implementation detail of the
installer makes not that much sense to me, what good does it do to have
it pluggable ? We got a perfectly fine text based install with some
top-grade docs. We don't have to plug-in different widget sets, who
cares if the installer is QT/GTK/etc. (yeah, i want a Motif installer,
I'm oldskool you know). It seems like over-complication of what is
already a complicated project to me.

> If the rest of the research people also feel that your proposal is more
> important to pursue right now and should be the second project instead of
> the installer, I have no problem with that.

Well, it's not about one or the other, we can have both. The menu thing
has been long going and actually just misses some left-over research and
mostly implementation. But at least it's a goal i think we can achieve
within a reasonable time frame and is an obvious improvement to the
desktop experience as a whole (as opposed to an installer). I see it
didn't even get discussed, which is a bit of a shame. I thought you were
on of the initiators behind the initial unified menu proposal spyderous
?

What I'm trying to say is that we shouldn't start out with
desktop-research staring blind at castles in the sky and in the end
achieve little. I think it's wiser to set shorter term goals and work
from there.

- foser


--
gentoo-desktop-research@g.o mailing list