Re: State of the Woody
Anthony Towns wrote:
> So I think the thing we probably need to focus first on is getting some
> way to install woody directly. Which means we need to decide whether
> we're going to keep hammering away at debian-installer until it actually
> becomes usable, or if we're going to bite the bullet and just accept
> that we're going to use boot-floppies again, not be able to improve
> them much and get flamed for it in all the woody reviews too.
I am sure that the debian-installer project will be a usable way to
install Debian by the time woody is released. It just won't work in all
situations. That may mean that we have not got it working on all
architectures, or what we only support network-install-from-a-single-floppy
(the first thing we are targeting), or that it just doesn't support some
minor detail like installing via nfs. Probably somewhere in between these
points.
We realized this a while ago, and thus decided that the boot-floppies
would have to remain the primary or at least back-up install method for
woody. And I believe that when Adam Di Carlo made that projection that woody
boot-floppies would take a month or two of work, he factored in an assumption
that much development effort would be sucked away by the debian-installer
project.
> If we do go with boot-floppies for woody, I'd be strongly inclined towards
> downplaying debian-installer so as not to divide the testing effort and
> leave us with a choice of two installation methods that both suck. To get
> boot-floppies to work with woody at about the same quality as potato's
> install, will likely take a couple of months; to get debian-installer
> to work at a similar quality could take... four months? six?
To get debian-installer to work at the same level will take at least one
cycle of testing by a sizable fraction of our entire user base. There is
no way we will catch every edge case without that. 5% would probably do;
tucking it away in an obscure directory and relying on word-of-mouth
that this option exists if you can find it will probably NOT do.
I don't know what variety of downplaying you have in mind; my plan has
been to have it available in the main archive (as it is) and probably on
CD's, and have a single pointer to it in the installation manual, with
hopefully a sizable buzz about it amoung those who use it succesfully,
and no downplaying of it at all to the people who choose to develop it.
Much more downplaying has the possibility of killing the project; not
getting us enough feedback so it is not fully usable and viable when
sid's release rolls around.
--
see shy jo
Reply to: