Developer Only Discussion List

[Andrew <afrayedknot AT thefrayedknot.armory.com>]

> I am sad to see that a Team Lead will be leaving and appears to have
> decided to release the ISO on his own terms. I guess we can all just
> throw the roadmap away and wait and see if the iso makes our Feb 2004
> deadline.
> 
> I don't really know what else to say except, thoughts on the matter
> (developers only as it should be on this list)?

I feel compelled to share some insight. Hamish's post taps into what
appears to be some pretty well grounded frustration with the state of
this distro.

Why is Hamish frustrated? Hamish has put a lot of time and effort into
things here because he wants smgl to be something worthwhile, instead
of yet another sorry broken distribution. We are on the brink of '1.0'
whatever that means. To some, its just another stupid version of another
open source software project, and if history repeats itself, it will suck,
like debian and redhat and all those other sorry broken distributions. To
others, like myself, and probably Hamish, it means attention. We are
planning on hitting up a number of well read news sites for attention,
why are we doing this? we want a /.ing, we want our distrowatch rating
up, we want more users, better yet, more developers. Hamish (and my)
thinking is, that if 1.0 sucks, then we've just wasted our chance at
getting a larger user base, and a larger pool of developers. At least
until we go for 2.0, which is a ways away, but would be a whole lot
better with more developers.

Why else is Hamish taking the iso into his own hands? No offense to any
former iso team members, but there hasn't been a lot of new work with
the iso, basically the same overly simple installer has been floating
through the past 6 releases. No real improvements were made despite
lobbying for it many times, we always got the same answers. A lot of
Hamish's improvments should have been worked on a long, long time ago, but
instead they just went by and were mostly ignored, despite my (and several
other people's) suggestions/complaints. Its really a shame considering
how much sorcery has grown in the past year in comparison to the iso. But
enough on that, Hamish will get us a great iso, something that we can
rely on, and will be useful to new users (after all, the iso is the
first thing people use). If its not done by our 1.0 release date,
then thats that, so what if we have to 'throw away' the roadmap? By
my reasoning these improvments are months late, and if it weren't for
Hamish we'd release 1.0 without them.

I think Hamish is also frustrated by our lack of communication. We all
keep our sections up to date (*smirk*) or get features added/fix bugs in
our respective projects, then we attempt to have a monthly mammoth irc
meeting. I think we should have more meetings, they dont have to be huge,
just weekly team meetings so we know whats going on. Then maybe
a bi-monthly meeting for team leads to get together on whats going
on. Maybe we should have apache-like voting on things. iirc, this is
something like
-1: i dont like it
-0: i dont care either way
+0: I like it but I wont/cant do any work on it
+1: I like it and i'll help
Team leads could have more votes or something.

The first thing Hamish talked about was how a rock solid sorcery is the
number 1 goal for 1.0. While the bug count is zero, thats hardly what
we see. A number of people have lost their /etc/services file, for a
while I had mine up on ftp so people could get it back. xinetd files
are getting installed for no reason, same for init.d files (I asked
prometheus, and if I didnt have him ignore all that stuff, you'd get a
crapload of bugs about it). Maybe thats all been fixed, maybe there are
more bugs we dont know about. Personally Im terrified of having to help
out newbies on irc if/when a nasty sorcery bug crops up in 1.0. Just
because the bug count is zero doesnt mean there aren't any bugs.
Enough on that, the sorcery team has done a great job over the past year
and a half.

So where do we go from here? We find some iso volunteers to help Hamish
implement and test these new features. We figure out a new 1.0 release
date which will line up with when our iso is ready. We start squashing
the 300+ grimoire bugs, a lot of these are bugs prometheus found and
are likely to be easier to solve. We have a meeting to get all this
hashed out in person. We figure out where we are going after 1.0 (new
sorcery?, reorganization? etc). And most importantly, we compromise,
Hamish wants to possibly postpone our March 1 release date, so we push
it back and instead of crying about our precious roadmap, we can move on,
and make this distribution something we can be proud of.

Just my 2 cents.

-Andrew