Skip to Content.
Sympa Menu

sm-discuss - Re: [SM-Discuss] project organization

sm-discuss AT lists.ibiblio.org

Subject: Public SourceMage Discussion List

List archive

Chronological Thread  
  • From: Jason Flatt <jflatt AT sourcemage.org>
  • To: SM Discuss <sm-discuss AT lists.ibiblio.org>
  • Subject: Re: [SM-Discuss] project organization
  • Date: Tue, 14 Feb 2006 08:20:55 -0800

I have so many things I want to say to this and to other messages that have
been posted recently, but I cannot seem to come up with the right words, and
I can't seem to get the time set aside to stop and think and come up with the
right words. (Not really a good thing for someone in charge of
documentation.) I also don't want to wait until this is a moot point.

I, too, have been bereft of time to spend in my newly re-appointed position:
I
have a new baby, my largest client has started requesting more time from me,
such that I am now working about 60 hours in the field (It seems that the
more I try to get away from that sort of work, the more I do), and I've been
trying to get my head back into an area of computing that I have largely
ignored for a few years.

I have great, big, wild, wonderful ideas about improving the state of the
documentation and the Web site (areas I think of as the "face" or "public
interface" of Source Mage), yet I find that currently I can barely keep up
with the e-mail (which is really not that much, even with me being subscribed
to all of the lists).


I believe there is more here than just the one issue of whether to have a
Projet Lead (PL) or a Council of Elders (CE). It feels to me, to a small
degree, like there is already a CE in place, it's just not called that. I
feel like the Team Leads (TL) that are currently in place step in to fill in
the gap or help in an emergency already, even if it's not "their area".
True, there is no redundancy, but I think we are too small to have a lot of
that right now. Where I feel we are lacking is not in leaders, but in
leadership.

The project feels to me to be very lopsided. There are many "teams" of only
one or two people, and most of those people are labeled leaders. On the
Developers page, there are teams that have many people, but only one person
actually doing anything, whether leading or working. The concept of a leader
is a person who is directing others (as well as him or herself) and a project
or sub-project toward a goal or end point of some type. A team of one is not
a team and does not have a leader. It is one person doing a job or task of
some sort and should probably be defined as such. The only team that really
feels like a team is the Grimoire Team, and only because there are so many in
it.

The Grimoire Team has not felt like a "real" team for a long time. I believe
it is because more and more people are signing up as General Gurus, rather
than Section Gurus. They take no ownership of the project or sub-project
they are a part of, and therefore do nothing, or they do little with no real
direction or guidance. I still feel very much like the Printer Section is
"mine" because I took ownership of it while I was a guru. I have to keep
reminding myself that it is no longer my responsibility and I need to focus
on other things, especially when someone else makes an update to it.

Another point I would like to make about leadership is that it is not the
leader's "job" to do the work, but it is the leader's "job" to get the work
done, usually by directing others to do the work. Because we are so small
and lopsided, it is usually the leader that is doing all of the work, and
because of the state of the project, is expected to do the work. In my
opinion, the PL should not be fixing bugs or implementing features in the
Grimoire or Sorcery nor creating & updating documentation, but directing the
leaders (or workers, if there are no team leaders) of those areas to do that
work. That goes the same for all areas of the project. I, as the Tome Team
Lead, should be doing the work of getting the job of documenting the project
done mostly through the use of others' help and efforts. I, as a Tome Team
member, should be doing the work as directed either by the Tome Team Lead, or
by the PL, if I am the only member of the Tome Team.

I am not suggesting that we need to take on the attitude of micro-management,
but that of actual leadership. We should not be telling others how to do the
job, unless it matters to getting the job done. For example, a member of the
sorcery team should not be implementing a new feature in PERL or Python,
because it has been established that we are using Bash. We should not be
telling that person how to make that feature happen, as long as they use the
established coding standards, as long as the feature works correctly, and as
long as it does not break anything else. (Please don't nit-pick the
examples, if they are not 100% accurate. They are only meant to be examples
to hopefully clarify my points.)

Also, I feel there is no clear direction for the project nor the
sub-projects,
and hasn't been for a looong time. I have felt for many years that we were
all generally working toward this non-discript 1.0 target, but I have not
really felt that I knew what that target was. Yes, we have a roadmap, and
yes it has some set mile markers, but I feel like no one /really/ knows where
that road leads to, just "some where over there". Everyone is just kind of
doing their own thing progressing toward a Source Mage GNU/Linux 1.0, but it
hasn't really been set down as to what that 1.0 goal will look like. Mostly
we all just kind of keep adding features and fixing bugs and "doing" things
as it suits us. We have had meetings in the past where certain things are
set as necessary for 1.0, and by the time we are half-way toward the intended
release date, that list of things has changed. It makes it really hard to
get to 1.0 when the target keeps shifting. We need to take the hard line, if
we really want a 1.0, and specify what 1.0 will really look like. Then we
need to say "no" to people's pet needs for 1.0, when they crop up, especially
once the definitions have been set, and say "yes" to making it happen,
whatever it takes.

Finally, while we are a project of many diverse people, and we are trying to
work together as a team, I feel we spend faaaar too much time debating the
minutiae and far too little time actually moving forward. I don't think it
is a bad thing to present items for discussion and recommendations and
approval, but I think there needs to be more nipping in the bud of
unnecessary conversions and more actions taken. It seems like we debate
things for weeks, then the topic finally dies with no action being taken,
whether that action is good or bad, by anyone (whether because everyone has
lost interest in the topic through the discussoin or what, I don't know), and
then gets brought up again 6 mos or 2 yrs later only to repeat the same
process. Someone (the PL, Assistant PL or the relevant Team Lead) needs to
make a decision, end the conversion and start things rolling. The endless
debating over the same points is deadening to the topics and the project in
general.


I think that is it. I'm sure later today I will remember something I forgot
to add, but I'm late for work and don't want to delay sending this another
day.


One last thing I would like to add, the following link was forwarded to the
Drupal development mail list. It is an interesting message that I believe
speaks to the current state of the Source Mage project quite well. I think
if you replace GNOME with Source Mage, it will paint an accurate picture of
our project.

http://mail.gnome.org/archives/desktop-devel-list/2006-February/msg00115.html


--
Jason Flatt
Source Mage GNU/Linux: Linux so advanced, it may as well be magic.
http://www.sourcemage.org/

Attachment: pgp88D0vKcfuQ.pgp
Description: PGP signature




Archive powered by MHonArc 2.6.24.

Top of Page