sm-discuss AT lists.ibiblio.org
Subject: Public SourceMage Discussion List
List archive
[SM-Discuss] Remove details from SMGL ( An SMGL fairytale )
- From: Hamish Greig <hgreig AT bigpond.net.au>
- To: sm-discuss AT lists.ibiblio.org
- Subject: [SM-Discuss] Remove details from SMGL ( An SMGL fairytale )
- Date: Sat, 3 Jul 2004 12:01:27 +1000
I am resending this email for the second time. The first time It awaited
moderator's approval, but unfortunately it was deleted without any notice or
reason sent to me, so I resubscribed to the list for the express purpose of
sending this.
What follows is a brief history of SMGL for the period I was a developer, it
explains why I left, and why I am now asking for my name and email to be
removed from any file in the grimoire. A piped grep, p4 and sed command would
achieve this, but if it is too much work then contact me privately, give me
access, I will remove myself and then you can remove my access.
The History:
+++++++++
SIck of rpm dependencies and not truly having control of the software on my
computer I left RedHat and installed "Sorcerer GNU/Linux" a few weeks before
the fork. I stayed with SMGL and later I reached a point where the grimoire
bugs I was fxing in my local files were only recently fixed when I went to
report them to bugzilla (fix committed but a new tarball had not yet been
generated). This lead me to volunteer my services as a developer, I could fix
the problems I found and someone with better skills would then have more time
to concentrate on more important issues.
After I became a developer I noticed a lack of active planning and of
communication within the project and between it's teams. There were around
forty developers listed on the website. But ... as we were all volunteers and
because our levels of commitment differed, I accepted it as a symptom of the
open-source, volunteer environment and didn't think of it as actual poor
management or incompetence on any single person's part.
Largely due to Eric Sandall's open leadership, through his ability to support
new ideas, his ability adapt and thanks to his perceptive understanding of
events and problems, many things were discussed and decided by the grimoire
team during my time as a developer. Unfortunately, SMGL as a project did not
rapidly or smoothly progress towards any of it's roadmapped goals. There was
lack of action in many events (like the squabbling after the devel grimoire
was stopped from being packaged). We weren't allowed to discuss and update
the project roadmap (I was in fact told on several occasions that the roadmap
was in the wiki, go read it and pull my finger out. At the time I didn't take
it personally, I was sure the project lead couldn't be talking directly to me
because I already did between 20 and 40 hours a week updating spells,
rewriting spells, submitting new spells and fixing bugs from bugzilla, so he
must be talking in general to the smgl user base and developer community).
Attempts to discuss the ISO/Cauldron teams problem meeting deadlines, to plan
or achieve goals and the teams lack of interest in working on filed bugs were
unwelcome, apparently developers had no right to discuss or plan goals of a
team that showcased their work. After each developer meeting I attended (at 4
or 5am, all but one in the time I was a developer) I was left with a feeling
of unease about it's outcome.
When the project lead, Eric Schabell, decided he would take over the position
of Sorcery Lead as well as Project lead for the period upto the (now
postponed) 1.0 release I was concerned. I hadn't actually seen evidence that
he had the advanced skills needed to hack sorcery, but he himself told me he
had personally made several of the previous sorcery releases., so I remained
quiet. I also wasn't sure it was a good thing for the project lead (by
definition an impartial facilitator, not an active developer) to also be a
team leader, but one more time I gave him the benefit of the doubt and
remained quiet. Like many developers I also knew of many annoying and
inconvenient bugs in sorcery that would need fixing before we could honestly
announce we had a stable 1.0 release, so I was actually happy that someone
would begin active work on sorcery.
Knowing the hours spent working on the grimoire were under-utilised and were
poorly showcased by an ISO that was almost never released on time and was
all too slowly adding the features requested by the community, several weeks
before Christmas, the third time I contacted Schabell about it, he again told
me to take over if I wanted more work done, so I finally said yes. I knew
there was no point in my continuing to develop the stability of the grimoire
when the ISO was so far behind it. When I asked him how he wanted to announce
it and how I got access to the things I needed he told me I would not be
taking over the ISO until after the 1.0 release in three months time !
(Luckily, in another 6 weeks time the ISO lead stood down early and I had the
chance to make the progress that you all saw.)
At around the same time (before Christmas), I received some notification from
bugzilla that the annoying bugs I had filed against sorcery had been moved to
feature request category for after the 1.0 release. I didn't think much of it
at that time, I simply thought there must have been more important bugs and
that a plan for the release and list of priorities had been drawn up.
Christmas came and several of my machines stopped booting after an update and
I had to take the prometheus cluster I was running offline and spend some
considerable time before finding that devel sorcery had stopped installing
any initscripts. I decided it was my own fault for running devel sorcery on
those machines, then corrected the problems, and changed the cluster
machines to run stable sorcery. Six weeks and nine releases of "stable"
sorcery later, after filing many bugs, after being told they weren't bugs
they were feature requests, after discovering that Schabell was moving every
sorcery bug into feature request regardless of it's severity, after being
told on several occasions "I am Sorcery Lead, I will do as I see fit" (only a
valid argument at kindergarten, as far as I know), after having another three
machines stop booting due to poor,untested code in the "stable" release of
sorcery, I finally managed to get all four of my cluster machines installed,
updated and rebuilt, then to actually boot without any manual intervention
on behalf of an unstable "stable" sorcery. I could once again run the
prometheus testing program without it reporting copious errors related to
sorcery problems. Unfortunately it was time to turn off the cluster and start
on the ISO work.
Being interested in supporting everyone's requests and showcasing all
developers efforts, I looked in bugzilla at the entire history of ISO bugs
and also looked through old meeting logs, forum postings and news site polls.
Very soon, I found that of everything ever requested in developer meeting, by
poll or by bug, only a fraction had been roadmapped or achieved. I looked at
what had actually been requested for the 1.0 release , looked at the problems
with sorcery, at the time left before the major release and found that
realistically nothing significant could be finished in that time frame. Any
release would be less than requested, less than indicative of the majority of
developers work and would be immediately followed by bug release after bug
release.
So I made a plan of what should be done for the ISO, published that plan and
"told" everyone I would not release 1.0 until that plan had been achieved. A
vote was called on the mail list and shortly after a news site poll was
started. Both the mail list vote and the sneaky poll agreed that the
requested features and bugs for the ISO and sorcery should take priority over
some arbitrary deadline set over a year beforehand. If the users and
developers had wanted a 1.0 release before any thing else I would have stood
down and immediately retired from SMGL.
The leadership changed, new Project Lead and Grimoire Leads started, but until
Andrew Stitt volunteered to be Sorcery Lead, Schabell stayed on as sorcery
lead. I announced my intention to leave SMGL development after achieving my
ISO roadmap and then put my head down and started working.
I had "founded" the NGA in june (or july?) of the previous year and I had
changed my perforce review listing to cover the entire project, as I didn't
want to be out of touch with the other developers and especially wanted to
make sure my work didn't conflict with anyone else's. We regularly used that
notification list for impormptu discussion of issues and as it took less than
5 minutes a day to scan the incoming submissions, it was an invaluable aid in
my work.
You can imagine my amazement then, when I found just before the leadership
changed, that the project lead, the person who was (in title) the supervisor
of all developers, the person who was in charge of the project roadmap, the
person who had been elected to represent all developers goals and wishes, the
person whose opinions and actions I never understood, the person who tried to
railroad a substandard 1.0 release against the wishes of the development
team, the person whose unskilled attempts to hack sorcery had introduced
months of instability, Eric Schabell, had neglected to subscribe to
notifications for any of the devel,test and stable grimoire repository . For
the previous year, he had no idea that of the 40 listed developers less than
20 were active and less than fifteen were actually responsible for almost
30,000 changes to the grimoires. He was in no position to lead anything,
having no knowledge of the majority of work done in the project. It went a
long way towards explaining all the problems encountered in the previous
year. The lack of open communication, lack of forethought, foresight and
planning, lack of follow up on events, ideas and suggestions. It explained my
dissatisfaction after developer meetings because they lacked forward planning
and were instead a method of reporting to an incompetent project lead. It
explained why he wasn't supporting the active developers, he simply didn't
know who they were. It was the root of so many of the problems so I passed
the information on to the other senior developers, but wishing he might
retain some small dignity for any good work he had done in the past I wrongly
decided not to alert the entire project and the user base of all these
problems. (you will notice in all my attempts to discuss problems around that
time I never (almost, I am human after all :) ) actually referred to anyone
by name, it should have been a discussion of events and problems not a full
blown witch hunt.)
At this stage any good fairytale has one of two endings
1) the villain is ousted from power or killed outright and the townsfolk live
happily ever after, having learned a valuable lesson and never again falling
under the influences of evil again.
2) the villian finds redemption, admits all his failings, learns his lesson,
hands over power to someone else and either banishes himself or becomes the
town streetsweeper in an effort to make amends for the errors of his previous
life.
SMGL has a new and different ending.
Ignore the gross incompetence, ignore the attrocious attitude, ignore the
lack of honesty, ignore the lack of respect for other peoples work, ignore
the lack of common sense, ignore the inability to accept blame or be
accountable for mistakes, ignore the poor communication skills, ignore the
personal abuse (cowardly) emailed to me (and others ?) in response to factual
but critical bugzilla posts, ignore the fact that the trust of every
developer
was abused and say
"oh well, he isn't in charge now, what damage can he do ?".
Once is an honest mistake, twice is an avoidable error, three times would be
incompetence and everything after that is gross incompetence and
irresponsibility.
I left SMGL not because of any single problem I had with Schabell, but because
I was unhappy that Schabell was not held accountable for anything, there had
been no public recognition of all the problems that had occurred and I
couldn't continue volunteering to support a project that wouldn't openly deal
with it's problems. I left SMGL knowing I had contributed a lot in my time and
thinking that those contributions would be continued by a team that had
learned a valuable lesson, even though they were reluctant to openly disclose
it.
I was contacted recently and notified he had finally quit the pretence of
being a sorcery developer because he couldn't fulfill the stricter
competency requirements of the new team lead. I was surprised though that
the cowardice displayed in posting the bug link to the mail list, and the
irrational comments in the bug itself had not been officially responded to and
was also worried that the gross incompetence, bad attitude and lack of
common sense had now been welcomed into the grimoire team and of course, he
still wasn't receiving perforce notices for any other developers. If you don't
learn from the past, the same mistakes will be made in the future.
And it is for this reason that I want my name and email removed from all files
in SMGL, I no longer want my good name associated with the timebomb that is
counting down in SMGL.
Hamish Greig
Games Guru
Nomadic Guru's Association founding member
Assistant Grimoire Lead
ISO/Cauldron Lead
SMGL developer no longer.
- [SM-Discuss] Remove details from SMGL ( An SMGL fairytale ), Hamish Greig, 07/02/2004
Archive powered by MHonArc 2.6.24.