Public SourceMage Discussion List

[George Sherwood <pilot AT beernabeer.com>]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Thu, 8 Mar 2007 14:36:42 +0400
George Sherwood <pilot AT beernabeer.com> wrote:

> 
> I accept the nomination and will also try to get an email out over the
> next few days.

First just a little bit about my involvement with Sourcemage.  Looking
at the activity file on my laptop, it goes back to January 2002 and
Sorcerer.  So I have been around this group a long time, but that
doesn't really say anything about why you should vote for me to be
Grimoire Lead.

Since becoming a developer a few  years ago, I have tried to be as
active as possible and don't think you would have to be around very
long to realize that I commit a large number of updates to the
grimoire.  But this alone isn't a good reason to vote for me either.  I
don't just sit around and talk hoping that Sourcemage will get better,
I have invested a large amount of my time to make sure it does.  I work
very hard on routine updates that there are only a small number of
people doing.  Not sure how many people recall, but during the
transition from BUILD_API=1 to BUILD_API=2, I personally fixed an
extremely large number of spells allowing progress to be made in that
area.  I can also point to complex spells such as firefox and php that
I have seen growing stale and I have worked hard to make them work for
our users.

I will now take a shot at some of the questions from our PL:

> - Approach to release cycle.  The project has basically started doing
> these on our own, but I expect the control to go back to the Grimoire
> Lead at
>some point, even if that's just to make sure the current process
>continues.  There are open policy questions around the current process
>  that need to be addresses as well, including target release cycle
>  timeline, policy and timeline for adding new siupported spells,
> policy for integrations to stable between releases, etc.

I was one of the people that started the push to getting stable moving
again.  We were stuck on a process that didn't work and we allowed
stable to become unstable mostly due to its age.  I still believe we
need to find a proper release timeline, but in my opinion somewhere in
the 4-6 week range is a good target.  We do not have the developers to
support much quicker releases and we don't want to let it get too old
once again.  We need to continue to work on a good procedure for these
releases so it isn't completely dependent on a few people taking charge
and pushing the release.

> - Policy and proposal for making sure we do security updates in a
> timely manner.

The only time in my experience that security update process was working
well, was when there was a security lead that was dedicated.  We need
to find a person that is willing to spend the time to make these
updates happen.  I am not nearly as concerned with the security mailing
list as just ensuring that these updates get in the grimoire and are
integrated to stable in a timely manner.  All developers need to keep
their eyes open and ensure bugs are written for any security updates
they see even if they don't have time to do the update.

> - Plan for getting our documentation current relative to the grimoire.
Documentation has always been one of the weak points for this
distribution.  I certainly don't have any magic bullets for this
problem, but we will need a grimoire lead that will spend the time to
understand what needs to be updated and attempt to point our limited
resources at the problem.  This will not be easy since as we all know
most people do not like to work on the documentation. 

>- Your expected availability, and how you'll make plans to let things
>move
> forward in your absence.
In the time that I have been a developer I have made two moves one of
which was overseas and a new job.  Over this time frame there has been
little time that I wasn't available in IRC and making updates to the
grimoire.  I don't see this changing in any way if I am elected to
Grimoire Lead.  I think we need to take a hard look at the Assistant
Grimoire Lead position and ensure that person is empowered and willing
to act as Grimoire Lead anytime the Grimoire Lead is not available.

>- Possible bugzilla usage changes; this one is probably minor for this
>kind
>  of discussion but it's on my mind because as I'm filling in doing
>  triaging or whatever I'm finding it's difficult to follow things
> that are assigned all over the place, half the time to devs that
> aren't around anymore.  I think sorcery and cauldron have a better
> approach to this and would like the grimoire to consider adopting the
> way they do it.

I believe that our bugzilla at this time has become nearly unusable.
It is my opinion that all the bugs need to be scrubbed and closed
ruthlessly due to their age and in ability to be reproduced.  I don't
feel that bugzilla in its current state is useful to any one, be it a
developer or user.  Currently there are very few active section
maintainers.  For the most part it is not useful to have bugs assigned
to any one or any section.  I think all bugs should be sent to the
mailing list and let people fix the ones from the list that they fell
then can.  But most importantly the number of bugs needs to be reduced
to some level that is workable and this is going mostly likely include
closing old bugs in large numbers.

I have really enjoyed my time with working as a developer on Sourcemage
and feel that I could contribute greatly to the success of the the
distribution as Grimoire Lead.  I freely admit that my technical skills
are not as good as some of the other candidates, but I can promise that
I will give my time and will seek out the solutions to problems that I
don't fully understand.

George Sherwood
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (GNU/Linux)

iD8DBQFF7/iEkVJnfkgKg60RAv1OAKCGFIvfi0tnQ4FogfZ6R+bxR0grTwCfUq80
+Xto8cRI6llApeZUy73/YZw=
=zpeu
-----END PGP SIGNATURE-----