Public SourceMage Discussion List

[Jeremy Blosser <jblosser-smgl AT firinn.org>]

On Mar 05, Jaka Kranjc [lynx AT mages.ath.cx] wrote:
> On Monday 05 of March 2007 20:53:22 Jeremy Blosser wrote:
> > I've been working on the project list... started a form of it for ideas 
> > for
> > the TX bugfix weekend.
> So it has mainly lightweight stuff on it?

No, if I've got these people in a room for a weekend, I don't plan to waste
it all on just lightweight stuff.  The current list for the wkd looks like:

project                   difficulty    priority
================================================
spell updates via quill   low           med-high
new spells via quill      low           low
doc cleanup               low           high

doc organizing            med           high
quickfix bugs             med           high

high profile bugs         med-high      high
new spells manually       low-high      low-high

integrate git and wiki    high          med
implement another init    high          med-high
iso refactoring           high          high

Where "low/med/high" are very relative to the group I've got coming.
Probably the guys who aren't at all SMGL devs now will play in the top half
of that list, the rest of us will play in the bottom half.

The full project list I need to get together to track all the stuff the
project has going on will of course be different, but there is some
overlap.

> > I guess right now I'm skeptical we have something that fits.  The grimoire
> > is basically day-to-day updating, as you noted.  Sorcery would fit but 
> > most
> > of the big things are already there (castfs would probably have been a
> > great project for this).  The cauldron work we need right now is
> > *simplifying*, not doing more advanced things. 
> Who said the project has to complicate the matter? Making things complex 
> and 
> obfuscated is easy, simplifying is the real challenge. And cauldron doesn't 
> have only one tip, there are other things beside the installer one could 
> work 
> on (the test case bug comes to mind). But I'll let the cauldron speak for 
> itself.
> 
> > I definitely thing the SoC is an interesting thing and would like to see 
> > us
> > in a place to take advantage of it someday but our current focus is
> > "stability and repeatability" and I'm not sure that fits what they are
> > looking for.
> I didn't find any text saying they are only after new feature additions (s. 
> stricto), it doesn't necessarily conflict with our "stability and 
> repeatability" goals.

If someone wants to come up with something that gets accepted I certainly
don't have a problem with it.

> http://wiki.sourcemage.org/InitMetadata has a good intro and even a spec 
> draft - I think it is good enough for a start. The students have 3 months 
> to 
> do their thing and I think this is enough to handle this as one project.

I would really prefer at this point that init refactoring work go more
along with the goals I spelled out broadly in a couple of mails last week.
I think the metadata is more than we really need right now and don't want
to see us fail to get anything done because we're fixated on that.  But if
someone really wants to take it all on as a single project that actually
gets completed, it would certainly be good to have, provided it was done
well.

Attachment: pgpco8J2cHl8c.pgp
Description: PGP signature