Skip to Content.
Sympa Menu

sm-discuss - Re: [SM-Discuss] Questions to the nominees

sm-discuss AT lists.ibiblio.org

Subject: Public SourceMage Discussion List

List archive

Chronological Thread  
  • From: Arwed von Merkatz <v.merkatz AT gmx.net>
  • To: sm-discuss <sm-discuss AT lists.ibiblio.org>
  • Subject: Re: [SM-Discuss] Questions to the nominees
  • Date: Thu, 8 Mar 2007 19:58:20 +0100

On Sun, Mar 04, 2007 at 08:29:55PM +0100, Andraž 'ruskie' Levstik wrote:
> My few questions for the grimoire nominees.
>
> a) New grimoire features like keywords and other sich things.
> What do you plan on doing about them and any other new
> features?

Most of the time, new features available to the grimoire get picked up
by most through word of mouth and the commit list. Obviously this didn't
work for keywords, though I think the main reason for this is that
there's no actual use for them so far. scribe can create an index of
them, but no one does anything with the index.
So what needs to be done is to get some tools use keywords to make them
actually do anything. A first step would be to have a gaze search
-keyword, listing all spells for a keyword.

> b) Try to unify as many BUILD/INSTALL files and systems into
> a single default_build(something that the gnustep sections do)
> or the new spell file inheritance scheme.
> Your opinion on this?

I think this is exactly what spell file inheritance is here to solve.
Essentially it's just replacing the method of overriding default_build
with a generic BUILD file for a section.

> c) Maintaining of critical spells like glibc, gcc.
> How would new versions be handled? Will you do the version bump?
> Delegate it to someone else?

I've handled many of the gcc and glibc updates personally, but that's
mostly because I like tinkering with complicated spells. For the future,
I think having seperate devel-$FEATURE branches in git will make this
easier. Anyone can update them in a devel branch, then we need to define
some testing process that allows them to move to test grimoire, like
"X lead devs signed it off to go into test".

> d) What's te ammount of time you could give to supporting gurus
> with issues, questions, etc? Availability for gurus?

Not as much as I'd like to, but definitely more than I've been able to
for the last two months. I should have appointed a really official
assistant that could replace me during such times, and I will do that in
case I continue as grimoire lead.
To be more specific, I should be around at least for a couple of hours
each day on IRC, provided that people are patient when asking questions
(I might be doing unimportant stuff like cooking or eating ;), and
at least half a day on weekend, usually more.
Email should always reach me, though I'm still working out something for
the time where my home pc isn't running, as that home pc is essential
due to spam filtering.

> e) Long overdue project on the wiki:
> http://wiki.sourcemage.org/LicenseInventory
> Would be nice to acctually start enforcing this more and
> try to cleanup everything.
> Also an that same note: spellifing licenses(already have a
> base for this).
> Your opinion on this? Sholud we try to clean out badly licensed
> spells as a project or leave it in a fix it as you find it state?

It is long overdue, but it's really not a priority for me. If someone
wants to make it their pet project, I certainly see a use in this, but I
think we have more pressing matters on hand.

> f) Opinion on supporting z-rejected as a developer grimoire(getting
> that Free mark from FSF/GNU is acctually starting to count for a
> lot of people I talked with online when I recommended them source
> mage. Many didn't want to use it because of that.)?

Personally I like the goal of a totally Free distro, although we
actually fullfill *my* goal in that area. I think some of the
restrictions by the FSF for getting their "Free Distro" badge are just
not that useful. If a developer wants to work on both our main grimoire
and the nvidia-driver spell, should we really deny that option?
Overall I don't think there's that many people who don't use SMGL
because we have the *option* of making installation of proprietary
drivers and some restrictively licensed stuff easy.
That said, I think z-rejected needs a split into a binary grimoire and a
rejected grimoire. There's really no reason to push them into the same
except the notion of limited support for both.

--
Arwed v. Merkatz Source Mage GNU/Linux developer
http://www.sourcemage.org




Archive powered by MHonArc 2.6.24.

Top of Page