Re: [school-discuss] projects

[LM <lmemsm@xxxxxxxxx>]

On 5/26/13, Michael Shigorin <mike@xxxxxxxxxxx> wrote:
> It happens; might help to publish your patchsets in some form
> so that distributions could pick them up, I've seen things
> flowing upstream more easily after testing in distros.

That's an idea.

> Well your ideas are interesting but it's a hard endeavor
> to achieve that single-handed; hence my suggestion to trade
> something in and look at the existing distributions as a base.
> My colleague who maintains musl package in ALT might have
> proposed examining Alpine Linux for that matter I guess...

Didn't think it was going to be easy and since I'm the only one
working on the project, I certainly expect it to be a while before it
gets to the point where I'd be comfortable distributing it in general
(to non-programmers).  I like things highly stable and well tested
before I distribute them.  I've definitely been looking at existing
distributions for a base system.  Not my goal to reinvent the entire
operating system.  Just trying to improve functionality of pieces I
don't feel are as efficient as they could be or don't work together as
well as they could and to find interesting and useful programs that
may or may not necessarily be mainstream in Open Source distributions.

Would be nice to work with some other developers, but so far, every
lightweight system I've checked, the goals are similar, but not quite
similar enough that they're interested in sharing resources/code.  On
the plus side, I get to follow my own design decisions and make things
work the way I want them to, instead of having to compromise with
group goals that may not be a good fit with what I need.  I'm not
really starting completely from scratch because I'm using Open Source
code someone else has already created as the starting point.  I also
get to look at the code of various lightweight and embedded systems
and if I see anything I like that's Open Source, I can incorporate it
with my own designs.  With Open Source, programs often get written or
projects get created because someone has a need.  I have no idea if
anyone else has a similar need out there and if they do that would be
great, but at the very least, this is something I need in order to get
useful work done on my computer using an Open Source operating system.
 Will keep plugging away at it.

> Guess NetBSD's pkgsrc is probably the most appropriate common
> ground for efforts like that these days; I can refer you to one
> of its developers I know (he also wrote NIH package manager).

Would be curious if projects like pkgsrc work with the build tools
(make, cmake, gnu autotools) that each project already uses or whether
they script their own build scripts/makefiles and uses only specific
tools (like possible bsd make) to build.  I've seen some projects use
the first method and others that use the second.  I've been using the
first method for my own project, but at this point I'm re-evaluating
some design decisions and wondering how well the second method works
in certain cases.

Thanks for the suggestions.

Sincerely,
Laura
http://www.distasis.com/cpp
###
To unsubscribe from the schoolforge-discuss mailing list:
Send an e-mail message to majordomo@xxxxxxxxxxxxxxx with no subject
and a body of "unsubscribe schoolforge-discuss"