Re: [avr-libc-dev] Re: Newlib [was: Re: Release?]

[Joerg Wunsch]

As Paul Schlie wrote:

> - maybe I'm confused, but avr-libc seems to be predominantly an
> assembler optimization of a subset of newlib (as exists for a few
> newlib targets) where although avr-libc contains a few avr specific
> utility functions, they're relatively minor, but complementary in
> nature, so don't sense problems with their being able to be
> maintained together, although possibly in a utility sub-directory,
> as they're basically structured now?

I don't know too much about newlib, I confess.

I had a quick glance on their website.  There's one thing I don't
really like with newlib: the multitude of copyrights.  Eric just did a
really good job on cleaning up all of our contributor's copyright
statements, so we can now ship it under a single simple copyright --
something that has been requested by our user base.

I don't really believe we'll just gain developers by moving it over to
newlib, albeit of course, we would benefit from all development work
in platform-independent areas (as we do already in GCC).  On the con
side, I fear we would lose our relatively autonomous status where we
can decide about all issues ourselves.  (The dependency on savannah's
infrastructure is not too hard, since the big cracker's attack the
overall site availability is quite good, IMHO a lot better than e.g.
for sourceforge.)  As for gaining developers, I see better chances by
trying to actively acquire them on the AVR-specific forums and mailing
lists, where all the people interested in the issue are dominant.

-- 
J"org Wunsch                                           Unix support engineer
address@hidden        http://www.interface-systems.de/~j/