Re: [avr-libc-dev] Adding (some) Procyon AVRlibfunctionalitytoavr-libc

[Ruddick Lawrence]

Wow, step out for lunch and miss a lot!

The confusion I was talking about with LibAVR was not with the existing
Procyon library, but with avr-libc. Of course that only becomes a problem if
they are indeed separate projects, which brings me to my second point...

I agree that associating this new project with the avr-libc brand will be
beneficial for it in terms of attracting developers/establishing legitimacy,
and in turn, as it becomes more useful, it will contribute to the "prestige"
of avr-libc. However, I do think it should be separated because of the
different abstraction levels: if you want to develop in C on an AVR, the
only sensible thing is to use avr-libc. However, the stuff we're talking
about creating is not that "necessary". I could not imagine coding without
the PORTC definition, but I could live with having to "manually" send bytes
via UART instead of using a uart_sendpacket function. I support housing it
in the avr-libc project under a separate CVS subtree (personal pet peeve: is
it possible to use SVN instead? I hear baaaaad things about CVS). Don't
forget, this also does not have to be permanent, but is a good way to start.
In the future it can be moved to its own project if we find it necessary.

As for function naming conventions, etc, once we're settled in with a
mailing list and all, I agree that it's appropriate to come up with style
guidelines for the library, which we'll submit back to this list for review
(does avr-libc have such a style guide already?).

Another possible name: avr-highlib (it's a high level library....)

Ruddick

2009/9/18 Weddington, Eric <address@hidden>

>
>
> > -----Original Message-----
> > From:
> > address@hidden
> > [mailto:avr-libc-dev-bounces+eric.weddington<avr-libc-dev-bounces%2Beric.weddington>
> address@hidden
> > org] On Behalf Of Frédéric Nadeau
> > Sent: Friday, September 18, 2009 2:00 PM
> > To: address@hidden
> > Subject: Re: [avr-libc-dev] Adding (some) Procyon
> > AVRlibfunctionalitytoavr-libc
> >
> > On Fri, Sep 18, 2009 at 3:40 PM, Weddington, Eric
> > <address@hidden> wrote:
> >
> > > Certainly, a hosting location needs to be picked. If
> > Savannah is too difficult, it's certainly easy to get a
> > project started on SourceForge, and of course it's a popular
> > hosting site. I suppose Google would be ok, though it's not
> > as well known. In looking at Frederic's project, I don't see
> > a mailing list capability on Google and I think that would be
> > essential. I would volunteer to setup a project on SourceForge.
> > >
> > > Eric Weddington
> > >
> > We could certainly do a Google group for the mailing list and link it
> > to the Google code.
> >
> > (Sorry for the direct email)
>
> Oh, I didn't notice. No, I don't like dealing with Google groups. We just
> need a simple mailing list.
>
> > As a side note on the name. For the AVR32 they named their set of
> > driver/livrary "framework".  avr-framework is yet another idea.
>
> No, please, definitely not. I really don't want to get it confused with any
> kind of "framework" product that exists on Atmel's website.
>
>
> _______________________________________________
> AVR-libc-dev mailing list
> address@hidden
> http://lists.nongnu.org/mailman/listinfo/avr-libc-dev
>