RE: [Simulavr-devel] Is this project still alive?

[Eric Weddington]

> -----Original Message-----
> From: William Rivet [mailto:address@hidden
> Sent: Tuesday, August 07, 2007 9:11 PM
> To: Eric Weddington
> Cc: 'Michael N. Moran'; 'Klaus Rudolph'; address@hidden
> Subject: RE: [Simulavr-devel] Is this project still alive?
>
> I think the reluctance here is that we want to be sure we are
> pulling in
> the same direction.
>
> I'm not opposed to adding contributors, but I would like to set some
> goals and ensure we work toward them.
>
> Through various exchanges I think we have the following modest goals:
>
> 1.) Simulavrxx must be able to run the GCC toolchain test suite.
> 2.) Simulavrxx should build under Cygwin
> 3.) There should be better documentation on how to write simulations
> 4.) We should have a better way to accept patches to the project.
>
> I'm not sure how to proceed in terms of Eric and potentially Michael
> joining. From my perspective however, I want to get the build
> system in
> control first. I'm about half way through that right now. I have a few
> hours over the next few days for that effort.
>
> I'll consider the build aspect "complete" when it works for me in
> Gentoo, RHEL 4.0, Debian (Sarge I think?) and under Cygwin.
>
> If Eric or Michael would like to propose what they will contribute I
> don't see why we should not add them. I think I am a bottleneck at the
> moment however with the build system not working right.

Overall, I agree with the above.

I can help with testing for building under Cygwin, and running the GCC
regression test suite. I have a personal goal of wanting to port it to build
under MinGW, but that is very much a separate, more complicated issue. For
now, having it build under Cygwin is sufficient. I can also help somewhat in
the grunt work of actually committing patches that are written by other
volunteers, but I'll be the first to admit that I'm not familiar with the
code base, so actually *reviewing* the patches at this point would take more
time.

I would like to bring up another, minor issue, but I think important in
promoting this project: the name. I think SimulAVR is a fine name, but I
have problems with SimulAVRXX. IMHO, the programming language that the
project is written in is secondary to its purpose. I know the history and
reasoning for the current situation. But I would hope that once the
simulavrxx code base does everything that the simulavr code base does, then
the name should be reverted back to "SimulAVR", set a reasonable new version
number (even "1.0" would be sufficient), and move forward. Having the two
"X"s in the name looks strange, it certainly doesn't help when speaking the
name, and one has to constantly explain what it represents, and in the end,
it doesn't really matter to the end-users.

Thanks,
Eric Weddington