[Axiom-developer] RE: Google Code repository

[Page, Bill]

On Tuesday, September 19, 2006 4:19 PM Ralf Hemmecke
> 
> >> Should I put Silver on the repository?
> >
> Bill Page wrote: 
> > The terminology is still rather confusing, but from my point
> > of view the axiom.build-improvements branch is a branch of
> > Silver.
> 
>
http://lists.nongnu.org/archive/html/axiom-developer/2006-09/msg00344.ht
ml
> 
> The terminology is actually not confusing.

What I meant was that in spite of the email referenced above, it
is still confusing to *me*. I find it awkward not to be able to
use Silver to refer to other branches in the SourceForge SVN archive.
I know someone suggested that we start to use Bronze etc. but I am
not sure that that would help.

> The point is that we should *not* call the whole SourceForge SVN
> repository Silver, but only the trunk. Otherwise Silver =
> Sourceforge/Axiom (with now 3 or 4 (different) branches of the
> axiom Sources).

Are you referring to the CVS on SourceForge (presumably identical
to the CVS on Savannah)? I thought these only had a single branch.

If you are talking about the arch (tla) archive on axiom-developer
then, you are right. I was just about to ask whether we should refer
to these a "Gold" or not. There are actually more like 10 of these
branchs and one of them is the axiom--windows--1 branch. That is
the only branch that builds on Windows now. Do we call it Gold?

> 
> > I would still prefer that we duplicate the most experiment
> > branch i.e. build-improvements, at Google.
> 
> Whatever we put there, we are probably only doing this to be 
> present at Google. I guess nobody will seriously use the Google
> repository until build-improvements is merged with Silver.
> 

I hope that is not the case.

Well, this is open source development which mean "release early
and releas often". The more accessible our experimental branchs
are, the more they will be tested and the better our software
will become. Of course people have to be warned about exactly what
they are downloading so that they have the proper expectations
about how it might/might not work. But I think we want to encourage
people to at least try the latest experimental versions. As an
Axiom developer that is the only version I am really interested
in.

Regards,
Bill Page.