Re: Moving GDB sources to subversion?

[Bob Rossi <bob@brasko.net>]

On Sat, Oct 29, 2005 at 12:57:02PM +0200, Eli Zaretskii wrote:
> > From: Nick Roberts <nickrob@snap.net.nz>
> > Date: Sat, 29 Oct 2005 12:56:30 +1300
> > Cc: drow@false.org, brobecker@adacore.com, gdb@sources.redhat.com
> > 
> > We can assume that GCC developers have made a sound technical decision.
> 
> Yes, we can assume that.  But no one said that there's only one sound
> technical decision.  I'm sure there were downsides to that decision
> even in the context of the GCC project (as opposed to a general
> decision that _all_ GNU projects should adopt svn).  I'm sure that the
> decision they made was influenced, at least to some degree, by the
> persons who were involved in making the decision, and by their social
> dynamics.

Definatly. For instance, look at the Linux kernel. Linus has already
said 
 "PS. Don't bother telling me about subversion." at
  http://lwn.net/Articles/130681/

Also, The Subversion Development Team wrote a letter to tell people to
stop bothering Linus about subversion.
   http://subversion.tigris.org/subversion-linus.html

I personally don't see a large difference between the Linux kernel
development and the GCC developement stratagies. With that in mind, it's
hard for me to understand why GCC *is* a good choice for subversion and
Linux *is not*.

If it was up to me, I'd rather see GDB switch to a distributed RCS. I
usually have several tree's and it takes a long time to update them all
by hitting the internet each time. Until we get a distributed RCS,
subversion definatly seems like an improvement to CVS. I'd be happy to
see the change.

Bob Rossi