Open MPI logo

Open MPI Development Mailing List Archives

  |   Home   |   Support   |   FAQ   |   all Development mailing list

Subject: Re: [OMPI devel] Switching away from SVN?
From: Jeff Squyres (jsquyres_at_[hidden])
Date: 2008-03-24 16:22:43


SVN v1.5 has completely revamped how they do merging -- it is supposed
to be much mo' betta (although I have not tried it myself). The whole
business of needing to track what you have already merged on a long-
lived branch has been eliminated; in this way, SVN v1.5 will be much
more like git/hg/bzr.

However, a key difference over SVN will still be that git/hg/bzr are
*distributed* VCS's, where SVN is still *centralized*. Several of us
have been using distributed workarounds to SVN for a while (e.g.,
using git, hg, or svk as a gateway to the centralized SVN server);
having a true distributed VCS could be quite nice. There are other
advantages, too, such as publishing arbitrary temporary or research
branches (perhaps even on your own server; you don't need to sign the
OMPI 3rd party contrib agreement to get commit access to our server)
that aren't ready for prime-time, etc.

On Mar 24, 2008, at 4:13 PM, Edgar Gabriel wrote:
> generally, I have no objections to switch away from svn to another
> method, assuming that we do not give up much of the comfort that we
> have
> today, as George mentioned. One question related to that however is,
> whether upcoming svn releases would solve some of the issues which we
> have today with svn, especially with long-living branches?
>
> Edgar
>
> George Bosilca wrote:
>> After playing with hg and git for few days, I tend to agree with the
>> emacs guys. It looks to me that any of them will do the job (as did
>> svn). I don't really care which one will be selected by the
>> community as
>> long as we:
>> 1. Don't spend months in deciding which one to choose.
>> 2. Don't loose the nice integration o svn with our TRAC.
>> Independent on
>> how good/fast the dVCS is, the way svn integrate with trac is a real
>> time saver. Tracking bugs, linking to revisions and to the wiki are
>> really important features to me, and I think that whatever our
>> decision
>> will be we should not lose this.
>>
>> george.
>>
>> On Mar 24, 2008, at 2:12 PM, Roland Dreier wrote:
>>> LWN.net has an interesting article about how Emacs chose a new
>>> version
>>> control system: <http://lwn.net/Articles/272011/>
>>>
>>> They were back in the CVS stone ages, but their main contenders were
>>> the same big three of distributed VCSs: git, hg and bzr. The
>>> article
>>> pulls out a couple of very good quotes from their discussion. The
>>> one
>>> that caught my eye was from Richard Stallman:
>>>
>>> We already know the most important thing about what we will find
>>> from
>>> a careful study of git, mercurial and Bzr. We will find that
>>> each has
>>> its advantages and disadvantages -- but none of them conclusive.
>>> Each
>>> will be preferred by some people, but any one of them would work
>>> out
>>> well enough.
>>>
>>> - R.
>>> _______________________________________________
>>> devel mailing list
>>> devel_at_[hidden]
>>> http://www.open-mpi.org/mailman/listinfo.cgi/devel
>>
>>
>> ------------------------------------------------------------------------
>>
>> _______________________________________________
>> devel mailing list
>> devel_at_[hidden]
>> http://www.open-mpi.org/mailman/listinfo.cgi/devel
>
> --
> Edgar Gabriel
> Assistant Professor
> Parallel Software Technologies Lab http://pstl.cs.uh.edu
> Department of Computer Science University of Houston
> Philip G. Hoffman Hall, Room 524 Houston, TX-77204, USA
> Tel: +1 (713) 743-3857 Fax: +1 (713) 743-3335
> _______________________________________________
> devel mailing list
> devel_at_[hidden]
> http://www.open-mpi.org/mailman/listinfo.cgi/devel

-- 
Jeff Squyres
Cisco Systems