[OpenAFS-devel] A few questions about the current Linux implementation of the AFS client

[Derek Atkins]

Derrick, 

Have you considered changing over to using bitkeeper instead of CVS?
Then individual developers could have their own local repositories and
make local deltas, and then they could set up their changesets to
openafs-gatekeepers.  When you apply a changeset users could pull it
down into their own repositories.

It would give developers the chance to make local "commits" for change
history during development without requiring access to the global
repository.

-derek

Derrick J Brashear <shadow@dementia.org> writes:

> To be clear, I'm speaking for me only and this doesn't represent the
> opinion of the Council of Elders, the OpenAFS Gatekeepers, Carnegie Mellon
> University, or anyone but me. 
> 
> On Fri, 18 Jan 2002, Matt Peterson wrote:
> 
> > The purpose of this email is not to point out the problems of OpenAFS, but 
> > rather to get some information about the plans developers have to improve the 
> > implementation -- specifically the Linux implementation.  I would also be 
> > interested in some statement about how willing OpenAFS is to the possiblity 
> > of a development branch of CVS that would allow more agressive development 
> > that what is currently taking place.
> 
> No patches have been refused, literally everything passed to us has
> either been taken or very specific comments about how to make it suitable
> have been offered. The only thing stopping aggressive development is lack
> of developers.
> 
> []
> 
> > Is there interest in maintaining this type of development on OpenAFS, or is 
> > OpenAFS in "maintenance mode" only? If there is interest, how does one
> > go about getting involved?
> 
> Implement stuff. Send it to openafs-bugs@openafs.org. I assume you've not
> been following development effort thus far?
> 
> Truthfully I favor a mechanism for developers to have direct access to the
> source tree, but CVS is really crappy in that it offers no way for you to
> have a private branch that can't taint the mainline, and without that it
> means we have to know and trust every developer who we give access to the
> tree. 
> 
> 
> 
> _______________________________________________
> OpenAFS-devel mailing list
> OpenAFS-devel@openafs.org
> https://lists.openafs.org/mailman/listinfo/openafs-devel

-- 
       Derek Atkins, SB '93 MIT EE, SM '95 MIT Media Laboratory
       Member, MIT Student Information Processing Board  (SIPB)
       URL: http://web.mit.edu/warlord/    PP-ASEL-IA     N1NWH
       warlord@MIT.EDU                        PGP key available