[svn.haxx.se] · SVN Dev · SVN Users · SVN Org · TSVN Dev · TSVN Users · Subclipse Dev · Subclipse Users · this month's index

Re: major Subversion progress

From: Jeremy Blosser <jblosser-svn_at_firinn.org>
Date: 2001-07-13 22:01:27 CEST

James Duncan Davidson [duncan@x180.net] wrote:
> On Friday, July 13, 2001, at 08:44 AM, Karl Fogel wrote:
> > Well, your conjecture is right -- at least, I can't see how it would
> > be a good use of any Subversion developer's time to work on
> > svn2cvs. :-)
> >
> > Of course svn2cvs would be useful for some people; I guess either a
> > volunteer will need to step forward, or (perhaps) it will be important
> > enough to some organization that they will fund its development.
>
> If time cycles are going to get spent on this, though, I'd highly
> recommend layering a pserver impl on top of svn (if at all possible with
> at least some subset of semantics). That way, orgs that have developers
> comfy with their CVS proceedures could at least do checkin/out/diff
> against files.

<lurk>

As long as people are dreaming, I'll throw in that this would be incredibly
useful to shops like mine, if we assume that it will take some time for all
the svn GUI clients on various platforms to come into their own. The users
are very happy with the cvs user feature set and their individual clients
on their individual platforms... we're using 2 different MSWin clients, 2
(3?) different Mac clients, plus of course the command line client on Win,
Mac, and Linux. As an admin I really really really want to switch to svn
ASAP because I know it'll make my life a whole lot easier, but I'll be held
back implementing it in production until I can provide working, minimally
buggy client GUIs for the users. Having some kind of CVS emulation layer
implemented through a pserver would speed this up a lot, though admittedly
one of the main reasons I want to use svn is to have something better than
pserver doing the authentication (something that can be synched to an LDAP
service, preferably). But at least this would let me switch the backend
sooner and wait on replacing the clients til I had options.

I know this isn't likely to be something the developers want to do, and I
don't expect it will happen... but maybe it'd be useful to people. If I
had funds or developers to throw at it I would.

And I know someone is working on a Java client, but I doubt that'll be a
real option for me. My users use their various operating systems and
environments because they like the individual approaches to workstyles, and
having a client that looks and acts like a Windows app on the Mac or vice
versa or just like it's own thing on both isn't something they're going to
find appealing or usable. I'm sure there's a large group that would find
it useful for their own reasons, and it probably even makes some sense as a
prototype GUI svn client, but it probably just won't get much use here.

</lurk>

-- 
Jeremy Blosser
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Sat Oct 21 14:36:33 2006

This is an archived mail posted to the Subversion Dev mailing list.