> -----Original Message-----
> From: Johan Corveleyn [mailto:johan.corveleyn_at_uz.kuleuven.ac.be]
> Sent: Monday, June 15, 2009 12:58 PM
> To: 'Nathan Nobbe'; David Weintraub
> Cc: users_at_subversion.tigris.org
> Subject: RE: Distributed Subversion
>
[snip]
>
> For our situation (commercial env, local team with devs talking to
each
> other frequently, ...), I can currently only see two possible
> advantages of DVCS over SVN:
>
> 1) Ability to keep dev-local branches, work off-line, commit locally,
I like the idea of shelving and offline commits... it would appear to be
a great feature that everyone could use - ie I'd use it to commit
changes locally, then make 'experimental' changes and could revert back
to a working version. I could also commit locally several times if I was
working on several code changes and then have them all appear in the
repo as multiple commits rather than 1 big one.
However... the issue with this whole approach is that, as a corporate
developer, I do not want to lose changes at all. If I'm working off-site
then I would not be happy with all changes being made to the laptop
only. I would definitely want to get the developers hard work in my
central, backed-up repository. This applies whether using git or svn.
Having your laptop stolen is a real danger, losing a week of two of work
is not acceptable simply because you always committed locally thinking
it was ok.
I agree performance needs to be improved - but then, when is that never
the case :) Maybe the proposed server-side property storage would help
out here a lot, as would the local WC changes. Also, fixing the
temporary file and memory leak will help a lot. I'm sure performance
will keep improving and will be noticeable very soon.
------------------------------------------------------
http://subversion.tigris.org/ds/viewMessage.do?dsForumId=1065&dsMessageId=2362146
To unsubscribe from this discussion, e-mail: [users-unsubscribe_at_subversion.tigris.org].
Received on 2009-06-15 14:12:27 CEST