Question on work-flow for big binary files

[Marcus Lindblom]

Paul R wrote:
> Marcus> I'm trying to come to terms with that. :) But cloning
> Marcus> repositories where the working copy is 2-20 gb (and the entire
> Marcus> thing, with changes and branches might be even bigger) is not
> Marcus> really cheap?
> 
> I don't think any VCS will suit such a need. VCS are designed to track
> changes by patchsets. Usually patches on binary files does not make sens
> as they are not mergeable and their size is unpredictable. Therefore you
> may want to hack-up an alternative solution for your big blobs, as
> I don't know of any VCS designed with binary versionning needs in mind.

SVN claims to have an efficient server-side diff algorirthm that also 
works on binary files. I think it's rather recent.

> I think rsnapshot or rdiff-backup, with a central-lock mechanism on
> push, will be a good start.

Yup.

> If you do so, I'd be very happy to know what you've done at the end, but
> this would probably be off-topic here, so please feel free to mail me if
> you want.

It might and it might not. VersionedBigFilesWithPruningExtension? :)

I do prefer public mail lists (for the greater good). Is there somewhere 
better to talk about these things, if it turns out to be something 
unrelated hg? (I'll probably ping you anyway. :)

Cheers,
/Marcus