Layered repositories

Radoslaw Szkodzinski astralstorm at
Sat Sep 17 08:35:49 CDT 2005

On Sat, 17 Sep 2005 14:16:25 +0200
Erich Schubert <erich at> wrote:

> This exactly what other SCMs call a merge... and it sucks. Having to
> merge all the time is exactly what I want to avoid!
> Because it's inconvenient.

If there are no conflicts, the merge is automatic. There are smarter and less smart algorithms for it, of course.

You can't do without merging. Two people can always create incompatible changes and then you have to merge or discard one set of them.
Tell me how to avoid it entirely.

> Secondly, I do not want distributed repositories. I want all the
> configuration archived on one central machine, so I can easily compare
> machines, prepare changes for a rollout etc.

Compare repos? It's still possible in distributed system if other people make their repos available. The only change required is a central repository of all machines in your repo-network.
(This starts looking like a spider web)

You probably want a two-way system. This means that you can push to all
cloned repositories forcibly. However, what will you do if they did
some local changes? You'll just force a merge on them? What if there are any conflicts?
Solving them by hand by the server operator seems extremely 
IMO, the merge should always be done by the client.

GPG Key id:  0xD1F10BA2
Fingerprint: 96E2 304A B9C4 949A 10A0  9105 9543 0453 D1F1 0BA2

-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
Url :

More information about the Mercurial mailing list