It’s not working out so well so far. I guess I’m too used to Hg; I found its workflow agreeable.
Things I miss or annoy me or I haven’t found in evidence so far (feel free to comment if I’m not understanding something accurately):
- Named branches in the repo. Bazaar branches == Hg clones. This is a big deal to me, I have some projects with several branches that I switch between rapidly. In other words, this became part of my workflow.
- Hard linked cloning. Hg uses hard links and copy-on-write. When I clone a bzr repo, it’s a copy of everything. So when I have to make clones of the repo to get branches, they take just that much more space. I can imagine how much space large projects with many open branches take on shared server.
- Instant local http server a la
- Hg has RSS feed generators built in to its server and CGI program; seems to be an addon for bzr.
- Not to mention automatic tarball creation for downloads.
- Specifying the remote bzr command is an environment variable, not an
option. Inconvenient. Also not in the help for
bzr cloneas a consequence.
- Remote urls are not relative to the log in directory.
- Come up with a creative and original icon and logo!
My initial take? It’s like losing some of the coolest features of Mercurial (low overhead cloning, insta web server), and gaining some of the things I like least of Git (packs…I have to GC my repo?) and Subversion (branching has to use a new repo/storage path?).
I’m sure Bazaar will have wide support. I might use it to see how it works for offline development of projects in Subversion.
I think for rapid indie development and the web work I do, it’s going to remain Mercurial. It just fits my workflow better and gives me more power off the shelf.