Plans, please, not excuses [was: Primer ... ?]

[Stephen J. Turnbull]

Ian Clatworthy writes:

 > 1. Bazaar started later than and hence behind the other tools.

That's very arguable.  Not only was the Bazaar-NG project announced
about 3 months before Linus and Matt Mackay started work basically
simultaneously, as Aaron points out, but the crew that defected from
Arch to Bazaar certainly had one heck of a lot more background in the
philosophy and practice of revision control than Linus.  (I don't know
about Mackay.)  They also took with them proven algorithms and
software, plus a lot of solid thinking about UI, workflow, and best
practices, while Linus invented his as he went along (at least that's
the legend).  If anybody in revision control was prepared to "hit the
ground running" on a new VCS project, it was the Collins-Bentley-et-al
team.  And from what I hear, Canonical beefed up the team with help
from guys who are just as clueful and smart and maybe smarter.

In any case the fact is that git and hg had better performance from
the very start than bzr does now, despite the public claims of recent
improvements.  What I conclude is that bzr was never well-designed for
performance at scale.  (Obviously, the effort both in design and
implementation went to UI and workflow stuff, it shows.)

Now, when Robert Collins says that the "pack" repo format is a good
foundation for performance and scaling work, that's good enough for
me.  But maybe it would be a good idea (from a PR standpoint) for
somebody to explain why that is so and what changes need to be made to
what bzr commands.  Otherwise, I don't see why anybody should believe
it until they see it.  Hopefully that would wet down some of the
flames on emacs-devel....