Feedback from evaluation in a corporate environment

[Stephen J. Turnbull]

Gotta run, but I'm very pleased by how nicely you were able to support
my guesses.  Obviously there's a well-designed, capable DAG handling
engine in there ... with a little study of your post, probably I can
give up my git "crutch" for explaining bzr. :-)

Two easy comments:

Robert Collins writes:

 > > when you need coherency in names of heads).  git goes this one better
 > > with reflogs
 > 
 > I don't see how reflogs really impact things here: yes it gives you a
 > way to address the two heads, but you still need to resolve things
 > eventually.

OK, git goes 0.5 better. :-)

I just like having a well-known name for the other head, rather than
having to dig it out with a tool I use once in a blue moon (I've
already forgotten your explanation of how to dig out the other head's
ID in bzr that you gave me once).  This allows me to postpone dealing
with the implicit branch indefinitely.

 > Doable but not a particular improvement IMO - just piping straight
 > through fastimport should behave very well for git itself, and tolerably
 > for bzr and other systems.

Not on the XEmacs CVS repository.  Gawd, what a mess. ;-)  As Uri said,
it's very useful to save the fastimport dump, because producing it is
by far the largest time sink.  But it's small enough (fastimport blob
file is 4GB, revision file is < 100MB) that I just keep it in raw form.
Uri's is about 2 orders of magnitude bigger.  Wow!