uncommitted changes in rsync and branch of bzr.dev

[Jamie Wilkinson]

This one time, at band camp, Wouter van Heyst wrote:
>On Mon, Mar 06, 2006 at 03:24:28PM +1100, Jamie Wilkinson wrote:
>> >Try a bzr revert. As far as I known, pushing switched from rsync to
>> >sftp, and now the working tree does not get updated anymore. It would be
>> >clearer if that just got deleted. (That, or working tree pushing with
>> >sftp implemented).
>> 
>> I wouldn't have expected the second branch:
>> 
>>  bzr branch bzr.dev bzr.dev.1
>>  cd bzr.dev.1
>>  bzr status
>> 
>> to also show the same output: does a local branch just copy the working
>> tree instead of doing the equivalent of bzr revert?
>
>Currently, yes. There has been some mild disagreement with this, see the
>'A couple of bugs...' thread for instance. 

Ok.  I register here my dislike of that behaviour, because I'm expecting to
branch from the latest committed revision and not have random uncommitted
crap from my working tree appear in the new branch (because almost certainly
I'm branching so the uncommitted code doesn't interfere with another line of
development).