Weird failure of a merge to local branch

[Stephen J. Turnbull]

Marco Pantaleoni writes:
 > I guess Mark intended that sync would be bidirectional: when it is
 > completed, both ends are updated with the changes from the other.

That would be a reasonable extension if selected by an option (eg,
"--bidirectional"), as an alias for

    { bzr merge \
      && bzr commit -m "Sync to remote $branch." \w
      && bzr sync --outgoing }

I think it would be a rather surprising default, though.

I didn't mention it because (a) it's not directly related to push and
pull as currently defined, and (b) probably most people would think
"bzr bind" is a more convenient way to accomplish the same thing.