RE: Pair Programming with Subversion
>In our present phase we both have our own working copies of the same development branch.
>We can separately make commits, and we can immediately update our own versions
>with the other's changes using the built-in merge capabilities of SVN (update).
This seems to be the most reasonable way to use the current Subversion,
with its broken handling of file permissions in the workspace,
for pair programming.
It is somewhat unfortunate: in classic "in the same directory"
pair programming, you can mess with temporary files that
do not get checked in. Since our work often involves large
temporaries, checking them in would slow us down a lot.
Related: I tend to check in a lot already (about every half hour)
and already get complaints. What you describe would require
even more frequent checkins.
Received on Thu Apr 1 20:54:11 2004
This is an archived mail posted to the Subversion Users