Re: 'svn revert' vs. 'svn resolve'

Garrett Rooney <rooneg@electricjellyfish.net> writes:
> Really though, I think 'svn resolve' is a perfectly fine name, and
> we're spending entirely too much time and effort on a problem that
> will never be completely solved. Users will always type the wrong
> command from time to time, it just happens.

I'd like to reiterate disagreement with the above sentiment.

Sure, users will always make mistakes, but good software won't
encourage it. You speak as if problems can only be solved completely,
never partially. If that were true, we wouldn't be in the revision
control business in the first place :-).

The "risk" here is the likelihood of making the mistake, multiplied by
the severity of the consequences.

   1. The names are similar enough that the chance of mixing them up
      is higher than for two randomly chosen commands (by now there's
      empirical evidence to support this).

   2. The severity is quite high -- you lose your changes.

Therefore, we ought to consider this a real problem. At some point,
when I come back from cvs2svn-land or just need an hour's break, I'll
change "resolve" to "settle", unless a) people strongly object, or b)
someone beats me to it :-).

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Tue Jun 10 18:28:29 2003