On Tue, Jul 14, 2009 at 10:36:26PM +1000, Ben Finney wrote:
> "W. Trevor King" <wking@drexel.edu> writes:
> > I've switched my branch over to the current url, and moved to
> > last-commit-timestamp version numbers.
> 
> Please, no. Timestamps aren't version strings, that's conflating two
> pieces of information with very different meanings. Correlating the two
> is the job of a changelog.

Which we don't bother keeping (also NEWS), since "bzr log" works so nicely.
If you really want an standard changelog, see
  http://mail.gnome.org/archives/desktop-devel-list/2007-September/msg00186.html

> > This removes the "prefered branch" issues with the old scheme, and
> > version numbers should increase monotonically
> 
> The English word “should” is ambiguous in this context. Are you saying
> this is desirable, or are you predicting that it will likely be the
> case?

Both.

> I don't see how it's either, so am doubly confused :-)

The timestamp should at least replace the patch release number, which
you agree is-desirable-to increase motonically ;).  I also predict
that it will increase monotonically for any given branch, since the
branch HEAD will have both the most recent commit and the highest
version number.  The only problem I can think of is having your clock
_way_ off, and that is unlikely enough to ignore.  If you hop between
branches, the timestamp is much more likely to increase going to the
more modern branch than the bzr revision number, which desynchronize
between branches fairly quickly.

> The convention for three-part version strings is often:
> 
>   * Major release number (big changes in how the program works,
>     increasing monotonically per major release, with “0”indicating no
>     major release yet)
> 
>   * Minor release number (smaller impact on how the program works,
>     increasing monotonically per minor release, with “0” indicating no
>     minor release yet since the previous major)
> 
>   * Patch release number (bug-fix and other changes that don't affect
>     the documented interface, increasing monotonically per patch
>     release, with “0” indicating no patch release since the previous
>     major or minor)

One problem is that we don't actually have "releases".  People just
clone a branch, install, and go.  If you're worried about stability,
just clone from a more stable branch (i.e., Chris' trunk).  I think
this is good for distributed development, but maybe makes it hard to
package into a conventional release-based system.  With the bzr patch
number in setup.py as the patch release number, I would be releasing
my 0.1.363 while Chris releases his 0.1.314, even though they're at
about the same point.  I would rather be releasing my
  0.1.20090714121347
while Chris releases his
  0.1.20090713154540
Since then the similarity is clearer.

At any rate, I think the two approaches are close enough that an
auto-updating timestamp beats a manually bumped patch number, since
no-one ever actually bumps the patch number ;).

-- 
This email may be signed or encrypted with GPG (http://www.gnupg.org).
The GPG signature (if present) will be attached as 'signature.asc'.
For more information, see http://en.wikipedia.org/wiki/Pretty_Good_Privacy

My public key is at http://www.physics.drexel.edu/~wking/pubkey.txt