Review Process Experiment (1 LGTM)

[Tim Penhey]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On 20/08/13 01:27, Nate Finch wrote:
> I accidentally only replied to John with this earlier:
> 
> I like the shared responsibility and the review of existing code,
> but I do lament the loss of that extra pair of eyes on the code.
> Even in my short time at Canonical, I have almost always seen the
> second review turn up things the first review missed. This is
> likely due to each reviewer valuing different things and therefore
> looking at those things more carefully, and probably also that the
> first one picked up the easy stuff, so the second one could ignore
> that and look for more subtle problems. Granted, most of the time
> it's not incorrect code, just something that could be tweaked to be
> better, but that still gives us higher quality code overall.
> 
> I do think that putting a higher bar on LGTMs is a good thing, and 
> hopefully that can help keep the quality high.

Well, we have a few things working for us here.

We use a DVCS which makes reverting a branch that causes real problems
very easy.  While this doesn't mean we should be committing just
anything, it does mean that we have an extra back-stop.

Also, this is just initially a one month experiment, with an end date
where we evaluate how things went.

Cheers,
Tim
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.12 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/

iEYEARECAAYFAlISi/kACgkQd1fvI4G7WRABdQCdF74H1hCrn5e3+R8dN3AhWb0l
WPEAoMTNYJiwJGsyj25LaIhsEnsnc3hY
=vz1a
-----END PGP SIGNATURE-----