launchpad-dev team mailing list archive

[Deryck Hodge <deryck.hodge@xxxxxxxxxxxxx>]

Hi, Robert.

I'm -1 on this.

On Mon, Oct 18, 2010 at 8:53 PM, Robert Collins
<robert.collins@xxxxxxxxxxxxx> wrote:
> I've chatted in the last few reviewers meetings about the review
> process, costs / benefits, what we want from it and so on.
>
> But thats all pretty vague stuff; I've got a specific experiment in
> mind that I think will at worst teach us what we value in reviews more
> clearly (e.g. for me its knowing that other folk can understand what
> I've done - I have problematic conceptual stuff hammered out way in
> advance).
>

It's the last sentence above that has me suddenly becoming as
ultra-conservative as the state I call home.

There are a number of ways to read that sentence, a couple of which include:

* I only get review so you can learn from me
* I did my home work and coded the implementation perfectly

Granted, I'm being a bit hyperbolic to make the point, and I hope you
take that with the great respect I have for you. :-)  But I do wonder
how much of this is really a problem generally and how much of it is a
problem specifically for you because of the pace at which you work or
less reviewers in your timezone.

This is not to say that we don't all feel the burden of code review at
some point, but in the realm of things that slow us down, I think
review time is quiet low on the list of things to improve the speed
of.  I can think of 2-3 times in the recent past where a developer was
frustrated by the time spent in review, and in the end, this had more
to do with the tone of the review, not understanding what was being
asked, and general frustration at the hidden complexity of what was
assumed a simple fix.  In the end, even the frustrating review was
worthwhile.

Can we not change the culture and speed of reviews given the current
process?  As others note, we already have rs reviews.  Why not use
those in this case?  Does it take that long to ping a developer and
say, "hey, I'm not doing anything but adding some better docstrings.
Can I have a preemptive rs?"

Cheers,
deryck


-- 
Deryck Hodge
https://launchpad.net/~deryck
http://www.devurandom.org/