I agree that the code needs to be self-explanatory enough to not require
annotations, but annotations can be useful - especially for larger changes.
Suggesting the order for code to be reviewed is certainly useful if you're
reviewing code in a part of the system you aren't familiar with


On Wed, Jun 25, 2014 at 10:25 AM, Jeroen Vermeulen <
[email protected]> wrote:

> On 2014-06-25 09:43, roger peppe wrote:
>
>  About pre-review annotations, I agree with Ian that the code should be
>> documented
>> well enough that someone coming to it from scratch can understand it, but
>> I also wonder if there is a room for review-specific comments, talking
>> about
>> reasons for the changes themselves in the specific context of that review.
>>
>
> There is, I think.  But should it be quite so close to the code, where it
> competes against commenting for the coder's time?
>
> Don't know if there's a definite answer, because either way we assume a
> human process to complement the technical solution.  But if a coder starts
> by reviewing their own code, perhaps they should also turn these notes into
> a single coherent "cover letter" and, in explaining, perhaps spot
> structural flaws or anticipate questions.
>
>
> Jeroen
>
>
> --
> Juju-dev mailing list
> [email protected]
> Modify settings or unsubscribe at: https://lists.ubuntu.com/
> mailman/listinfo/juju-dev
>
-- 
Juju-dev mailing list
[email protected]
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/juju-dev

Reply via email to