[sakai2-tcc] Which parts of the code need a mandatory review of changes

Jean-Francois Leveque jean-francois.leveque at upmc.fr
Wed Jul 17 07:50:51 PDT 2013


On 17/07/2013 16:27, Neal Caidin wrote:
> Hi J-F,
>
> I'm not sure if I understand your comment. Is it about having code reviews?  What kind of list do you want?
>
> A couple of thoughts/questions wrt Code Review:
>
> 1) Maybe discussing the review process is a question for the CLE release team and not TCC?

Because this was part of a TCC meeting, I thought it was up to the TCC 
to decide, even if the release team contribution is welcome.

> 2) The existing "process" seems to be to ask for help in reviewing, over email list or on the CLE release team calls.
>
> 3) Do we need a better defined process for reviews, and if so, how realistic is it that the process will be complied with?
>
> 4) I'm still looking into Crucible, but that is just a tool to make reviews easier, it does not solve the process issue.
>
> 2 cents.
>
> Cheers,
> Neal

My feeling is we really need to have, for the parts that need double 
checking even when changes come from trusted contributors, a clear 
process in order to avoid reviews to be waiting for a long time.

When it's kernel, portal, access (or other parts of the code not yet 
mentioned), I think we need to have a list of explicitly approved for 
one important part of the code and also available reviewers. We also 
need to know they will be reviewing on a regular basis, even if it's no 
more than one submitted fix a week.

Contributing to the community could be less frustrating. :(

2 sous.

Cheers,

J-F

> On Jul 17, 2013, at 8:40 AM, Jean-Francois Leveque<jean-francois.leveque at upmc.fr>  wrote:
>
>> Hi,
>>
>> This subject was part of the Thursday meeting in San Diego.
>>
>> I remember kernel and access being mentioned and I understand portal is
>> also important.
>>
>> I would really like to have a list instead of learning by trying because
>> this might not be obvious for all of us.
>>
>> I would also like this review process to be defined. The questions I
>> could ask start with How, Who and How often.
>>
>> Cheers,
>>
>> J-F


More information about the sakai2-tcc mailing list