Re: Review D38047 ... and then there was one....

In message <AE37187C-79D0-4B5E-87F0-6BB52822F03B@gmail.com>, Enji Cooper 
writes
:
> 
>
> > On Oct 7, 2024, at 08:08, Cy Schubert <Cy.Schubert@cschubert.com> wrote:
>
> > Yes. I was about to suggest this. Plus, any proposed commit log message
> > must answer the questions why, what and how. With special attention to why=
> .
>
> I have the same feelings as Cy.
>
> FWIW, part of the reason why large/complex changes like this languish in my r
> =
> eview queues is in part due to reasons like this.

Totally!

>
> Unless I am a SME in the area who is driven to understand what the change ai=
> ms to achieve, I will not take the time to review large/complex chances. I h=
> ave a lot of other things in my life which take priority over large code rev=
> iews.

Agreed. In the case of nscd, I applied one fix to it years ago. Like you, I 
do not feel qualified to review a large and complex jumbo patch (group of 
patches).

>
> Please break the large change down into a smaller set of changes/reviews to m
> =
> ake it easier to review the overall change effectively.

Yes. I have the same issue at $JOB. We use github enterprise. I refuse to 
review such patches and yet others, less experienced, summarily approve 
such patches without even looking at them, breaking our infrastructure.

The choice is clear.

>
> Cheers,
> -Enji=


-- 
Cheers,
Cy Schubert <Cy.Schubert@cschubert.com>
FreeBSD UNIX:  <cy@FreeBSD.org>   Web:  https://FreeBSD.org
NTP:           <cy@nwtime.org>    Web:  https://nwtime.org

			e^(i*pi)+1=0