Re: How can we decrease the cognitive overhead for contributors?

[Simon Tournier]

Hi,

On Sat, 26 Aug 2023 at 19:42, kiasoc5@disroot.org wrote:

> I would like to hear from committers if non-committer reviews are 
> helpful, because I don't really know how or what I can comment on for 
> incoming patches on packages I'm not really familiar with.

Well, I do not have enough committer experience.  However, I commented
several patches and my aim was the one I would like to see for helping
me as committer before pushing.

Because I am signing-off the commit, it is my responsibility to check
that the commit is compliant with Guix standards.  So, whatever I will
read as other review, I will check myself.  However, if another person
has also checked, it helps me because it avoids me to double or
triple-check in order to be sure I am not missing something.  Somehow,
another person looking at the patch reinforces my confidence that all is
fine.  Other said, without diluting the responsibility, it dilutes the
pressure, somehow.

Well, from my point of view, the comment “build locally“ is not very
helpful.  The comment “work locally” is a bit more.  IMHO, the most
helpful appears to me, in that order:

 1. the check of license – sometime upstream says the license is GPL
    when it’s true for most of the code but a tiny subpart uses some
    non-free license, such comment is very helpful.

 2. the good writing of synopsis and description.

 3. the ’home-page’ field is correct.

 4. some comments about the ’arguments’ field.

And the next level appears to me:

 5. apply the comments about #2 or #3 or #4 and re-send another version
    (reroll-count).  That’s what committers are often doing and except
    some cases, that is not a co-author ship. :-)


Cheers,
simon