Re: [PATCH RFC 2/2] docs: process: submitting-patches: clarify imperative mood suggestion

[Jonathan Corbet]

Ahmad Fatoum <a.fatoum@xxxxxxxxxxxxxx> writes:

> Hello Jon,
>
> On 30.12.24 19:40, Jonathan Corbet wrote:
>> Ahmad Fatoum <a.fatoum@xxxxxxxxxxxxxx> writes:
>> 
>>> While we expect commit message titles to use the imperative mood,
>>> it's ok for commit message bodies to first include a blurb describing
>>> the background of the patch, before delving into what's being done
>>> to address the situation.
>>>
>>> Make this clearer by adding a clarification after the imperative mood
>>> suggestion as well as listing Rob Herring's commit 52bb69be6790
>>> ("dt-bindings: ata: pata-common: Add missing additionalProperties on
>>> child nodes") as a good example commit message.
>>>
>>> Signed-off-by: Ahmad Fatoum <a.fatoum@xxxxxxxxxxxxxx>
>> 
>> I'm rather less convinced about this one.  We already have a whole
>> section on describing changes.  Given that this crucial document is
>> already long and hard enough to get through, I don't really think that
>> adding some duplicate information - and the noise of more labels - is
>> going to improve things.
>
> Do you agree with the content of the patch in principle?
>
> My changes were motivated by a disagreement about the necessity of having
> to use the imperative mood throughout as I described in my cover letter,
> so I still think think that a clarification is appropriate.
>
> Would a v2 without the example at the end be acceptable?

I will always consider a patch, but the example isn't the concern,
really.  The information you are trying to add to an already too-long
document is already present there; I think that repeating it, and making
this crucial document that much more unapproachable, would actively make
things worse.

Thanks,

jon