Re: Pending Reviews page

On 11/14/24 14:38, Alexandre Julliard wrote:
> Rémi Bernon <[email protected]> writes:
> 
>> On 11/14/24 11:42, Alexandre Julliard wrote:
>>> Folks,
>>> In our grand tradition of shaming people into doing work, after the
>>> Regressions page I'm proud to present the Pending Reviews page:
>>>     https://source.winehq.org/reviews
>>> As with the regressions page, the goal should be to *not* be number
>>> one
>>> on the list. But feel free to congratulate Huw anyway ;-)
>>> Note that the list is a little overestimated, commenting on the diff
>>> counts as a review but simply adding a note doesn't, so you may see some
>>> MRs in your list that you have actually commented on already. If people
>>> find this thing useful we can try to make it smarter. Feedback is
>>> welcome!
>>>
>>
>> Nice!
>>
>> Does it mean there's now a difference between a MR assigned to you and
>> one that isn't assigned? :)
> 
> Not much difference, but hopefully making the unassigned MRs more
> visible will encourage people to voluntarily assign themselves to the
> ones where they feel they can help.
> 

I think many times it's difficult to look at some unknown code area and 
believe you can give some meaningful guidance. In addition it seems like 
that assuming reviewer-ship and approving a MR is considered by many as 
a big deal, whereas I'm not sure it should be so much.

I don't have much suggestions on how to make it less overwhelming but I 
wonder if using Gitlab merge request tags could help us give more 
detailed feedback to contributors, and perhaps offer an intermediate 
step for drive-by developers feedback.

We have the Draft state already, and the page already kind of have some 
another class of 'pending further action' MR, although they are not 
explicitly classified as such on Gitlab. We could have some additional 
tags like 'needs tests', or whatever else would useful.

I've also been told, and I tend to agree, that it would be good to 
consider having some sort of 'Acked-by' status, where a MR has been 
looked at, at least briefly, and looks good but needs further review 
(not necessarily by the same person).

-- 
Rémi Bernon <[email protected]>