Wine staging patches submission

[Kai Krakow]

Hey Alistair!

2018-05-26 12:49 GMT+02:00 Alistair Leslie-Hughes <leslie_alistair at hotmail.com>:
> Hi Kai,
>
> Thank you for your feedback.
>
> On 26/05/18 19:34, Kai Krakow wrote:
>> Resent to wine-devel.
>>
>>
>> 2018-05-26 11:33 GMT+02:00 Kai Krakow <kai at kaishome.de>:
>>> Hello!
>>>
>>> First words: Thanks to you and many other people here for a really
>>> fantastic job of bringing thinks forward in wine-vanilla. I think (and
>>> can confirm personally) that Wine made huge steps forward in the past
>>> months, and I attribute that mostly to the efforts that have gone into
>>> reviving the staging project and getting patches into mainline. For
>>> me, a lot of games have seen vast improvements without using the
>>> staging patchsets. I only cherry-pick a few patches from staging (like
>>> nvcuda/nvapi) to get things work. I don't use the patchset workflow
>>> but instead rely on the all-patches-rebased branch from wine-staging
>>> to do that, rebasing my own cherry-picks. And I've seen lots of
>>> improvements here, too, for wine-staging: Rebase conflicts have gone
>>> down a lot. And especially this is a huge time saver... Time for
>>> putting efforts into other parts of development.
> Its great to know that our hard work is paying off for the community.

Giving some positive feedback is the least I can give. From personal
experience I know that positive feedback is important for motivation,
and it's the part of development that very often is forgotten about.

The Wine project made very impressive steps forward in the last
months. IMO, the new release cycle system is probably what kicked all
these great improvements off. But of course it rendered staging into a
difficult state to keep up. So it's great to see that staging thinks
about new ways to integrate with the main project and starts working
on the patches again. I think there's sleeping great potential in many
of those patches. This is why I think the staging project needs some
reshaping, needs starting to think different about how to handle it,
without throwing away hard work, without forgetting about the workflow
that devs arranged with over many years (which probably has a steep
learning curve).


>>> So great thanks to all people involved, even when some decisions may
>>> cause some rumblings here and there, I can imagine the bigger plan of
>>> this with a great result.
>>>
>>> But I think it makes sense to not blindly accept everything and add
>>> _valuable_ discussion.
> As patches are brought to our attention, they are assessed on a
> per patch basis.  Greater chance if they contain tests ;)

Heh, now this wasn't exactly what I meant - but yes: patches review is
important, and feedback is important for high quality.

But what I was talking about is: One shouldn't blindly accept any
proposed change. I can hear Dmitrys point.

I don't know his exact intention but it probably made quite obvious
that changes should be discussed first, or at least be explained more
in detail. Some devs may feel upset to hear about how their workflows
they've become comfortable with over the past years, are ripped to
peaces. But clearly, some change is needed as otherwise wine-staging
will completely die or become a very closed/private area of abandoned
patches.

Thanks,
Kai