Re: Release plans

[Dan Kegel]

Looks like Alexandre must have fired up his time machine again :-)

And as long as it's up and running, how about a look forward
at the 1.4 release plans?

Release plans

[James McKenzie]

On 9/12/10 6:36 AM, Dan Kegel wrote:
> Looks like Alexandre must have fired up his time machine again :-)

And I thought I was alone...
> And as long as it's up and running, how about a look forward
> at the 1.4 release plans?

Would be nice to know what has priority for this release.  I would love 
to see the DIB Engine be the 'deal maker'.

James McKenzie
>

Release plans

[Dan Kegel]

James McKenzie wrote:
> Dan Kegel  wrote:
> > And as long as it's up and running, how about a look forward
> > at the 1.4 release plans?
>
> Would be nice to know what has priority for this release.
> I would love to see the DIB Engine be the 'deal maker'.

Ain't gonna happen.  Too hard, not enough manpower
going into it.

A more realistic goal held over from 1.2 might be
DX10 support.  There are several people chugging away,
getting bits and pieces of that committed.

I proposed a few smaller goals for 1.4 at
   http://wiki.winehq.org/WineReleaseCriteria :

Bug 6971, the mouse problem affecting many FPS-style games,
 IIRC Alexandre says it would take him four weeks to clean
 up the XInput2 patches

AcceptEx (bug 280) - needed for Warcraft III and a number of other
games (Mike Kaplinskiy's real close on this, so maybe it doesn't
even bear mentioning as a 1.4 goal)

Antialiasing/Multisampling (Roderick's got a patch that needs a few
weeks of cleanup)

- Dan

Release plans

[James McKenzie]

On 9/12/10 12:29 PM, Dan Kegel wrote:
> James McKenzie wrote:
> > Dan Kegel  wrote:
> > > And as long as it's up and running, how about a look forward
> > > at the 1.4 release plans?
> >
> > Would be nice to know what has priority for this release.
> > I would love to see the DIB Engine be the 'deal maker'.
>
> Ain't gonna happen.  Too hard, not enough manpower
> going into it.

Then the second goal is not going to happen either :(  Emmanual 
Malliard? and others have stated that the DIB engine is a pre-requisite 
for it.
> A more realistic goal held over from 1.2 might be
> DX10 support.  There are several people chugging away,
> getting bits and pieces of that committed.

The 'missing' pieces of DX9/DX10 would be a great deal maker for 1.4
> I proposed a few smaller goals for 1.4 at
>     http://wiki.winehq.org/WineReleaseCriteria :
>
> Bug 6971, the mouse problem affecting many FPS-style games,
>   IIRC Alexandre says it would take him four weeks to clean
>   up the XInput2 patches

This is under way by a few folks, right?
> AcceptEx (bug 280) - needed for Warcraft III and a number of other
> games (Mike Kaplinskiy's real close on this, so maybe it doesn't
> even bear mentioning as a 1.4 goal)

I hope that Mike can solve this one.
> Antialiasing/Multisampling (Roderick's got a patch that needs a few
> weeks of cleanup)

Again, this should be done quickly as well.

Wine 1.4 should have something in it along the level of x64 code or a 
bunch of updates/fixes.

I like the third one, and it is quite possible that this will be the one 
to make Wine 1.4 a possibility.  Many folks would love to use their 
favorite USB 'thingy', be it a dongle, iPod, etc.  that does not look 
like a serial device or a drive device.

Anyway, I'm plugging away at Max's last DIB attempt in bug 421 to see if 
it breaks on the Mac Intel platform.

James McKenzie
> - Dan

Release plans

[Edward Savage]

On Mon, Sep 13, 2010 at 5:29 AM, Dan Kegel  wrote:
> James McKenzie wrote:
> > Dan Kegel  wrote:
> > > And as long as it's up and running, how about a look forward
> > > at the 1.4 release plans?
> >
> > Would be nice to know what has priority for this release.
> > I would love to see the DIB Engine be the 'deal maker'.
>
> Ain't gonna happen.  Too hard, not enough manpower
> going into it.
>
> A more realistic goal held over from 1.2 might be
> DX10 support.  There are several people chugging away,
> getting bits and pieces of that committed.
>
> I proposed a few smaller goals for 1.4 at
>   http://wiki.winehq.org/WineReleaseCriteria :
>
> Bug 6971, the mouse problem affecting many FPS-style games,
>  IIRC Alexandre says it would take him four weeks to clean
>  up the XInput2 patches
>
> AcceptEx (bug 280) - needed for Warcraft III and a number of other
> games (Mike Kaplinskiy's real close on this, so maybe it doesn't
> even bear mentioning as a 1.4 goal)
>
> Antialiasing/Multisampling (Roderick's got a patch that needs a few
> weeks of cleanup)

I like the looks of this list, from a gaming point of view.  An
increasingly important missing feature is basic GFWL support (depends
.net 3.5sp1) which I hear is getting there slowly.  Would be nice to
have by 1.4.

Out of interest why are applications not considered release goals? I'm
sure there is a very good reason I'd just like to know it.  Just seems
to me that it would be a good idea to pick a handful of very popular,
but mostly ignored, applications and focus on having them work well by
release (CS5 is an example I can think of immediately).

Is there a time frame for 1.4? 1 year, 2 years, sooner?

Release plans

[Francois Gouget]

On Mon, 13 Sep 2010, Edward Savage wrote:
[...]
> Out of interest why are applications not considered release goals? I'm
> sure there is a very good reason I'd just like to know it.  Just seems
> to me that it would be a good idea to pick a handful of very popular,
> but mostly ignored, applications and focus on having them work well by
> release (CS5 is an example I can think of immediately).

One should not have to go out and buy a specific application to be able 
to work on a release goal. Another issue is applications that keep 
changing: we don't want to make Fozzle 7.0 into a release goal if one 
month from now all one can download is Fozzle 7.3. Applications that 
have neither issue could potentially be turned into release goals.

Release plans

[Scott Ritchie]

On 09/12/2010 06:36 AM, Dan Kegel wrote:
> Looks like Alexandre must have fired up his time machine again :-)
>
> And as long as it's up and running, how about a look forward
> at the 1.4 release plans?

I've discussed this idea with a lot of (non-Wine-Developer)
stake-holders, and a few things came up pretty consistently.

1) Have some sort of time limit for the freeze process to start.
Specific features are great for a release, but recognize that Wine gets
dozens of features every month in the form of bug fixes and new
applications working.  Add enough of them together and you have
something worthy of a release even without one big thing in particular.

2) That said, there are a few big things that would be great as release
goals in and of themselves.  That means substantial effort to include
and polish them on behalf of core Wine developers, followed by a freeze.
 Possible features to name here have already been listed by others and
in the wiki - DIB engine, Direct3D 10, OpenAL audio, etc.

3) Instead of a software developing perspective, we could instead
release based on a QA perspective.  This would mean ignoring which
features were ready (or almost ready) and instead focus on the QA
metrics we have - make the test suite pass everywhere (and on Windows),
make sure we have some metric of test coverage, make sure the number of
nonworking apps hasn't increased in some time, and so on.

These ideas can all be combined, eg we don't freeze unless we have a big
feature OR a whole year has passed, and once we freeze we don't release
until the test suite passes on Windows, Linux, and Mac.

Thanks,
Scott Ritchie