KWin does not generate window shadows for VLC

[Duncan]

Nikos Chantziaras posted on Fri, 30 Jul 2010 23:49:36 +0300 as excerpted:

> I'm using SMPlayer mainly, but I've come across some missing features
> (like the inability of mplayer to bring silent and loud parts of DTS and
> AC3 audio closer together, like PowerDVD on Windows is able to do) and
> so I thought I'll try VLC and see if it does better.
> 
> I'm on Gentoo too btw, so I know mplayer isn't crippled.

=:^)  I thought I recognized the name.

FWIW, I had written something additional about VLC and KDE, but deleted it 
as going OT.  But you (and perhaps others here) will be interested if 
you've not already seen it, probably due to following kde-planet (tho I 
recently quit following it as it was too much to keep up with, and a curse 
of mine is that I'm /somewhat/ interested in nearly everything, so if I'm 
following it, I find myself reading a lot more than I really have time 
for, but there are some useful infos and previews occasionally, for those 
with the time).

Some weeks (months?) before akademy, one of the phonon devs was blogging 
about it, qtmultimedia (IIRC that's what it's called) and VLC.  There's 
some history in the phonon backends.  The xine backend was first, and 
early on the only one fully developed and thus the default, but it turned 
out that as the phonon mission expanded, xine wasn't as well suited to the 
job of default backend as initially thought.  Among other things, when qt 
adopted phonon, it didn't work quite as well for what they needed as for 
kde.  Then came the gstreamer backend, which I guess the qt folks put some 
work into.  But then with the nokia refocus of qt to strongly emphasize 
phones and other handhelds, it turned out phonon wasn't as good a fit for 
them as originally thought, either.  The result is two-fold.  First, qt is 
continuing to support current qt-phonon status but not really putting 
anything into further development, with qt-multimedia getting the focus 
there, as it's more flexible for small-form-factor mobiles but with 
development should scale up to full computers better than phonon scales 
down.  Second, where phonon is with qt, the xine-backend is with phonon, 
it was the first reasonably mature backend, but they'd end up having to 
write the features needed for it in ordered to mature phonon further on 
that backend, so it's staying basically where it is.

That puts the gstreamer backend front and center for the short term, as 
it's the only other relatively mature backend.  (Unfortunately, I had 
always found gstreamer too complex for the level of use I got out of it, 
considering it primarily a gnome tech, so don't have it even installed, 
and continue to use the xine backend.  Oh, well...)

But, intermediate term, unlike either xine or gstreamer, VLC seems to 
already have all the features the kde/phonon folks are looking for in a 
backend, so, according to this blog, which described the outcome of a 
meeting they had (IIRC several of the media/phonon devs happened to be at 
plasma tokamak, and they had a mini-summit there where much of this was 
discussed, and preliminarily decided, folks that are interested can thus 
place the timing of this information a bit more precisely than I did 
earlier, by looking that up), it's likely to become the default phonon 
backend in the medium term.  My interpretation of that was 4.6 to 4.7, but 
AFAIK, that's purely my interpretation, and it would have been preliminary 
plans at that point anyway.

Meanwhile, longer term and extremely preliminarily at that point, given 
that kde is solidly built on qt already, the blog said that it may be that 
phonon will lose focus much as has the xine backend, with qt-multimedia 
taking over much of phonon's role.  However, that was labeled as pure 
speculation at this point, and it was admitted that at least thru kde4, 
phonon is a fixed part of the api that they really can't, nor are they 
planning to, get rid of entirely.  However, it may well be that at some 
point, what's left of phonon will be an API shim between qt-multimedia and 
apps still using that API, with all the exciting new development based on 
qt-multimedia directly, without the shim.

The obvious comparison is to arts, which became the no longer actively 
developed and only barely supported millstone around kde3's neck, well 
before kde 3.5, and /years/ before kde3 was declared dead.  However, 
there's both definite positives and definite negatives to be had from the 
comparison/contrast of the two.  Learning from arts, the phonon devs very 
specifically and deliberately made it modular instead of monolithic, and 
placed a priority on keeping the code clean and internally well 
documented, to ease long term maintenance even if the current maintainers 
should disappear (as happened with arts).  Thus, even worst-case, phonon 
cannot and will not be the millstone around kde4's neck that arts was 
around kde3's neck.  The modular system, clean code, and decent 
documentation (apparently at times at the cost of rejecting functionality 
until it was up to standard), should see to that.  However, just as kde3 
was the only arts user, kde4 is the only major phonon user, and other qt 
development using it now will likely migrate away from it in time, as kde4 
is the only one that has to maintain api compatibility across such a wide 
development ecosystem, so it's certainly possible that some of the 
warnings when kde set out to develop phonon, may come true to some degree, 
and people could then be shaking their heads and asking why kde didn't 
learn from the arts mistake (tho it did, as I explained above, and point 
of fact, it has indeed been a learning experience, nothing adopted then 
would have continued to serve kde4's needs without major changes, and the 
modularization and layered phonon API *WILL* and already *IS* standing kde 
in good stead as the backends change out from under it).

But, back to the topic of VLC.  My whole reason for writing all this is to 
point out that VLC is a very good thing for kde users to be learning about 
right now, as it's very likely that the VLC backend in development now, 
will become the recommended default, within the next three six-month 
feature release versions after the 4.5 about to come out now, thus, likely 
next year, but almost certainly by early 2012 if nothing big comes along 
to change things.  Provided that prediction is correct, VLC should then be 
safe thru 2012 and into 2013, at which point the long-term predictions 
involving qt-multimedia start coming into play -- but that's out there 
pretty far in terms of predicting FLOSS in any case, and there's a lot 
that could happen before then to change things, so the fate of phonon 
relative to qt-multimedia remains very much up in the air, even tho phonon 
support is assured thru kde4.  (OTOH, many kde3 users know quite well how 
well such promises of continued support as long as there are users are 
honored, that is, mostly in the breach thereof, when it comes to dropping 
old versions like hot potatoes before the new versions are properly and 
fully functional replacements for relied upon functionality, portions of 
which are even still not there.  Such promises are apparently about as 
long-lived as the photons of light impinging upon your eyes by which you 
read them.)

-- 
Duncan - List replies preferred.   No HTML msgs.
"Every nonfree program has a lord, a master --
and if you use the program, he is your master."  Richard Stallman

___________________________________________________
This message is from the kde mailing list.
Account management:  https://mail.kde.org/mailman/listinfo/kde.
Archives: http://lists.kde.org/.
More info: http://www.kde.org/faq.html.