|| ||Lennart Poettering <mzerqung-AT-0pointer.de> |
|| ||devel-AT-lists.fedoraproject.org |
|| ||Re: Is PulseAudio dead? |
|| ||Tue, 3 Aug 2010 03:06:13 +0200|
|| ||Article, Thread
On Mon, 02.08.10 23:50, email@example.com (firstname.lastname@example.org) wrote:
> > So, I guess what I want to say is: I will return full-time to PA not so
> > far away. And I have a queue of patches in my checkout (including volume
> > ramping and plug-in effects and similar). Also note that I'll run the
> > track about audio at plumbersconf again, so there's really no reason to
> > believe that I moved on or PA was dead.
> Which is great and I understand that but systemd will basically cover
> the release time frame for F-13 and F-14 and in that timeframe the
> support and issues for PA are going unfixed or even un triaged. Not
> great for a core sub system. So maybe it would be a good idea to train
> up a few people that can do the boring trage so you can get on with
> the upstream PA and systemd stuff so that the average end user doesn't
> need to wait for the bottle neck of a single person because presumably
> with other distros using it Fedora isn't the only distro demanding
> your time.
Audio hackers unfortunately don't grow on trees. In my counting, there
are 3 people paid in the whole industry who work on general purpose
audio infrastructure of Linux. Two of them are basically busy with
keeping the HDA driver up-to-date, if I am correctly informed. The third
one is me.
As long as things are that way the entire weight of fixing bugs all
across our consumer audio stack are basically lying on three pairs of
shoulders, and that defines the speed in which we process bugs. So,
please be patient.
One would wish that a certain other company with a clear focus on
desktop Linux (where consumer audio is a key part of) would want to
actually hire more folks in this area, but well, ...
I not sure whether it should be considered a failure of us consumer
audio hackers that we never managed to attract a bigger number of core
contributors. But well, I am not a Jono Bacon, and I have done quite a
number of talks about PA and related techs on many conferences, both
about the technical details and from a more user-related
perspective. While I like to believe that people did enjoy my talks they
didn't really have the effect of boasting the numbers of core hackers of
our audio infrastructure.
Audio hacking is often quite complex unfortunately. You need to have a
basic idea of signal processsing and RT stuff. The code involved is time
critical and usually very low-level. That makes the learning curve
steep, and doesn't help growing audio hackers.
Then again, something similar can probably be written about every other
part of our Linux infrastructure. I am still waiting for the project
that doesn't have too feww, but too many people making contributions
But anyway. We have come quite far in the last years, and I actually
think the status quo is not bad at all anymore. I have a pretty
positive view on things, so while it of course would be great if we
could fix all open bugs tomorrow, I don't think it should be considered
a catastrophic desaster if we didn't.
Lennart Poettering - Red Hat, Inc.
devel mailing list
to post comments)