On Tue, Feb 05, 2013 at 09:58:14AM -0500, Dave Phillips wrote:
> Too many distros.
What should be the problem here ? Natural selection will deal
with this. Those that 'work' will survive.
> Confusion re: desktops, and GUI toolkits.
That's actually two issues. Regarding GUI toolkits, you use
either Gtk, or Qt, or a thin layer on top of X11 as I'm
doing. All three of them will be available in any distro,
so I don't see a problem with those.
Desktops are another issue. IMHO any app that depends on a
particular desktop environment to work (or fully work) is
just broken. There's a lot of those around, but fortunately
most of them are not audio related.
> Too difficult to set up audio system.
It certainly was in its first years and for a large part that
was due to limited system support, immature real-time libs,
etc. Since longer than I can remember it 'just works'.
Assuming you want a system meant to produce music or process
audio, once you have Jack working that's more or less it.
And that hasn't been a problem (for me at least) for a long
time. It's years since I have compiled patched kernels, or
had to install Jack from source.
Fact is that the introduction of PA into mainstream systems has
created a lot a trouble - Jack and PA serve really different
purposes and don't go together easily.
Another problem is that there are still lots of apps around that
have no native Jack support, or broken support usually provided
by some 'cross platform' toolkit. Just dump those. Refuse to use
them. Don't advertise them.
> Not enough native plugins, esp. instruments.
Some things are missing. But why on earth should an instrument
be a plugin ? 'Real' instruments are not plugins, they provide
an input signal to whatever SW you use.
> Inconsistent support for VST/VSTi plugins.
Paul Davis has already pointed out that this can mean two
different things. If someone's workflow *really* depends on
native Windows plugins - repeat *if* repeat *really* - then
why on earth should such a person want to use Linux ?
As regards 'native Linux' VST, I don't see the point of it.
> Too many unstable/unfinished applications.
Agree 100%, not just apps but also plugins - there are lots
of those either don't do the thing they are supposed to do,
or do it very poorly, or crash when you give them the right
inputs. Basic lack of care for quality, ignorance, incompetence
or any combination of those. Just let them die and be forgotten.
> Too many "standards" (esp. wrt plugins).
None of them being really standard, nor really up to the task.
Which means there will even be more. I wouldn't really want to
see any of the existing ones becoming the only choice. Also,
IMHO there's nothing wrong with an ad-hoc plugin standard, one
tailor-made for a particular app. Since it won't try to be
everything to everyone, it would probably be a lot simpler and
easier to use than one that does. I'm currently developing two
app that will have plugins. They won't have a common standard.
But each of them will be easy to port your code to.
> Poor support for certain modes of composition (think Ableton Live).
I really couldn't care less. If someone wants to indulge in those
'modes of composition', he/she should just use the tools that define
them. I do not buy the idea that Linux should offer free clones of
whatever exists in the commercial world.
> Too much conflict/fragmentation within the development community.
Good. We are not a religion or a sect. People will disagree, compete
or just do their own thing and even then share their work anyway,
leaving you with the choice.
> Poor external/internal session management.
Those are real concerns.
A world of exhaustive, reliable metadata would be an utopia.
It's also a pipe-dream, founded on self-delusion, nerd hubris
and hysterically inflated market opportunities. (Cory Doctorow)
Linux-audio-dev mailing list