[linux-audio-dev] Re: [linux-audio-user] Free Software vs. Open Source: Where do*you* stand?

Pete Bessman ninjadroid at gazuga.net
Tue Feb 21 04:15:48 UTC 2006


On Mon, 20 Feb 2006 15:21:19 -0500, "Paul Davis"
<paul at linuxaudiosystems.com> said:
> On Mon, 2006-02-20 at 14:55 -0500, Pete Bessman wrote:
> > So let's hear it!
> >
> > WHAT is your NAME?
> >
> > WHAT is your QUEST?
>
> i can hack ardour part time while i work. i can hack ardour full
> time while i live the rest of my life. but i know for sure that
> i definitely can't get into philosophical debates with the
> recently deceased while i work full time on ardour and live the
> rest of my life.

Fool --- I'll eat your brains!

All seriousness aside, I'd like to address the myriad responses
I've received.

First, I want to reiterate that I don't want to get into a war about
morals.  Moral philosophy is an old enterprise, and it's rate of
progress in recent centuries has not been great (to quote Friedman).
I'm a libertarian, I believe in the right to contract, so I believe
in the right of a software producer to require a user to agree to an
EULA or walk.  I don't think this is really at all in conflict with
my belief that our current copyright law is a little whack --- I
think our contract law is just fine, and that's where I see the meat
of the validity and enforceability of proprietary software licenses
as coming from.

Now, naturally, you might disagree --- and obviously, plenty of
people here do. That's not the discussion I wanted to have, though!
Something of that depth is better served by a blog post than a flamewar
here, I think.

That said, I hate to say it, but I think I'm going to go ahead and stir
up some more controversy.  I agree that, in many instances, linux audio
probably meets and exceeds the "good enough" test --- that is, it either
gets the job done flawlessly, or can be made to done so with a little
work.  Straight recording, mixing, and mastering are probably at this
level, and I'd wager you can produce pro-quality stuff with relative
ease using current tools in this realm.

However, as far as synthesis, sampling, sequencing, and other assorted
electronica, I don't think it really cuts the mustard --- at least for
the cutting edge.  Sure, it can do scene music and minimalism and
abstract-avant-
garde stuff.  But take a listen to the latest from, say, Delerium, or
Front Line Assembly --- how in the world are you going to pull that off
with Hydrogen, ZynAdd, Seq24, rosegarden, muse, jack-rack, and
LinuxSampler?

Guitars and drums haven't changed in a while, nor has recording.  But
cutting-edge electronic music is defined by running on cutting-edge
electronic technology.  Go browse through musiciansfriend.com and take a
look at all the various proprietary offerings in this department.  Heck,
just look at Project5 or FL studio and the plugins it comes with.  You
simply have more tools at your fingertips in the proprietary world.  And
further, I agree with Russ --- the sequencing side of things just isn't
really there.  Maybe for stuff that's supposed to sound realistic.  But
try programming a complex synth line, replete with legatos and
extremely-
small interval fragments, and add in automation of filfreq, reso,
panning, etc.  It's doable, but it's painful.  And the more time you
spend fighting an app, the less time you spend making music.  There's a
reason tb303 synth apps are so popular, and it *ain't* just because the
tb303 hardware is rare --- it's because the truth is, it's a feckin'
PITA to use.

For the kind of stuff I'm interested in (see my above band references to
get an idea), I think it is eminently clear that Linux isn't good
enough.  Maybe it is just me, but I have been struggling intensely for
some time to get things to work out, and it just ain't happening.  If
anybody here has got some tracks that can prove me wrong, I'd love to
hear 'em.  And maybe I'll get back to Windows world and have the same
problems --- but I sincerely doubt it.

I'm not sure what it is --- some times, open source whomps up,
sometimes, it just can't keep up.  For the realm of music I'm in, it
lags.  Same thing for videogames (obviously). Maybe this will change
with time, but with all the new stuff that's been going on, the gap has
been widening if it's moved at all.  This causes a lot of cognitive
dissonance for myself, and I'm not sure why.  Maybe I drank too much GNU
kool-aid as a kid, or maybe it's all the years I've logged in emacs.
Most likely, I'm just obsessive compulsive, and I like to be able to say
"this is this and that is that." The fact that I can't yet figure out
the rule for when open source is going to produce the better product is
probably driving me slightly mad.  That, and the fact that years of
believing in Stallman gospel have probably deranged me a bit.

Man, I'm not sure where I'm going with this.

I suppose my overarching point is, don't confuse "good enough" with
"best."  From my moral perspective, at least, it's simply a matter of
what gives you the best cost-benefit ratio, and there are many, many
factors which make up both parts of that ratio beyond simple dollars.

I'll end this on a slightly unrelated note.  When I see a beautiful
piece of software, I want to remunerate the creators.  Be that the
gorgeously bump mapped characters in Halo 2, or the ultra slick
interface of FL studio, I have a sincere appreciation for the time and
effort that went into the production of that software, and a desire to
share some of the joy it's use has given me with the producers in a way
I know they'll appreciate --- cold, hard cash.

Functionality is just as important as looks, and that's another thing
which compels me to compensate the producer.  Even if we ruled out the
idea that I need to pay to play anyway, using a wonderful product
makes me want to shell out.  It's a way to ensure future output from
the author.

But what about when you have not-so-wonderful product to work with?
Paying for that is a risk then.  Assuming I didn't kill Specimen, if I
asked you for a $200 donation to support the future of Specimen, you've
got to hope that your cash is actually going to result in Specimen
becoming, like, totally sweet.  Considering that you don't really know
me, that's understandably a bit to swallow.  But it's worse than that
--- you don't know if anybody else is going to donate!  Your $200 might
be wasted if you have no way of knowing if anybody else will contribute.
$200 isn't even going to pay my rent --- if I only get your donation, I
can't exactly quit my day job.

The way I see it, if everyone had some way of knowing for absolute fact
that if they put in their $200, they will get an absolutely awesome
product, then there would be no problem with the quality of open source
software.  And a smarter person than myself might figure out a way to do
just that.  But as things are now, funding open source software via
donations is a bet on the users part. And so long as the funding problem
isn't solved, open source is not going to consistently lead the way.
The fact of the matter is, people got bills to pay.

There are other problems (not sure that's the right word), but this
strikes me as perhaps the most fundamental one.  If we figured out how
to end-run it, then everything else might be moot and open source might
rule the world. But so long as the problem exists, I don't think we're
ever going to be in a world where the technical merits of open source
beat proprietary for every possible task.  And while technical merits
aren't the whole deal, they are a signifcant part of it.

Goddamn I'm getting long winded in my old age.  There's something to
chew on. Like Paul's brains, because I'm a DAMN ZOMBIE!

-Pete



More information about the Linux-audio-dev mailing list