From alsa-devel-owner@alsa.jcu.cz  Sat Feb 27 08:57:11 1999
Received: from neutrino.kek.jp (root@neutrino.kek.jp [130.87.218.139])
	by marvin.jcu.cz (8.9.1a/8.9.1) with ESMTP id IAA16874
	for <alsa-devel@alsa.jcu.cz>; Sat, 27 Feb 1999 08:56:57 +0100
Received: from lbscissors.kek.jp (root@lbscissors.kek.jp [130.87.218.173]) by neutrino.kek.jp (8.8.8+2.7Wbeta7/3.4W3) with ESMTP id QAA17398; Sat, 27 Feb 1999 16:56:40 +0900
Received: from lbscissors.kek.jp (really [127.0.0.1]) by neutrino.kek.jp
	via in.smtpd with esmtp (ident sharkey using rfc1413)
	id <m10Gec3-000sLNC@lbscissors.kek.jp> (Debian Smail3.2.0.102)
	for <alsa-devel@alsa.jcu.cz>; Sat, 27 Feb 1999 16:56:39 +0900 (JST) 
Message-Id: <m10Gec3-000sLNC@lbscissors.kek.jp>
To: alsa-devel@alsa.jcu.cz
Cc: kduffey@inprise.com
From: sharkey@ale.physics.sunysb.edu
Subject: Re: Trident's contribution, a draft 
In-reply-to: Your message of "Fri, 26 Feb 1999 11:29:27 PST."
             <3.0.32.19990226112926.009575f0@mail.inprise.com> 
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Date: Sat, 27 Feb 1999 16:56:39 +0900
Reply-To: alsa-devel@alsa.jcu.cz
Sender: alsa-devel-owner@alsa.jcu.cz
Precedence: list

> I also believe that while this is all good about some companies releasing
> source, we really need support for hardware accellerated soundcards like
> the SB Live, Diamond MX series, etc. I haven't seen much yet..just started
> with the group..but to be THE premiere system for delivering sound,
> supporting devices (soundcards, MIDI, digital audio etc), we really need
> solid hardware support. My suggestion, since the GPL doesnt permit
> non-source drivers, is to somehow release our specs (GPL should take care
> of this) so that other vendors, such as Creative, can write their own Linux
> drivers that will interface with ALSA.

The entire ALSA code-base is open.  Trident doesn't seem to have had any
difficulty writing an ALSA driver for their hardware, so I don't see why
Creative would need any more information than is currently available.

> This means they can freely
> distribute the driver (as they do on the wintel platform) but not have to
> give up their source code. The only downside is we can't package it with
> ALSA, but instead would have to point people to yet another site to
> download from.

No, that's not the only downside.  The other downside is that it will
discourage the creation of an open driver.

Binary drivers are not an acceptable solution to me and to many other people.
I've discussed the reasons for this previously on this list.  Check the
archives.  I'll send you a pointer if you can't find it.

If there is a binary driver, Creative will be that much more reluctant to
release the specs necessary for an open driver.  They'll point to the binary
and say, "Look, there's your driver.  You don't need our specs.".  The other
problem is that binary drivers are acceptable solutions for some people,
and it will decrease the demand for these specs, so for those of us for
whom binary drivers just don't cut it, we'll have a smaller mob behind
us when we storm the gate of Creative's castle.

No, this is not something we want to encourage.  We want to spit on Creative
and its driver.  Turn our backs and tell them that we simply are not interested
in playing that game.  If they don't want to release specs, fine, we don't
want to buy their cards.

> The good side is that we may get support for some hardware
> that otherwise just wont be supported because of the GPL. I love the GPL,
> but there has to be a way around some of the restrictions. 

A way around the restrictions?  Those restrictions are the very core of the
GPL.  They're there to help protect you and your freedom.  They're the reason
why we use the GPL in the first place.

If you don't want that kind of protection, then you don't use the GPL.  That's
what the BSD style licenses are for.
 
> If anyone could enlighten me if the above would be an acceptable way to
> allow support for devices that the companies dont want to release the
> source for, I would be apprecative of knowing. Thanks.

If companies don't want to release driver source, then they need to release
hardware interface specifications.  If they release neither source nor specs,
then they simply are not worth doing business with.

You're right, thought.  ALSA does need good support for quality hardware.
In the short term, it would be good if ALSA could accept these drivers, but
the long term detriments far outweigh those small benefits.

Eric


