Subject: Re: binary-only code is evil
To: Garrett D'Amore <garrett_damore@tadpole.com>
From: Greg A. Woods <woods@weird.com>
List: current-users
Date: 09/30/2006 18:50:20
--pgp-sign-Multipart_Sat_Sep_30_18:50:18_2006-1
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: quoted-printable

I'm jumping in late here, but at least with a more relevant subject
header, just to add a couple of points that don't seem to have been
stated clearly before....

At Sat, 02 Sep 2006 09:25:21 -0700,
Garrett D'Amore wrote:
>=20
> Of course I would prefer to have _source_, but I also recognize there
> are cases for binary blobs.  I believe ath(4) is one such case.  I've
> worked with another WIFI architecture that used blobs as well, and it
> worked out okay.
>=20
> My biggest complaint with blobs is lack of portability, [....]

Lack of portability is, quite simply, just another type of bug.

I.e. the main real-world problem with binary-only code is inability to
fix bugs, add and/or remove features, etc.

Free systems must allow their users to fix any and all bugs, including
those that affect or prevent portability.


> I totally disagree with any stance we take that prevents binary blobs
> from being used with the operating system.  It will prevent NetBSD from
> being used in a number of scenarios going forward.

Well it's pretty much impossible to _prevent_ binary code from being
used in an otherwise open system.  :-)

However no binary code should ever be distributed with the source of an
open-source system such as NetBSD, particularly not one with a primary
goal of being "extremely portable" (even when a BLOB driver, for
example, might actually be usable on every platform it's associated
hardware might ever be successfully attached to, and even when the
driver might be useful to boot the system -- custom install media can
easily be created by anyone, but not from the main source tree please).

(Note a BLOB that's not part of the system proper is a different matter,
for example firmware that's loaded into some device is OK, though of
course it's always nice to have source and development tools for such
firmware as well.)


> If folks want NetBSD to thrive, the best way to do so is not to go
> around trying to piss off vendors.

Do you really think that refusing binary-only code contributions from
vendors will make them angry?  Do you really think that not buying
unsupportable hardware from such vendors will make them angry?

Do you really think that angry proprietary-only hardware vendors can in
any way stifle any open-source operating system project?

--=20
						Greg A. Woods

H:+1 416 218-0098 W:+1 416 489-5852 x122 VE3TCP RoboHack <woods@robohack.ca>
Planix, Inc. <woods@planix.com>       Secrets of the Weird <woods@weird.com>

--pgp-sign-Multipart_Sat_Sep_30_18:50:18_2006-1
Content-Type: application/pgp-signature
Content-Transfer-Encoding: 7bit

-----BEGIN PGP SIGNATURE-----
Version: PGPfreeware 5.0i for non-commercial use
MessageID: xujdcSPZ+kkvWxEH3vGh9oC8DiFniu4G

iQA/AwUBRR70q2J7XxTCWceFEQK9EACgsNnAcUWCbAjFDFGSM38eicB5ZNMAnRvL
uWDSAGy5YPFo2D1D7oGQPzrl
=e5X3
-----END PGP SIGNATURE-----

--pgp-sign-Multipart_Sat_Sep_30_18:50:18_2006-1--