Re: Hardware licensing (fwd)

[graham@belegost.mit.edu]

From pmaupin@jump.net  Mon Oct  9 12:10:14 2000
Return-Path: <pmaupin@jump.net>
Received: from mail11.jump.net (mail11.jump.net [206.196.91.11])
	by belegost.mit.edu (8.9.3/8.9.3) with ESMTP id MAA21488
	for <graham@belegost.mit.edu>; Mon, 9 Oct 2000 12:10:13 -0400
Received: from jump.net ([216.30.74.184])
	by mail11.jump.net (8.10.2/) with ESMTP id e99GA9F17951;
	Mon, 9 Oct 2000 11:10:09 -0500 (CDT)
Message-ID: <39E1EE3E.44067833@jump.net>
Date: Mon, 09 Oct 2000 11:11:43 -0500
From: Patrick Maupin <pmaupin@jump.net>
X-Mailer: Mozilla 4.5 [en]C-CCK-MCD compaq  (Win98; U)
X-Accept-Language: en
MIME-Version: 1.0
To: graham@belegost.mit.edu
CC: rms@gnu.org
Subject: Re: Hardware licensing
References: <200010090921.FAA16753@belegost.mit.edu>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

>Is it ok with you and Patrick if I forward this exchange of emails
>to the hardlicense-discuss mailing list?

OK by me.  I didn't start there because I wasn't sure if I
was retreading old ground.  (After this email, I will start
copying the mailing list, but I'll let you post them in order
up to here.)

I think you and I are in agreement that:
a) the GPL as written doesn't provide any protection against
    people distributing hardware without distributing the underlying
    source code. (I think it's safe to say that Mr. Stallman also agrees
    with this position, since he may have been the first to articulate it.)
b) it might be possible to provide this protection anyway,
    if the license is worded properly, but we need to tread
    very carefully here, and get a good copyright lawyer to
    back this up.

However, I think the resultant license would be much stricter than
the GPL, but not in any practical sense.  In other words, the GPL
does not prohibit anything you might do in private with GPLed
code, but once you "distribute" you have to basically do a full
disclosure, if anybody cares to ask.

This sequence of events does not work for a hardware license,
because once the hardware is distributed, the cat is out of the
bag.  Instead, the licensee must be prohibited from making any
"transformations" to the code (modifications, compilations, etc.)
unless they agree with the license.  The hardware equivalent of
GPL clause 5 would state that they don't have to agree with the
license, but they have no right to make any modifications or
copies if they don't.  (Since it's not audio, and we're not talking
about consumer issues here, the "fair use" rights to use a whole
design are minimized, but as Mr. Stallman points out, once we
have our theories figured out, we certainly need to validate
them with a good copyright lawyer.)

Existence of hardware based on the code, in some sort
of violation of the license (e.g. without accompanying source),
would be prima facie evidence that the license had been violated
(because intermediate copies had to have been made during
development of the hardware).

Practically speaking, in some cases it would be impossible to
prove that a given piece of hardware was derived from a
given piece of source code; however, a business trying to
"get away" with a license violation would have to keep this
a secret, even though (in a typical design) dozens of employees
would have seen the source and would probably have realized
where it came from.  Most medium to big sized businesses
these days are fearful of legal and political fallout from violating
licenses, and they realize it takes only one disgruntled employee
to put them in really hot water.

Just as with the GPL, the real value would not be in having to
prove it in court, but to have it be believed that the risks of
violating the license (some chance it would stand up in court
plus community wrath) would outweigh the benefits.

Thanks,
Pat Maupin