comp.lang.ada
 help / color / mirror / Atom feed
From: "Marin David Condic" <marin.condic.auntie.spam@pacemicro.com>
Subject: Re: GPL and Plug-INs and XML\Ada
Date: Wed, 2 May 2001 16:46:54 -0400
Date: 2001-05-02T20:46:57+00:00	[thread overview]
Message-ID: <9cpro1$p8a$1@nh.pace.co.uk> (raw)
In-Reply-To: uy9sfvdaf.fsf@gsfc.nasa.gov

This would seem to be impossible to enforce. If its impossible to enforce,
its a good bet that it won't stand up in front of a judge.

If you had some sort of dynamic binding to something else that is not part
of your .EXE - no matter what the specific mechanism is - it would seem that
the license on the thing you bind to can't really impose something on your
code. (If it could, don't you think Bill Gates would have claimed ownership
of everything ever run on a Micro$oft OS?) If you make reference to someone
else's book while writing your own book, does this violate the other guy's
copyright?

That is to say, if you don't have a translation of the library included in
your .EXE and if you don't distribute (copy) the other guy's DLL onto your
disk, you have not violated his copyright. So claims based on copyright
don't seem to hold up - you don't really copy anything, right? (There may be
other problems - is the API to his DLL some sort of trade secret? etc? Or
could someone else "reverse engineer" it without violating any
restrictions?) If the copyright owner extends the right to you to copy and
distribute his DLL separately, then putting it on a separate disk - or on
the same disk - seems to be nothing but a splitting of hairs that makes no
difference. They remain separate products that function independently & can
be distributed separately.

Obviously, this gets significantly more complicated if you're talking about
something that *isn't* dynamically linked, such as compiler libraries &
runtime kernels - but that wasn't the original question. If I were the
judge, I'd look at such a situation on the basis that the compiler is pretty
well useless to a developer unless the developer can retain full ownership
of the product of his labor. Compiler writers used to put all kinds of
restrictions on their runtime code, but you don't see that anymore. Either
a) it wouldn't hold up in court (the "That dog don't hunt!" principle of
arbitrary judgement) or b) the American Free Enterprise System saw to it
that such compiler vendors got spanked for doing it.

It seems logical that calling a DLL couldn't impose an open source rule on
my code. Otherwise, any software that ran on Linux would have to be open
sourced because at the end of the day, Linux is really equivalent to a big,
giant DLL. If making Posix calls on a Linux development platform could force
open source on me, I bet there'd be a lot of developers in trouble right
now. :-)

Has there ever been a court challenge based on GPL'ed DLL's?

MDC

--
Marin David Condic
Senior Software Engineer
Pace Micro Technology Americas    www.pacemicro.com
Enabling the digital revolution
e-Mail:    marin.condic@pacemicro.com
Web:      http://www.mcondic.com/


"Stephen Leake" <stephen.a.leake.1@gsfc.nasa.gov> wrote in message
news:uy9sfvdaf.fsf@gsfc.nasa.gov...
> The key is the linking mechanism. I asked essentially the same
> question on a Windows CVS mailing list; apparently the Free Software
> Foundation position is that a "dll" (which stands for Dynamic Link
> Library) is _linked_, which is the term used in the GPL. So GPL dlls
> do "infect" the full app.






  reply	other threads:[~2001-05-02 20:46 UTC|newest]

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2001-05-01 18:35 GPL and Plug-INs and XML\Ada David Botton
2001-05-02  0:48 ` tmoran
2001-05-02  1:26 ` Frank Ranner
2001-05-02  8:31 ` Tarjei T. Jensen
2001-05-02 15:05 ` Ted Dennison
2001-05-02 18:03 ` Stephen Leake
2001-05-02 20:46   ` Marin David Condic [this message]
2001-05-03  2:07   ` tmoran
2001-05-03 14:32     ` Marin David Condic
replies disabled

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox