Re: lynx-dev Re: Licensing Lynx (fwd)

[Klaus Weide]

On Sun, 26 Sep 1999, Brett Glass wrote:

> At 10:27 AM 9/26/99 +0100, David Woolley wrote:
> 
> >Unfortunately the copyright status is not clean as there is no single
> >legal person that owns the copyright.

The fact that there is no single legal person that owns the copyright
doesn't make the copyright status unclean (and neither unclear, in
case that is what you meant to write).  All it does is make the identity
of the owners difficult to determine.

That fact that not all such owners are easily findable by Mr Brett does
not detract from their rights.

> Worse still, the license under which Lynx is CLAIMED to be distributed
> cannot be applied to several parts of it.

Several parts??  You have shown *one* possible inconsistency.  Relating
to some code portions that were put into lynx, as best as I can determine
from the change logs, around 3-19-94 (i.e. under the responsibility
of the University of Kansas as the original Lynx copyright owner).
Maybe someone should look into this and resolve the apparent
inconsistency.

Now you'd have to show how this is relevant.  There isn't a problem
between Bellcore and any Lynx authors (nobody has attempted to "use
the name of Bellcore in advertising or publicity pertaining to
this material", to my knowledge).  There is no actual problem for
you, as a recipient of the program under the license, unless for
some strange reason you actually plan to use "Bellcore" in advertising.
There might be a problem for you in the role of distributor of copies
(or derived works), in that you want to know what conditions you have
to pass on to *your* recipients, but really there isn't: comply with
COPYING and with the Bellcore notice, and pass the same on, nothing
more is required of you.

De facto the license is not purely GPL, but GPL + Bellcore Clause
(where "Bellcore Clause" has no practical significance for any
realistic scenario, and applies to a small portion of code only).
That is the license the original copyright holder who placed the
Program under this License, the University of Kansas, in effect chose.
There are two flaws, (a) that the additional condition is not
mentioned in COPYING (or elsewhere except the source file), and (b)
that Lynx is said to be licensed under the GPL (instead of, more
correctly, "GPL-with-bellcore-clause".  Regrettable, but so what.
You still have a valid license for "copying, distribution, and
modification" given to you by the text in COPYING.  (Well, I do.)  If
you chose to advertize your code as "Contains Super-Cool Mailcap-
Reading Technology from Bellcore" or something stupid like that, you
would be in violation of the Bellcore condition (unless you had
cleared it with them), but not of the license in COPYING.

The text in COPYING forbids *You*, the recipient of the license, from
"imposing further restrictions".  It does not forbid this to the
original copyright owner.  He was free to choose the license as he
wished, of course.  The fact that he did so by unwisely and unclearly
combining the GPL text with something else does not make the license
invalid.  I don't see anything in COPYING that would make it invalid
in this case.

When you or I pass on lynx code (or derived work) without removing the
relevant code in HTInit.c and its associated notice, we are not
"imposing any further restrictions" in the sense of section 5.  We are
merely allowing "the recipient automatically receives a license from
the original licensor" to happen.  It is not our responsibility to
"fix" the original license should it be unwise or unclear, only to
abide by it.

Presumably additional authors to the lynx code since the University of
Kansas released it under their GPL-with-bellcore-clause license agree
to have their contributions fall under the same license.  Hypothetically
one of them might complain that he was mislead into thinking his code
contribution would be under a "pure" GPL.  Play this through if you like; 
I doubt you can come up with any real scenario that would result in
"there IS no license".

Regarding flaw (b), well we're at least in good company.  The license
under which the Linux kernel is provided is usually referred to as just
GPL, not GPL-with-special-provision-for-modules.

Anyway, you should address these questions to Michael Grobe
<address@hidden> as the file COPYHEADER advises you.  Maybe he can
clarify the license for you.  Of course you would have already
contacted him if you had a sincere interest.

> So, one might say, it is
> "out of control" -- there IS no license, really.

You can logically either take the view that the license is valid anyway.
So you are bound by it and reap its benefits.  Or you take the view that
there is no valid license.  Then copyright reverts to its default state,
"nothing else grants you permission to modify or distribute the Program or
or its derivative works.  These actions are prohibited by law if you do
not accept this License" (COPYING, 5.).  In no case would you gain any
additional rights.  So what's your point?


> To apply the word "commercial" to code which is GPLed is, alas,
> an oxymoron.

Only with a narrow-minded definition of the adjective "commercial".

> Anything distributed under the GPL cannot be commercial, because the
> entity that distributes it cannot license the code itself;
> it does not
> own it. It can only create copies of the code.

Only with a narrow-minded definition of the adjective "commercial" and
the verb "to license".

> it does not
> own it. 

You own (in the copyright sense) what you wrote, and that for which
ownership has been transferred to you.

> It can only create copies of the code.

You can create derived works.

    -----

I see no reason for furhter detailed responses, they have already been
given.  Moreover, Mr Glass is already intimately familar with this
topic and knows the answers, contrary to his feigned ignorance about
how projects such as this work and his "somebody must be in charge here"
incredulity.  Still, on the off chance that he might be more than a troll, 
and for the sake of argument:

So tell us more about your "new program for blind users" for which
"Lynx is currently the best available code".  What exactly is it
about?  Tell us about your company (the "we" in your messages), your
financing, your release schedule, how you are going to distribute and
market it.  (Why should you have to tell?  Well, remember, you were
trying to get a special license from me (as well as 324[*] others).  I
wish to know what you are going do with my code before I can consider
the question further.)

Since you are going to "as much as possible, seek grants and other
funding" in order to *pay* us for licensing code that you are already
licensed to use *without* payment, I'm sure your account of how you
get that grant money will be truly fascinating.  Let *them* give *you*
money so that *you* can pay *us* so that you'll end up with special
license rights and no obligation to share.  As opposed to, let them
pay *you* so that *you* can develop code that in the end will benefit
*you* and *us* and *everyone*.  Yeah, I'm sure foundations and stuff
will go for your approach.

[*] arbitrary guess.


    Klaus