Subject: Re: copyright questions
To: Thomas Graichen <graichen@rzpd.de>
From: Chris G. Demetriou <cgd@pa.dec.com>
List: current-users
Date: 06/12/1997 11:56:26
[ This is, of course, not in any way the opinion of Digital Equipment
  Corporation, and the entire discussion of copyright on code that
  I own doen't in any way apply to code that they or any of my
  previous employers might own. ]

> i have a little question about the copyright's used in the NetBSD
> sources:
> 
> at the NetBSD homepage i find the following under the "goals":
> 
> ...
> In summary, the people involved in the NetBSD Project use a
> Berkeley-style license where possible because it closely matches our
> goal of allowing users to do whatever they'd like with our software,
> while making sure that we get credit for the work we have done.

Yup.


>  * Redistribution and use in source and binary forms, with or without
>  * modification, are permitted provided that the following conditions
>  * are met:
>  * 1. Redistributions of source code must retain the above copyright
>  *    notice, this list of conditions and the following disclaimer.
>  * 2. Redistributions in binary form must reproduce the above copyright
>  *    notice, this list of conditions and the following disclaimer in the
>  *    documentation and/or other materials provided with the distribution.

I.e., the copyright notice and license terms must go into a document
shipped with binary distributions (and it could be argued that in the
case of many free software projects, that the documents associated
with binary distributions are the source code).


>  * 3. All advertising materials mentioning features or use of this software
>  *    must display the following acknowledgement:
>  *      This product includes software developed by the University of
>  *      California, Berkeley and its contributors.

In other words, even if you write a critical piece of system software,
_unless advertising materials specifically mention features of use of
your software_ you don't get any credit.


> looks fine so far ... now let's have a look at the copyright of
> another file in the source tree - src/sys/arch/alpha/alpha/locore.s
> (it's not the only file with this style of copyright):
> 
> /*
>  * Copyright Notice:
>  *
>  * Copyright (c) 1997 Christopher G. Demetriou.  All rights reserved.
>  *
>  * License:
>  *
>  * [ ... ]
>  */

Ahh, you've run across my new license.

Yes, it requires credit in more cases, specifically because I tend to
write software which _is_ critical to the operation of (certain ports
of) the system, yet which never has its features or use specifically
mentioned.

If you use or distribute my software, I want credit for it.  period.
Since in the past, people have _not_ given me credit for it, I decided
to go to a lawyer and craft a more complete and correct (for my
purposes) set of license terms.

I _do_ intend to enforce that license (including registration of
copyright on large packages of new software I write).


> Yes, I _do_ plan to enforce the license, and yes I 
> i would be interested to hear what other people think about this and
> what are the reasons for the ones using this kind of licence not to
> stay with the berkeley style licence - is the credit you get from it
> not enough ? ... is it really required to do such stuff ?

The credit you get from the Berkeley license is very limited.

(1) you get your copyright and license terms put into the front of a
manual, or whatever, if you wrote software that contributes to the
product that the manual describes.  In practice, last I saw, _none_ of
the freely-redistributable BSD systems does this.

(2) you get an attribution string mentioned _IF_ "advertising
materials" mention features of use of your code.  However, as noted,
type of software that I write (typically base, low-level system
software) is _not_ the type of thing that people mention either
features or use of, and therefore I _don't_ get credit for it.


Also, the attribution clauses of the Berkeley license aren't very
enforceable.  There are too many gray areas.  (What is advertising,
for instance?  Print advertising, as was the original intent of the
license?  There was no WWW when the Berkeley license was written!)

The "don't remove this copyright notice" and "don't use our name for
advertising" are very enforceable, but they don't do anything to get
the author credit for their work.


In short, I did not think that I was getting appropriate credit for
the work that I was doing.

The _spirit_ of the berkeley license says "give copious credit
whenever anybody does work that you use," and that is indeed what I
do.

However, other people and organizations (e.g. OpenBSD, for example;
for a very long time, I wasn't listed as a a contributor to the
OpenBSD/alpha code, even though it's copied largely from my work)
don't do such a good job.  Quite often, no more credit than was
_legally_ required by the Berkeley license was being given, and given
the type of software I write, I decided that that was not satisfactory
for me.

Since "the spirit of the license" isn't really anything that can be
enforced, I had a lawyer formulate a new license which is much more
clear, has fewer grey areas, and which requires credit in the
situations where I think it's appropriate.

I OK'd it with 'core', and am now using it on all new code that I
write (that I own, of course 8-), with the exception of minor changes
to existing files and with the exception of changes that I contribute
to NetBSD or other groups (e.g. the FSF).  The new license terms still
allow modification and redistribution in source or binary form, they
are just more specific and complete about when they require credit.

FYI, some of the clauses, e.g. the requirement that the copyright
notice be included in the binaries, exist strictly to make the
copyright more enforceable.  I'm not keen on doing that sort of thing,
but it was strongly recommended by my lawyer.


You may not like it, but your actions, as one of the people working on
OpenBSD/alpha (which for the longest time _didn't_ give me credit
e.g. on the web pages) are in fact directly responsible for my new
license.  Now you have to live with it.

As noted above, I do intend to enforce the new license.  As noted in
the license terms, people who have a desire to distribute the software
under different terms (e.g. CD-ROM distributors or third parties who
want to distributed my code as part of their product) and who have
historically respected the spirit of the copyright I've distributed
software with can ask me for an exemption or special distribution
terms.

BTW, though i'm very "down" on the practices of some organizations,
I've had _very_ good experiences with others.  My change of license
should by no means indicate that all, or even most, groups are out
there to steal credit for code that people write.


Really this boils down to a simple truth.  If somebody does something
that makes you unhappy, you have two choices: be unhappy and live with
their actions now and in the future, or do what you can to prevent
future unhappiness from the same sort of cause.  I don't like to be
unhappy.  8-)



chris