Subject: Re: what happened to the lm75(?) driver?
To: None <tech-kern@netbsd.org>
From: Greg A. Woods <woods@most.weird.com>
List: tech-kern
Date: 09/12/1999 23:19:48
[ On Monday, September 13, 1999 at 11:58:52 (+1000), Julian Assange wrote: ]
> Subject: Re: what happened to the lm75(?) driver?
>
> I agree strongly with this. However, as your license is functionally
> the BSD license, so why not simply use the original BSD license?

Well, actually, it's not quite the same and though the differences are
perhaps subtle I feel that they're very important.  (I think my wording
is far superior too, but then I'm not a lawyer!  ;-)

> The problem for third party developers is that they have to legal every
> damned one of these things. Now while yet-one-more isn't going to bring
> down civilization as we know it, clearly having to deal with thousands
> of varying copyrights is non-optimal.

That's their job -- i.e. if they want to use someone else's code for
free and they want to be sure they do so with good legal footing then
indeed they do have to be sure to verify that each and every copyright
applicable to that code permits them to use it for free.  TANSTAAFL

Of course any reasonably sane person with a working knowledge of the
copyright law in their jurisdiction should be able to spot any
questionable copyright licenses very quickly and ensure that only those
were issues are likely to develop are verified by their much more
expensive legal team.

The physical cost of providing the individual text of my particular
copyright notice is negligible, at least so long as either manual page
sources and/or header files are included, intact, in the final
distribution.  I realize that the complete copyright can't be included
in the .o files, and that it's prohibitive to try to include the
complete copyright in the documentation, but I've not (so far as I can
tell) required that it be (and if the wording of my terms does in fact
seem make such a requirement then that was not my intention and
hopefully my wording can be fixed).

The only thing I see that I'm asking for that's out of the ordinary and
that's not already done for the distributor by the files I'm already
providing is that a one-line acknowledgment be included in some (any!)
form in the final distribution.

If the issue is that the TNF is trying to take legal responsibility for
verifying the conditions of each individual contributor's unique
copyright license have been met and that the result is a distribution
that's freely redistributable then might I suggest that they're taking
the bull by the horns without having the necessary clowns in the wings
to rescue them.  TNF should only ensure that their own distribution is
legal and that's it.  If any third party distributor wants to have TNF
assist in verifying the re-distributability of NetBSD then they should
pay the legal bills TNF would accrue in doing this verification
otherwise they must do this verification all by themselves and TNF can
only offer guidelines and non-legally binding advice as to where the
snakes may lay in the grass.

> On the other hand, the situation is so bad as it is, I'd be inclined
> to just give up entirely on uniform copyright adoption. The battle is
> lost. If we accept that we can move forward in an more inclusive way,
> concentrating on features and not playing amateur lawyer.

TNF could completely and totally cover all of our butts by making the
computer do the dirty work for us and making it possible to have it
collect all copyright notices from all files, merging them down to a
collection of unique notices and a list of which files each pertains to.
This wouldn't have to be AI and the result would be a canonical
"copyright" declaration for the entire distribution.  If TNF took the
repsonsibility to do this completely then third-party developers could
simply crib the entire list and should not have to do any additional
legal leg work.

I still think though that what I've asked for in the form of a one-line
entry in a list of contributors, as well as keeping my unique copyright
intact in every source file, is a lot more fair and somewhat more
generous than what TNF is asking for.  I wouldn't want to begrudge
anyone else from asking for more, but perhaps my terms are indeed a good
middle ground and perhaps they can be used as the basis for a comromise
on this issue.  I don't believe that the original BSD license with just
a name substitution will serve my needs exactly.

If we are really talking about actually having uniform copyright terms
across the board on every file then that's an entirely different matter.

I.e. I'm trying to say that if a uniform third-party copyright will make
NetBSD more palatable to third-party use then TNF should be paying a
very good IP lawyer to write a very solid copyright license that can
meet all of my needs as well as those of every other contributor.  If
TNF can offer a template for a license that clearly and obviously meets
my needs then I'll be more than happy to use it.  This is what I've been
asking all along.  So far all of the proposed templates have missed one
or more very important terms.  BTW, if *I* would have to pay a lawyer to
OK such a license for my own code then I suspect it would not be a good
enough license for everyone either.

Please everyone let's be crystal clear about exactly what's required
here.  If it's a uniform copyright license then let's come up with one
that's acceptable by consensus with *all* contributors and I'll be more
than happy to assist in getting everyone to switch to it.  If it's just
one or more terms in my specific coypright license that are
objectionable then please say exactly that.  I get the distinct feeling
that there's no public consensus at all yet on what's really required to
make NetBSD's over-all copyright status acceptable.

-- 
							Greg A. Woods

+1 416 218-0098      VE3TCP      <gwoods@acm.org>      <robohack!woods>
Planix, Inc. <woods@planix.com>; Secrets of the Weird <woods@weird.com>