matt@group-w.uchicago.edu (Matt Crawford) (05/05/90)
I do hope this can be the last word on the subject. (But then, doesn't
everyone?) If Dan's willing, we'll confine this to alt.flame after this
point.
brnstnd@stealth writes:
) Matt Crawford has seen fit to take a personal vendetta to this group.
Vendetta? What have I ever done to you, Dan, except to argue with you?
Come to think of it, one of our previous arguments was about your
vocabulary. You didn't know what "proselytize" meant. Now you don't
know what "vendetta" means. (Oh, and didn't you lose an argument with
someone else about your misunderstanding of the legal effects of
compiling with GCC?)
) First, anyone who does not time-date copyright limitations is deluding
) himself and anyone receiving the copyrighted work. 17 USC 203(a) says
) that after a certain period of time an author can regain the exclusivity
) of his copyright by informing recipients that he intends to do so. In
) fact, (a)(5) says that it doesn't matter what the author has done to
) give away his rights; he can still get them back. Either Richard Stallman
) knows this and is tricking people into thinking that Emacs is truly free,
) or his lawyers haven't told him basic copyright law.
You are either confused yourself, or are deliberately obfuscating.
The Free Software Foundation never gave away its copyright on any of its
products, so the question of "regain[ing] the exclusivity" does not
enter. Are you confusing a license with a copyright? I know that you
know that there is a difference.
) Finally, once you've legally obtained a copy of software, it's yours.
) Nobody can take away your right to use it, to make backup copies, to
) make changes for your own use, to give it to a friend, or any of the
) other rights normally associated with ownership. Remember that.
I see. So what you're saying is, I can take my legally-obtained copy of
Macsyma, make a change so that it doesn't check the hostid and
expiration date against the secret password, and then give it to a
friend. Thanks. When I need a lawyer, I'll be sure not to call you.
) > By now many of you will have noticed some absurd restrictions placed in
) > the copyright notices on sources submitted by Dan Bernstein to
) > comp.sources.unix. The conditions include
) > o No rights to the recipient except those explicitly granted.
)
) That's a shameless lie, Matt.
OK, you tell me what "all rights reserved" means. You use that phrase
quite frequently in your code.
) As you should know, recipients are given quite a few rights under the
) copyright law.
And those are explicitly granted, are they not? But not by you. I
don't spend time reading law, as you seem to do. (And as you seem to
expect all of us to do.) So, having done all that research, why don't
you share the knowledge by enumerating some of those rights along with
your own legalese?
) Nowhere do I pretend that I'm *restricting* the recipient. In contrast,
) GPL section 4 is exactly such a pretense.
Funny use of "restrict". Between "all rights reserved" and "you have
the right to make and distribute exact copies", it sounds pretty
restrictive to me. No more restrictive than the minimum rights the law
itself grants, but far, far more restrictive than is the norm for
USENET.
GPL section 4 begins:
4. You may not copy, sublicense, distribute or transfer GNU Emacs
except as expressly provided under this License Agreement.
Is this a restriction which you do not make on your own work? If so,
then we can ignore your terms completely, can't we?
) > o The right to make and distribute exact copies, but not any
) > derivative forms, as long as no fee is charged.
)
) What's wrong with this? I just want people to check with me before
) distributing modified versions. That way I can find out who's using what
) versions, know where to report bugs in particular derivatives, etc.
If your product is good, so many people will use it that you won't
*want* to try to keep track of them. If it's good, people will want to
adapt it to oddities of their own system. You've told them they can't.
(But later you hint that just possibly you might grant them a special
exception.)
) > o Those rights, what few there are, are revocable upon
) > "written, oral, or other" notice, and are "automatically"
) > revoked on a certain date in the future.
)
) That's the legal protection that I talk about above.
Protection for whom? For me? No thank you. It makes a risk for me, if
I forget to delete your code from my disk. (Don't worry, I did that
already as soon as I read as far as your copyright and license terms.)
Protection for your code itself? If the code is any good, it will live
on its own merits without your legalese, and improved versions will
drive out obsolete versions.
Protection for you? Maybe in your own mind. You'll have to explain it
to me though. If the world is forced to delete "buggy" versions after 5
years, will your (imaginary) reputation for writing only bug-free code
thereby be protected?
) Matt, I'm not
) responsible for your failure to read or understand the copyright law;
) but I won't have people who use my programs similarly deluded.
Failure to read. I imagine that if I read it, I'd understand some of
it. How many potential users of your code do you expect to have read
copyright law? All of them? I mean, from reading your license terms,
it sounds like I don't even have the right to COMPILE the damn thing,
but you assure us in a separate article that some law gives us that
right.
) A few months ago, a poor soul confused by the proliferation of GPL-like
) unenforceable licenses wrote me asking why my ``license'' didn't give
) him permission to modify squeeze for his own use. As I told him, he's
) given that right by the copyright law.
Well then he wasn't really confused was he? Your license didn't give
him the right, the law did. The difference he actually saw was that
GPL-like licenses allow him to share his modifications but yours
doesn't. Would it be getting too close to the original point to claim
that the former is a better way to promote free software?
) It's a shame that inexperienced
) pseudo-lawyers like Matt who don't even bother reading the law have
) created an environment where such confusion can exist.
"Don't even bother..."??? Laws are not written to be read. A sad fact,
but true.
) As always, I solicit suggestions, questions, criticism, and any other
) comments.
But of course criticism may be construed as vendetta ...
) By the way, Matt, UUNET is not in violation of my ``nutty terms,'' nor
) is anyone else who is distributing exact copies for free.
UUNET forwards usenet for a FEE. Chinet lets people download your
software for a FEE. Portal is probably similar, and maybe The Well is
too.
Matt Crawford
vendetta (ven'det&schwa.). [a. Ital. vendetta:-L. vindicta vengeance.
Cf. VINDICTIVE a. ]
1 A family blood-feud, usually of a hereditary character, as customary
among the inhabitants of Corsica and parts of Italy.
2 A similar blood-feud, or prosecution of private revenge, in other
communities.
(Gracious, if he thinks it's a vendetta, no wonder he sounds paranoid!)brnstnd@stealth.acf.nyu.edu (05/05/90)
In this article I will restrict myself to exposing some of Matt's
failures in legal understanding, viz.:
1. The FSF has not given away any of the exclusivity of its copyrights.
2. The GPL is as enforceable as the Macsyma license.
3. The phrase ``all rights reserved'' takes away ownership rights.
4. Individual rights granted by statute must also be granted by me.
5. ``All rights reserved'' also takes away First Sale doctrine rights.
6. If not all rights are reserved, then no rights are reserved.
7. Copyright law prevents modification of a program for private use.
8. Notifying the consumer of legal pitfalls doesn't protect him.
9. You need explicit permission to compile a program you legally own.
10. Patches to a program are derivative works of that program.
11. If you forward communications without regard to their content, then
copyrights upon those communications apply to you.
Followups to misc.legal. Sorry, Matt, if you were expecting a response
to any of the rest of your article, but I don't see anything there that
deserves a response.
In article <983@gargoyle.uchicago.edu> matt@group-w.uchicago.edu (Matt Crawford) writes:
> brnstnd@stealth writes:
[ I point out that under 17 USC 203(a), an author can regain the ]
[ full exclusivity of his copyright after a period of time, no ]
[ matter what he's given away ]
> ) Either Richard Stallman
> ) knows this and is tricking people into thinking that Emacs is truly free,
> ) or his lawyers haven't told him basic copyright law.
> The Free Software Foundation never gave away its copyright on any of its
> products, so the question of "regain[ing] the exclusivity" does not
> enter.
With the GNU Public License, the FSF has given away some of the
exclusive rights of a copyright. To be sure, any restrictions upon the
recipient contained in that license are unenforceable; but any benefits
it provides to the recipient are perfectly valid.
> ) Finally, once you've legally obtained a copy of software, it's yours.
> ) Nobody can take away your right to use it, to make backup copies, to
> ) make changes for your own use, to give it to a friend, or any of the
> ) other rights normally associated with ownership. Remember that.
> I see. So what you're saying is, I can take my legally-obtained copy of
> Macsyma, make a change so that it doesn't check the hostid and
> expiration date against the secret password, and then give it to a
> friend.
No, you can't, because *before receipt* of Macsyma, your university
signed an agreement taking away some of its First Sale doctrine rights.
This has nothing to do with software that you pick up by anonymous ftp
or through USENET. We're talking about free distribution here.
> ) > By now many of you will have noticed some absurd restrictions placed in
> ) > the copyright notices on sources submitted by Dan Bernstein to
> ) > comp.sources.unix. The conditions include
> ) > o No rights to the recipient except those explicitly granted.
> ) That's a shameless lie, Matt.
> OK, you tell me what "all rights reserved" means.
It means that I'm protecting my default copying rights in certain South
American countries, where without that phrase a copyright notice is
simply invalid. Redundancy never hurts anyway.
> ) As you should know, recipients are given quite a few rights under the
> ) copyright law.
> And those are explicitly granted, are they not?
Yes, they are, by the copyright law.
Are you asking ``And those are explicitly granted by each and every
author, are they not?''
No, they are not. I don't have to tell you about your ownership rights
for you to have them. I have mentioned some of them here and in other
articles, and I'm sure you can take the time to consult your lawyer for
more.
Matt, I do not give you the right to life, liberty, or the pursuit of
happiness.
> GPL section 4 begins:
> 4. You may not copy, sublicense, distribute or transfer GNU Emacs
> except as expressly provided under this License Agreement.
> Is this a restriction which you do not make on your own work?
Yes, it is a restriction that I do not make on my work. You can, for
example, transfer my programs with all backup copies to a friend, even
if I take away all my copyright limitations. You can use my software
on your machine for as long as you wanted, even if I take away all my
copyright limitations. I never pretend that I can, should, or want to
restrict anyone's ownership rights.
> If so,
> then we can ignore your terms completely, can't we?
No, you can't, because copyright law says that you can't copy,
distribute, prepare derivative forms of, or publicly use a copyrighted
work unless I say you can. Just because you have full ownership rights
doesn't mean you have full copying rights.
> ) > o The right to make and distribute exact copies, but not any
> ) > derivative forms, as long as no fee is charged.
> ) What's wrong with this? I just want people to check with me before
> ) distributing modified versions. That way I can find out who's using what
> ) versions, know where to report bugs in particular derivatives, etc.
> If your product is good, so many people will use it that you won't
> *want* to try to keep track of them. If it's good, people will want to
> adapt it to oddities of their own system. You've told them they can't.
Wrong again. The copyright law says that anyone may modify any program
for his own use on any machine. And if he wants to distribute patches,
he can distribute the patches separately---provided they don't include
more of my code on the side than fair use would allow. I can't restrict
these rights, nor do I want to.
> ) > o Those rights, what few there are, are revocable upon
> ) > "written, oral, or other" notice, and are "automatically"
> ) > revoked on a certain date in the future.
> ) That's the legal protection that I talk about above.
> Protection for whom?
For the consumer, who might be deluded (as you are) by the GPL and
similar licenses into thinking that a limitation of copyright is
permanent, when in fact the law says it cannot be.
> it sounds like I don't even have the right to COMPILE the damn thing,
Did it ever occur to you that legislators adapt the law to take care of
all these sensible cases, so that authors shouldn't have to explain each
and every one? Of course you have the right to compile a program you
own. Duh.
> GPL-like licenses allow him to share his modifications but yours
> doesn't.
A patch is not a derived work. Are you confused by the GPL again?
Say I put together a package consisting of suggested code examples to
fill in some of the gaps in BSD PS2. The code makes no sense without
PS2; for best results, you should patch the package into the nroff
versions of the manuals and print them out again. Does this make the
package a derived work? Of course not. It contains none of the original
text.
Now, there are legal issues arising from the fact that some of the
original code is included in a context diff, so patches do have some
slight derivation from the original. As long as the amount of code in
question isn't significant, this should be safe under fair use criteria.
In other words, there's no basis under copyright law for restricting the
distribution of a patch, unless the patch includes a significant portion
of the original. Though I would like to see patches and if possible add
them to future versions, I don't worry about the problems of supporting
unofficially modified versions. After all, the patches are distributed
separately, rather than merged into the original; and so each recipient
will know the patches he's applied and I won't have to go hunting for
bugs in my code when they're really in someone else's. Doesn't this make
sense to you?
> ) By the way, Matt, UUNET is not in violation of my ``nutty terms,'' nor
> ) is anyone else who is distributing exact copies for free.
> UUNET forwards usenet for a FEE. Chinet lets people download your
> software for a FEE. Portal is probably similar, and maybe The Well is
> too.
But UUNET doesn't exert control over the contents of what passes through
any particular newsgroup over its paid communications lines, or over the
distribution of programs in its anonymous ftp archives. There's another
section of law that you haven't read saying, in effect, that copyrights
don't interact with such middlemen. Paying for an uncensored line has
nothing to do with paying for what comes over that line.
Have you ever been in apartment complexes or hotels where television
signals are relayed through a local center into each room? Where using
the television costs a lot more than electricity? It's exactly the same
situation: just because a commercial happens to be copyrighted doesn't
mean the hotel has to worry about that copyright.
---Danggw@wolves.uucp (Gregory G. Woodbury) (05/06/90)
In article <983@gargoyle.uchicago.edu> matt@group-w.uchicago.edu (Matt Crawford) writes: > >OK, you tell me what "all rights reserved" means. You use that phrase >quite frequently in your code. Jezeuz! The phrase "all rights reserved" is explicitly required so that the author doesn't loose rights under pan-american copyright. Without that phrase certain south american countries will take all sorts of liberties with the item being copyrighted. It also helps to limit the stuff that the government can do to the item if it is somehow transmitted to them. Talk about basic confusion. -- Gregory G. Woodbury @ The Wolves Den UNIX, Durham NC UUCP: ...dukcds!wolves!ggw ...mcnc!wolves!ggw [use the maps!] Domain: ggw@cds.duke.edu ggw%wolves@mcnc.mcnc.org [The line eater is a boojum snark! ] <standard disclaimers apply>