What would people think of binary-only software on Linux?
Ken Arromdee
[quoting rms]
>People often speculate about technical procedures that might
>circumvent the GPL in some way. For example, they may suggest a
>modified version could be cut artificially into two pieces, one free
>and one proprietary, that are called two independent programs.
>This kind of scheme is based on the premise that the legal system
>operates in the fashion of a stupid computer program, and that
>superficial manipulations of the way files are grouped and labeled
>would fool it. While the legal system often does seem stupid and
>easily fooled in comparison with common sense, the FSF's lawyer told
>us that it would not be stupid about this.
>The lawyer said that such a scheme would fail because a judge would
>regard it as a subterfuge. The judge would conclude that the two
>parts are really one program in disguise, and go on from there.
Remember this is a double-edged sword.
For instance, the GPL is supposedly not a shrink-wrap license. Since you
did not sign it, you are not bound by it; however, the only thing that allows
you to make copies of GPL software at all is the license, so if you choose
not to be bound by it, you cannot legally copy GPL software.
However, according to rms, the legal system does not work like a computer
program in that you cannot create a B that works like A without falling under
the laws concerning A. Even though B does not literally fit the exact wording
of laws regulating A, B was clearly _intended_ to really be an attempt at doing
A while circumventing the law.
If rms is right here, there is no reason why this should apply only in cases
where the laws benefit the FSF. A judge could just as well conclude that the
GPL, although not literally what the law considers a shrink-wrap license, is
really a subterfuge. The GPL is really a shrink-wrap license in disguise, and
the judge would go on from there.
If "technical procedures" to get around the GPL are unacceptable because the
law recognizes subterfuge, why isn't the GPL itself also unacceptable because
it uses a subterfuge to get around the illegality of a shrink-wrap license?
--
"On the first day after Christmas my truelove served to me... Leftover Turkey!
On the second day after Christmas my truelove served to me... Turkey Casserole
that she made from Leftover Turkey.
[days 3-4 deleted] ... Flaming Turkey Wings! ...
-- Pizza Hut commercial (and M*tlu/A*gic bait)
Ken Arromdee (arro...@jyusenkyou.cs.jhu.edu, arro...@jhunix.hcf.jhu.edu)
Ross Biro
For those in gnu.misc.discuss: The question is weather or not it is ok
to distribute proprietary .o files to be linked into the Linux kernel
(a GPL covered work.) I am arguing that it is not.
In article <1mqu7c...@senator-bedfellow.MIT.EDU> j...@athena.mit.edu (John F Carr) writes:
>In article <1993Feb28.1...@leland.Stanford.EDU>
> bi...@leland.Stanford.EDU (Ross Biro) writes:
>
>> The GPL has never been tested in a court of law, so legally it
>>is an unknown. I do know the intent of the GPL is to include object
>>modules whose primary purpose is to be linked into a GPL'ed package.
>
>Authors of GPL software can claim that intent, and I'm sure RMS would
>appreciate it if people acted that way, but there is no way that my
>software, not derived from other copyrighted source, can be held to the
>terms of the GPL, the legal authority of which derives from copyright.
Here is the relavent sections of the GPL
These requirements apply to the modified work as a whole. If
identifiable sections of that work are not derived from the Program,
and can be reasonably considered independent and separate works in
themselves, then this License, and its terms, do not apply to those
sections when you distribute them as separate works. But when you
distribute the same sections as part of a whole which is a work based
on the Program, the distribution of the whole must be on the terms of
this License, whose permissions for other licensees extend to the
entire whole, and thus to each and every part regardless of who wrote it.
Thus, it is not the intent of this section to claim rights or contest
your rights to work written entirely by you; rather, the intent is to
exercise the right to control the distribution of derivative or
collective works based on the Program.
In addition, mere aggregation of another work not based on the Program
with the Program (or with a work based on the Program) on a volume of
a storage or distribution medium does not bring the other work under
the scope of this License.
...
3. You may copy and distribute the Program (or a work based on it,
under Section 2) in object code or executable form under the terms of
Sections 1 and 2 above provided that you also do one of the following:
a) Accompany it with the complete corresponding machine-readable
source code, which must be distributed under the terms of Sections
1 and 2 above on a medium customarily used for software interchange; or,
b) Accompany it with a written offer, valid for at least three
years, to give any third party, for a charge no more than your
cost of physically performing source distribution, a complete
machine-readable copy of the corresponding source code, to be
distributed under the terms of Sections 1 and 2 above on a medium
customarily used for software interchange; or,
c) Accompany it with the information you received as to the offer
to distribute corresponding source code. (This alternative is
allowed only for noncommercial distribution and only if you
received the program in object code or executable form with such
an offer, in accord with Subsection b above.)
...
4. You may not copy, modify, sublicense, or distribute the Program
except as expressly provided under this License. Any attempt
otherwise to copy, modify, sublicense or distribute the Program is
void, and will automatically terminate your rights under this License.
However, parties who have received copies, or rights, from you under
this License will not have their licenses terminated so long as such
parties remain in full compliance.
....
7. If, as a consequence of a court judgment or allegation of patent
infringement or for any other reason (not limited to patent issues),
conditions are imposed on you (whether by court order, agreement or
otherwise) that contradict the conditions of this License, they do not
excuse you from the conditions of this License. If you cannot
distribute so as to satisfy simultaneously your obligations under this
License and any other pertinent obligations, then as a consequence you
may not distribute the Program at all. For example, if a patent
license would not permit royalty-free redistribution of the Program by
all those who receive copies directly or indirectly through you, then
the only way you could satisfy both it and this License would be to
refrain entirely from distribution of the Program.
End of GPL quotes
So it looks as if distributing proprietary .o files is acceptable.
However linking them in (being a form of copying) is questionable at
best. I would say that since it makes a copy of GPL covered code
which cannot be redistributed is a violation of the GPL under section
3, and hence makes the .o distributions worthless.
As for the Linux kernel, I may change the licensing agreement on
all modules which I've contributed to explicitly disallow linking to
proprietary modules. This would not affect modules that have been
distributed in the past, but it probably would affect ones in the
future.
Ross Biro bi...@leland.stanford.edu
Member League for Programming Freedom (LPF)
mail l...@uunet.uu.net to protect your Freedom
Joel M. Hoffman
real question is whether or the the GPL has any legal standing. The
point is, I cannot make up an arbitrary restriction on my sortware,
and expect you to abide by it. Copyright law applies to everything
that is copyrighted. Other than that, both parties have to agree.
For example, what if I publish software and clearly indicate on it that
"You may use this program exactly once. If you use it more, you agree
to pay the purchase price of $1,000,000 upon demand."?
You STILL incur no obligation unless you sign a contract indicating
your approval of the conditions.
While I am very much in favor of the GPL, I wonder if it is really a
legal contract, or a request.
-Joel
Matthew J Brown
A section of the GPL (version 2) I consider relevant:
> 6. You are not required to accept this License, since you have not
> signed it. However, nothing else grants you permission to modify
> or distribute the Program or its derivative works. These actions
> are prohibited by law if you do not accept this License.
> Therefore, by modifying or distributing the Program (or any work
> based on the Program), you indicate your acceptance of this
> License to do so, and all its terms and conditions for copying,
> distributing or modifying the Program or works based on it.
In other words, it is trying to work by virtue of the fact that you
must either accept the licence, or not; you cannot accept parts of it,
and refuse to accept others. Only if you consider the GPL binding do
you have the right to distribute software under it. If you break the
terms of the licence, and then distribute software under it, the FSF
can sue, not for breach of contract, but for copyright violation.
Whether this is a valid technique or not, I'm not sure; but the FSF
employs good lawyers.
-Morven
--
| Matthew J. Brown | Dept. of Computing | If God intended for us to go to |
| m...@doc.ic.ac.uk | Imperial College, | lectures He wouldn't have created |
| mj...@cc.ic.ac.uk | 180 Queen's Gate | double-sided photocopiers. |
| Morven on Lambda | LONDON SW7 2AZ | -IC RagMag 1991/92 |
Joe Buck
>I fear, however, that the present discussion misses the point. The
>real question is whether or the the GPL has any legal standing. The
>point is, I cannot make up an arbitrary restriction on my sortware,
>and expect you to abide by it. Copyright law applies to everything
>that is copyrighted. Other than that, both parties have to agree.
This point is true, but it is addressed in the GPL, as follows:
-----------
5. You are not required to accept this License, since you have not
signed it. However, nothing else grants you permission to modify or
distribute the Program or its derivative works. These actions are
prohibited by law if you do not accept this License. Therefore, by
modifying or distributing the Program (or any work based on the
Program), you indicate your acceptance of this License to do so, and
all its terms and conditions for copying, distributing or modifying
the Program or works based on it.
---------
>For example, what if I publish software and clearly indicate on it that
>"You may use this program exactly once. If you use it more, you agree
>to pay the purchase price of $1,000,000 upon demand."?
This would be invalid. However, the point in the GPL above is correct:
FSF owns the copyright; you may not copy GPL'd software or derivative
work of GPL'd software without their permission.
Now, one could argue that the phrase "you indicate your acceptance" is
not quite right, but either you accept it when you copy, distribute, or
modify or you don't, and in the latter case you have broken the law.
>While I am very much in favor of the GPL, I wonder if it is really a
>legal contract, or a request.
It is not a legal contract, because you have not signed it. It is not a
request, because it is a legal document, specifying under what conditions
FSF will give up its legal monopoly on making copies of GNU software.
--
Joe Buck jb...@ohm.berkeley.edu
Charles Hannum
In article <1mr6fb$k...@agate.berkeley.edu> jb...@forney.berkeley.edu
(Joe Buck) writes:
>
> FSF owns the copyright; you may not copy GPL'd software or derivative
> work of GPL'd software without their permission.
A not-so-minor nit: The FSF does not own the copyright on all software
distributed under the terms of the GPL.
--
\ / Charles Hannum, myc...@ai.mit.edu
/\ \ PGP public key available on request. MIME, AMS, NextMail accepted.
Scheme White heterosexual atheist male (WHAM) pride!
Ross Biro
>bi...@leland.Stanford.EDU (Ross Biro) writes:
>>For those in gnu.misc.discuss: The question is weather or not it is ok
>>to distribute proprietary .o files to be linked into the Linux kernel
>>(a GPL covered work.) I am arguing that it is not.
>
>license by using my software.
>
>Assume I do the following:
>
> 1. Write a device driver in C for Linux. I use my own header
> files rather than any Linux header files.
>
> 2. Compile this with Metaware C under DOS.
>
> 3. Run a program of my own that converts object files from
> Metaware's format to Linux's format.
>
>I'm not using any GPL'ed software, and so have not had to agree to GPL
>for anything I've done.
>
>Assume you do the following:
>
> 1. Obtain Linux.
>
> 2. Use it.
>
>You have agreed to GPL and you are bound by it.
>
>Now, I post the binary of my object file to the net. You grab it, and link
>it into Linux.
>
>*You* are still bound by GPL. *You* cannot distribute your resulting version
>of Linux. If you do, and someone asks for source, and you can't give it
>to them, it's you who will get sued[1], not me. The worst that will happen
>to me, legally, is I might be called as a witness.
>
>If the owners of the Linux copyrights wanted to prevent the above
>scenerio, they would have to do something like claim a copyright
>on the driver interface itself.
I maintain that the act of linking the proprietary .o file into
the GPL'ed kernel would be a violation of the GPL. Otherwise I could get
around the GPL by simply distributing my derived work in three parts.
1) gpl.o all the gpl'ed parts ( I would include source for this.)
2) proprietary.o (the proprietary part which you are not
aloud to re-distribute.)
3) an install script to link them together.
This is clearly a violation of the GPL as RMS has mentioned.
Furthermore since I would be providing materials whose sole use and
purpose is to help people to violate a contract, if the copyright
holders on the GPL'ed software didn't sue me, my customers would.
I think one of the best arguments that this is the case is that Next,
Lucid and all the other companies which have modified GPL'ed software
have not tried it. Furthermore this is exactly the case that RMS
talked about in the message that was posted earlier.
Basically the benefits to be gained by attempting to get around the
GPL in such a fasion are not worth the risk.
Tim Smith
>For those in gnu.misc.discuss: The question is weather or not it is ok
>to distribute proprietary .o files to be linked into the Linux kernel
>(a GPL covered work.) I am arguing that it is not.
Your mistake is an implicit assumption that you can put me under a
license by using my software.
Assume I do the following:
1. Write a device driver in C for Linux. I use my own header
files rather than any Linux header files.
2. Compile this with Metaware C under DOS.
3. Run a program of my own that converts object files from
Metaware's format to Linux's format.
I'm not using any GPL'ed software, and so have not had to agree to GPL
for anything I've done.
Assume you do the following:
1. Obtain Linux.
2. Use it.
You have agreed to GPL and you are bound by it.
Now, I post the binary of my object file to the net. You grab it, and link
it into Linux.
*You* are still bound by GPL. *You* cannot distribute your resulting version
of Linux. If you do, and someone asks for source, and you can't give it
to them, it's you who will get sued[1], not me. The worst that will happen
to me, legally, is I might be called as a witness.
If the owners of the Linux copyrights wanted to prevent the above
scenerio, they would have to do something like claim a copyright
on the driver interface itself.
[1] OK, maybe they would sue me. You can sue anyone for anything, pretty
much. It wouldn't get to trial, though. It would be thrown out after
the first round of interrogatories.
--Tim Smith
Tim Smith
>"You may use this program exactly once. If you use it more, you agree
>to pay the purchase price of $1,000,000 upon demand."?
>
>You STILL incur no obligation unless you sign a contract indicating
>your approval of the conditions.
You example probably would not stand up. However, GPL would almost
certainly stand up.
(Let's assume we are talking about something whose copyright is held by
FSF, so that I can type "FSF" rather than "the copyright owner" from now
on...)
Copyright gives FSF the right to control distribution. In the absence of
an agreement to the contrary, you have no rights to copy FSF software.
The GPL is *NOT* restricting you from doing something that you otherwise
would not be able to do. Rather, it is saying that FSF will *NOT* enforce
a legal right of theirs (to prevent you from copying their work) if you
don't do anything to violate the GPL. It's a restriction on them. If
you violate GPL, they won't sue you for violating GPL -- they will sue
you for violating copyright.
As a contract, it may or may not turn out that GPL is valid. However,
as a practical matter, FSF knows, or should know, that users rely on
the promise made in GPL by FSF to not sue if you follow GPL, and I
would expect that this would prevent FSF from suing someone who
follows GPL. (This is called "promissory estoppel").
In summary, you obligation to FSF comes from copyright law. GPL is
FSF's way of releasing you from part of that obligation.
--Tim Smith
Tim Smith
>In other words, it is trying to work by virtue of the fact that you
>must either accept the licence, or not; you cannot accept parts of it,
>and refuse to accept others. Only if you consider the GPL binding do
>you have the right to distribute software under it. If you break the
>terms of the licence, and then distribute software under it, the FSF
>can sue, not for breach of contract, but for copyright violation.
>
>Whether this is a valid technique or not, I'm not sure; but the FSF
>employs good lawyers.
I think it's better to not even try to think of GPL in terms of
it being a license or a contract. Think of it as a statement of
under what conditions the FSF (or whoever) will not sue you to
enforce their copyright rights.
--Tim Smith
Tim Smith
>It is not a legal contract, because you have not signed it.
Most legal contracts are not signed. Most are not even written down.
--Tim Smith
K.X. Saunders
<where the laws benefit the FSF. A judge could just as well conclude that the
<GPL, although not literally what the law considers a shrink-wrap license, is
<really a subterfuge. The GPL is really a shrink-wrap license in disguise, and
<the judge would go on from there.
<
<If "technical procedures" to get around the GPL are unacceptable because the
<law recognizes subterfuge, why isn't the GPL itself also unacceptable because
<it uses a subterfuge to get around the illegality of a shrink-wrap license?
This is all well and good, but you did not indicate how the GPL is a
SWL in disguise. A SWL is what it is since the user has to accept the terms
of the license in order to be able to read those terms. The GPL is not like
that. A user is quite able to read its terms without being bound by them.
- Kyle
--
_,--\_/>> kxs...@ultb.isc.rit.edu // Amiga weenie
`"""-(_-\ ky...@nick.csh.rit.edu \\ //
_,// \_\ \X/
*""' "" not representing Phoenix Systems Inc.
Tim Smith
> I maintain that the act of linking the proprietary .o file into
>the GPL'ed kernel would be a violation of the GPL. Otherwise I could get
Under my scenerio, *I'm* not linking, so even under your interpretation,
*I'm* not violating the GPL. (By the way, we don't even have to look
at GPL to see this. If I'm not using GPL'ed code, then it has no
power over me whatsoever. Person A and person B can't make a contract
or license that tells person C what to do unless they already have some
legal power over person C.)
--Tim Smith
Michael Golan
>I think it's better to not even try to think of GPL in terms of
>it being a license or a contract. Think of it as a statement of
>under what conditions the FSF (or whoever) will not sue you to
>enforce their copyright rights.
Oh give me a break. The FSF can sue anyone for copyright violations,
but they are sure to loose. Especially if a wierd document like the GPL
is behind. the FSF software is freely distributed. You don't need to
read the GPL to know this, they keep flashing it on the screen. As such,
limiting it in funny ways, or sueing only specific people is unlikely
to work. Just imagine if the GPL said that anyone can distribute the
software, except for young black males. Do you think a judge would
allow the FSF to sue a young black mail for not following the GPL?
[note: there is no special discrimination law against this. the copyright
law gives you some control over distribution, but not in ridicules way
like this (and most say, like the GPL)]
The FSF never sued anyone and probably never will. I wonder if they even
registered the work with the copyright office. The legal procedure is
far too expensive, and besides, what would the FSF want, money? their
intent is not to make money, but make software widely avaialable ...
making the source free? no judge would order such specific results ...
I like RMS's claim that the GPL, even if not legally binding, achieves
its goal. As if other public-domain or free software doesnt! The only
one being stopped down by the GPL are large companies that refuse
any legal hassle[1], and "decent people" who respect the GPL. They loose
the most, having to reimplement parts of GPL code on their own, for use
in commerical software (e.g. like a piece of code to manange regex
that they are already familiar with, because they have contributed to
the GNU project!).
-- Michael Golan
m...@cs.princeton.edu
[1] It is foolish to think that large companies, e.g. AT&T, being stopped
by the GPL is good. You & I don't like AT&T, but slowing them down like
this hurt society in the long run (which pay higher phone bills). If you
think AT&T is using bad legal practices, etc, use your VOTE to get the
government to regulate them better. For example, today's AT&T TV ad claimed
that other phone companies loose twice as many 800 calls. I'll bet thats
like 2 our of 1000 compared to 1 of 1000. Not very fair. The FTC can and
should stop such things (see latest Supreme court rulling re Kraft cheese),
Ken Arromdee
><where the laws benefit the FSF. A judge could just as well conclude that the
><GPL, although not literally what the law considers a shrink-wrap license, is
><really a subterfuge. The GPL is really a shrink-wrap license in disguise, and
><the judge would go on from there.
><If "technical procedures" to get around the GPL are unacceptable because the
><law recognizes subterfuge, why isn't the GPL itself also unacceptable because
><it uses a subterfuge to get around the illegality of a shrink-wrap license?
> This is all well and good, but you did not indicate how the GPL is a
>SWL in disguise. A SWL is what it is since the user has to accept the terms
>of the license in order to be able to read those terms. The GPL is not like
>that. A user is quite able to read its terms without being bound by them.
You are pointing out differences between the GPL and a shrink-wrap license.
The GPL does not, indeed, fit the definition of a shrink-wrap license, so of
course there are differences. But, according to rms's article, this is not
enough: the legal system does not work like a dumb computer program where you
are exempt from laws about something just because you don't literally fit the
definition embodied in those laws. So if rms is right, the fact that the GPL
does not fit the definition of a shrink-wrap license does not mean that it
might not be legally treated as one anyway.
--
"On the first day after Christmas my truelove served to me... Leftover Turkey!
On the second day after Christmas my truelove served to me... Turkey Casserole
that she made from Leftover Turkey.
[days 3-4 deleted] ... Flaming Turkey Wings! ...
-- Pizza Hut commercial (and M*tlu/A*gic bait)
Ken Arromdee (arro...@jyusenkyou.cs.jhu.edu)
John L. Coolidge
>In article <1993Feb28.2...@blaze.cs.jhu.edu> arro...@jyusenkyou.cs.jhu.edu (Ken Arromdee) writes:
><where the laws benefit the FSF. A judge could just as well conclude that the
><GPL, although not literally what the law considers a shrink-wrap license, is
><really a subterfuge. The GPL is really a shrink-wrap license in disguise, and
><the judge would go on from there.
> This is all well and good, but you did not indicate how the GPL is a
>SWL in disguise. A SWL is what it is since the user has to accept the terms
>of the license in order to be able to read those terms. The GPL is not like
>that. A user is quite able to read its terms without being bound by them.
I believe that this is an poor definition of a SWL. A SWL is a license
in which cannot view the license before a sale has occurred. Thus,
many people hold that SWL's are invalid because a sales transation (a
different, and unlicensed, form of transaction) has occurred; the SWL
would be an invalid post-sales contract, unsupported by consideration.
The GPL is similar to a SWL, but it doesn't meet my definition above
(nor yours). However, there is a _serious_ flaw in the GPL reasoning
that accepting the GPL is the only thing allowing you to copy the
program -- the initial copying phase of the program. When I ftp to a
server and grab emacs (for example), I am making a copy there and
then. At this point I cannot necessarily be presumed to have read the
GPL. Thus, downloading a GPL-protected program may be an illegal act,
since I cannot be presumed to have accepted a license I have never
read. The illegality of the act of downloading cannot be removed by a
later acceptance of the license.
Note, also, that this clears the way for the 'user-does-the-link'
workaround. The scenario is: person A creates proprietary.o which
must be linked with a GPL-protected program to work. Person A then
distributes proprietary.o (presumably for some cost), with or without
GPLProgram. Person B obtains proprietary.o and GPLProgram. Assuming
person B has legally obtained proprietary.o and GPLProgram, there is
no legal way to bar person B from creating a derivative work _for
their own use_. Person B, by both the GPL and (presumably)
proprietary.o's copyright and/or license, cannot redistribute their
derivative work, but they are legally entitled to use it themselves.
There is no way, under the GPL, to stop person A from distributing
GPLProgram along with proprietary.o (the mere aggregation clause).
While RMS has indicated that such a technical workaround would be
viewed by the courts as an attempt to sidestep the intent of the GPL,
there really is no way for the court to rule in favor of the FSF.
Every step is _entirely_ legal. There is simply no room for a court to
make a ruling against person A -- they have broken no laws, and they
have never distributed a GPL-derivative work, so proprietary.o cannot
have fallen under the GPL. Person B has also acted in an entirely
legal way. Indeed, if the FSF did indeed sue either person A or B,
it's perfectly likely that a countersuit for interference would
succeed.
The only way for the FSF's viewpoint to succeed here is for the
internal interfaces of GPLProgram to be legally protected. Then
proprietary.o can be considered a derivative of those interfaces and
thus it would fall under the GPL. The GPL itself seems to indicate
that this might be the FSF's view. However, it seems implausible to me
that the FSF would really be in favor of a system where the interfaces
of a software program could be considered protected works. Such a
system would (for instance) have allowed AT&T to have copyrighted the
interface to UNIX and prevented the GNU project in the first place.
Disclaimer: I am not a lawyer. However, I've had some good contract
law classes and I've looked at a number of court decisions in this
area.
--John
The fundamental principle of science, the definition almost,
is this: the sole test of the validity of any idea is
experiment. -- Richard P. Feynman
+++John L. Coolidge++++++++...@apple.com+++++++++++++++++++++++
I speak for myself, not for Apple Computer. Copyright 1993 John L. Coolidge.
Copying allowed only if attributed, and if all copies may be further copied.
Craig Burley
Oh give me a break. The FSF can sue anyone for copyright violations,
but they are sure to loose. Especially if a wierd document like the GPL
is behind. the FSF software is freely distributed. You don't need to
read the GPL to know this, they keep flashing it on the screen. As such,
limiting it in funny ways, or sueing only specific people is unlikely
to work. [...]
The FSF never sued anyone and probably never will. I wonder if they even
registered the work with the copyright office. The legal procedure is
far too expensive, and besides, what would the FSF want, money? their
intent is not to make money, but make software widely avaialable ...
making the source free? no judge would order such specific results ...
I like RMS's claim that the GPL, even if not legally binding, achieves
its goal. As if other public-domain or free software doesnt! The only
one being stopped down by the GPL are large companies that refuse
any legal hassle[1], and "decent people" who respect the GPL. They loose
the most, having to reimplement parts of GPL code on their own, for use
in commerical software (e.g. like a piece of code to manange regex
that they are already familiar with, because they have contributed to
the GNU project!).
I assume this person is unfamiliar with the Software Publisher's Association
(SPA), which successfully sues (and collects damages from) companies
that violate copyright when making copies of software in a manner that
violate license agreements legally similar to the GPL (in terms of how
they are "accepted" by users).
It's always funny watching people who know few facts post stuff like this,
anyway. Good for a laugh. "GPL...not legally binding"? If it isn't,
then copyright law itself probably isn't either. As if being freely
distributed had anything to do with whether a copyright (and a license
specifying terms of copying) were legally binding! What a laugh.
--
James Craig Burley, Software Craftsperson bur...@gnu.ai.mit.edu
Member of the League for Programming Freedom (LPF) l...@uunet.uu.net
Craig Burley
The FSF never sued anyone and probably never will. I wonder if they even
registered the work with the copyright office. The legal procedure is
far too expensive, and besides, what would the FSF want, money? their
intent is not to make money, but make software widely avaialable ...
making the source free? no judge would order such specific results ...
I forgot to mention how this paragraph alone shows incredible ignorance
of copyright law, the Berne Convention, and legal procedures in general.
If anyone finds any of this guy's post anything like factual, send me
email, I've got plenty of things I'd like to sell you. :-)
John Lazzaro
Is this true? I certainly don't think its a sign of weakness if it is,
when someone has a strong hand the other side will prefer to negotiate.
Michael John Haertel
The FSF never sued anyone and probably never will. I wonder if they even
registered the work with the copyright office. The legal procedure is
far too expensive, and besides, what would the FSF want, money? their
intent is not to make money, but make software widely avaialable ...
making the source free? no judge would order such specific results ...
The FSF routinely registers everything with the copyright office.
Elizabeth Haley
The GPL is similar to a SWL, but it doesn't meet my definition above
(nor yours). However, there is a _serious_ flaw in the GPL reasoning
that accepting the GPL is the only thing allowing you to copy the
program -- the initial copying phase of the program. When I ftp to a
server and grab emacs (for example), I am making a copy there and
then. At this point I cannot necessarily be presumed to have read the
GPL. Thus, downloading a GPL-protected program may be an illegal act,
since I cannot be presumed to have accepted a license I have never
read. The illegality of the act of downloading cannot be removed by a
later acceptance of the license.
Ummmm, someone correct me if I am wrong, but in my memory, nothing
prevents you from copying GPL'ed stuff, whatsoever. That would be
silly. You signal acceptance of the license by *distributing* it.
Note, also, that this clears the way for the 'user-does-the-link'
workaround. The scenario is: person A creates proprietary.o which
must be linked with a GPL-protected program to work. Person A then
distributes proprietary.o (presumably for some cost), with or without
GPLProgram. Person B obtains proprietary.o and GPLProgram. Assuming
person B has legally obtained proprietary.o and GPLProgram, there is
no legal way to bar person B from creating a derivative work _for
their own use_. Person B, by both the GPL and (presumably)
proprietary.o's copyright and/or license, cannot redistribute their
derivative work, but they are legally entitled to use it themselves.
There is no way, under the GPL, to stop person A from distributing
GPLProgram along with proprietary.o (the mere aggregation clause).
While RMS has indicated that such a technical workaround would be
viewed by the courts as an attempt to sidestep the intent of the GPL,
there really is no way for the court to rule in favor of the FSF.
Every step is _entirely_ legal. There is simply no room for a court to
make a ruling against person A -- they have broken no laws, and they
have never distributed a GPL-derivative work, so proprietary.o cannot
have fallen under the GPL. Person B has also acted in an entirely
legal way. Indeed, if the FSF did indeed sue either person A or B,
it's perfectly likely that a countersuit for interference would
succeed.
But this isn't a "GPL" work around, is it? No-one is selling any
GPL'ed code or a derivative work, just a third party object file.
The only way for the FSF's viewpoint to succeed here is for the
internal interfaces of GPLProgram to be legally protected. Then
proprietary.o can be considered a derivative of those interfaces and
thus it would fall under the GPL. The GPL itself seems to indicate
that this might be the FSF's view. However, it seems implausible to me
that the FSF would really be in favor of a system where the interfaces
of a software program could be considered protected works. Such a
system would (for instance) have allowed AT&T to have copyrighted the
interface to UNIX and prevented the GNU project in the first place.
Disclaimer: I am not a lawyer. However, I've had some good contract
law classes and I've looked at a number of court decisions in this
area.
Again, correct me if I am wrong:
The GPL prohibits: distributing GPL'ed code as your own (standard
copyright)
distributing partial packages
Or otherwise preventing someone from getting
all the code to a package
The Library license allows you to distribute your proprietary program
made with GPL'ed *libraries* if you provide your *object* files and
the GPL stuff separately, in ready to link form, so that the user may
choose to go with other libraries, or link in new GPL'ed stuff or
whatever, in essense giving them the right to hack on the GPL'ed
stuff, even if they can't hack your proprietary stuff.
This is probably a gross simplification, but the essense of all this
stuff is trying to be:
If something is broke, let me have a hand at fixing it. It's not fair
to make me put up with something broken or stupid when I have the will
and knowledge to make it better.
--
Jesus saves sinners...
/****************************************************************************\
==============David Charles Todd, tHE mAN wITH tHREE fIRST nAMES==============
I/O Error: core dumped.
\*************************hac...@headcheese.daa.uc.edu**********************/
...in a shoebox in his mother's attic.
Craig Burley
arro...@jyusenkyou.cs.jhu.edu (Ken Arromdee) writes:
>definition embodied in those laws. So if rms is right, the fact that the GPL
>does not fit the definition of a shrink-wrap license does not mean that it
>might not be legally treated as one anyway.
The two cases I'm aware of wherre shrink-wrap licenses were tested in court
held that the shrink-wrap licenses were not valid. Also, of the several
law review articles I found that talked about shrink-wrap licenses when
I did a Westlaw search for such articles, none argued that they would be
valid.
I haven't kept up on this stuff _at all_, but my vague recollection is
that the main reason shrink-wrap licenses were held invalid (at least
in some older, i.e. 3-yr-old?, ruling) was that the person had _purchased_
a product, naturally assuming that he or she then owned the product and
its contents, and could not have read the license before the purchase nor
was made to sign it before the purchase. (That is, the person purchased
a copy of a product, of course -- not the rights to the entire product,
manufacturing and all.)
In other words, the license put the purchaser in the position of having
made the purchase under, essentially, false pretenses, and walked out of
the store, perhaps driven 50 miles home, only to discover that he or she
did not really own the software and had to make significant legal
commitments just to use it (at which point returning the package would
be difficult).
If that vague (and probably incorrect) memory is the case, then the GPL
shouldn't be lumped in with shrink-wrap licenses in general. As long
as the person obtaining a GPL'ed distribution has a means of learning
about the license and reading it prior to receipt (which is basically
always the case, if the GPL itself is followed), there should be no problem.
A particular distributor might shrink-wrap a product, label it "GNU EMACS",
and fail to clearly indicate that it is merely a (pricey :-) distribution
of licensed (and unwarranted) software, but that won't make the GPL
itself weak in court -- though that distributor might get in some hot
water (at least for violating the spirit, if not the letter, of the GPL)
and end up in court accused of infringing the copyright of the author
of the GPL-protected code (the FSF, in this case), and/or perhaps being
made liable for fixing defects found in GNU EMACS by that distributor's
customers.
Note: as far as I'm aware, the aspects of the shrink-wrap licenses that
were not held up were _not_ the copyright aspects, but the other ones --
like "no disassembly". At least, it makes sense that a purchaser realizes
up front that buying "Microsoft Word" no more gives that person to make
unlimited copies of the floppies and documentation than does buying
"The Way Things Ought To Be" by Rush Limbaugh (to pick a recent best-selling
example) gives that person the right to make unlimited copies of that book.
However, what also makes sense is that a person making such a purchase
expects to be able to take apart the _copy_ of the product (that they _own_)
and do with it what they will (as long as they don't distribute the
possibly-protected results to others). Shrink-wrap licenses typically
deny this freedom post-purchase, just as if you got a book home and
discovered you weren't allowed to read pages out of order or got a car
home and discovered you weren't allowed to look under the hood to learn
how it worked. I _suspect_ that such unusual post-purchase prohibitions
are the ones that are not being upheld in court.
Since the GPL doesn't put any unusual restrictions on distributions
(e.g. "no young black males may use this product", to modify a bizarre
example posted earlier) and is (or should be) used in a manner that makes
it clear to recipients _at_ or _prior_ to receiving a distribution that
it is an unwarranted distribution, or copy, of a licensed product they
are receiving, I'd be quite surprised if the GPL were challenged in
court, and even more so if it succeeded. Such a success would probably
result in the immediate removal of all non-PD literature (including
software) from all ftp sites, BBSes, and so on, since it would imply that
authors cannot place reasonable restrictions on further copying of
copyrighted works simply because those works were easy to obtain (i.e.
free, shareware, or whatever). It might even put most proprietary software
developers in a very awkward position, since their products inherently
exist in a medium that makes copies of them easy to obtain, as in "well,
I found a copy of FrameMaker on Bob's system, so I just copied it to mine,
that's like using ftp, right?", and cause them all to go back to
copy-protecting disks, using hardware/software keys for one-user or
one-node or one-network restrictions, etc.
The long and short of it is, I see little chance that a court will
decide to even listen to why the GPL should be singled out for
invalidation, and almost as little chance that all such licenses
would be invalidated by any court.
(If anyone knows for certain _which_ aspects of shrink-wrap licenses
were not upheld and which were, please post. It'd be something if
it turned out you _could_ make and distribute copies of a shrink-
wrap-licensed software product! I doubt that the presence of a shrink-
wrap license invalidates the author's copyright, however.)
Remember, the GPL, unlike shrink-wrap licenses, doesn't say what you
can and can't do with a distribution in general -- it just says
what you can and can't do when you _redistribute_ material from
a GPL-protected distribution. Though there are many people who
sincerely feel that the difference between these two things is vanishingly
small, most of society recognizes a wide gulf between the right to
scribble notes in one's copy of someone else's book and the right to sell
copies of that scribbled-in book without obtaining the original author's
permission. Shrink-wrap licenses typically prevent you from even
_reading_ the book (among other things), in a sense, and that's something
quite different from restricting how (or whether) redistribution happens.
Fortunately, the courts (and judges) generally understand these
distinctions fairly well.
Michael Golan
>I assume this person is unfamiliar with the Software Publisher's Association
>(SPA), which successfully sues (and collects damages from) companies
>that violate copyright when making copies of software in a manner that
>violate license agreements legally similar to the GPL (in terms of how
>they are "accepted" by users).
They sued for software published as free? Go read my post again. Copyright
law gives you certain rights, sure. It does not, according to many lawyers,
allow you to place any kind of restriction you like. Your nice post have
failed to address my point about young black mails, for example. Indeed,
you failed to address any of the points.
In comp.lang.c++ this week, someone asked for a "free C++ compiler" and got
the answer "g++". No one bothered to correct that the software is NOT free.
If the post has said "Borland C++" you bet your ass someone (probably
from Borland, too) would correct it. Such claims for "free software" about
the GNU software are common, and will most likely lead the
court NOT to accept the special binding of the GPL. Frequent violations
that go unnoticed (like posting diff's w/o reposting the GPL!) are another
serious problem if you ever going to sue for copyright/license violations.
>It's always funny watching people who know few facts post stuff like this,
>anyway. Good for a laugh. "GPL...not legally binding"? If it isn't,
>then copyright law itself probably isn't either. As if being freely
>distributed had anything to do with whether a copyright (and a license
>specifying terms of copying) were legally binding! What a laugh.
>--
It is always funny to watch a reply that flames the original but provide
no counter details.
Who said the copyright low isn't binding? I said it can not be stretched
as you wish, and that the GPL is probably stretching it too much. That
the FSF but its actions and no-actions, have lost the copyrights law
protection. There are clear similar cases, e.g. Intel's 8088 (?) code & NEC;
it is not so hard to loose your rights.
-- Michael Golan
m...@cs.princeton.edu
ps. referring to the cost of the legal procedure, I meant sueing, not
the registration; I'm glad the FSF register their code. good for them.
As for lack of prosecution being a position of strength, thats only true
if people are not violating your rights in public. If they do and you take
no action, you may very well loose your rights!
Craig Burley
The GPL is similar to a SWL, but it doesn't meet my definition above
(nor yours). However, there is a _serious_ flaw in the GPL reasoning
that accepting the GPL is the only thing allowing you to copy the
program -- the initial copying phase of the program. When I ftp to a
server and grab emacs (for example), I am making a copy there and
then. At this point I cannot necessarily be presumed to have read the
GPL. Thus, downloading a GPL-protected program may be an illegal act,
since I cannot be presumed to have accepted a license I have never
read. The illegality of the act of downloading cannot be removed by a
later acceptance of the license.
It cannot be possible to commit an illegal act that is made illegal
only after committing it (excepting such abuses of the law and the
courts as in the Twitchell manslaughter case in Massachusetts --
fortunately, similar cases in other states have had guilty
verdicts overturned by enlightened higher courts).
Further, the GPL does not prevent you from making copies of someone
else's GPL-protected software (e.g. ftp'ing it to your machine), it
only places restrictions on your further redistributing that software
(e.g. putting it up for ftp on your machine).
Consider a software package put up for ftp that had a license
agreement lurking in it that stated, in effect, "it is illegal
to copy this software -- if you have already copied this software
to your system, even if you haven't used it, you have committed
an illegal act unless you send $2000 to...". If you think the
courts would ever pay _any_ attention to supporting the legality
of such a thing, why do you think nobody puts such stuff up for
ftp? Answer: because it wouldn't work (aside from perhaps roping
in a few sad sacks, but claiming to be Jesus appears to work better
and get you more guns 'n' fun :-).
Remember, laws are not like software viruses that you can use
to trip up unsuspecting but well-intentioned people -- only lawyers
(and legislatures) are at all like that. :-) Really, if this sort
of thing worked, you could stand right at a corner on a busy street,
wait until somebody accidentally bumped into you, and get them
convicted of assault and battery, but that just doesn't work
(again, except on the gullible) -- the courts consider intent of
all parties in matters like these. (E.g. if you put it up for ftp,
your intent must have been for people to copy it to at least determine
if they wanted to keep it and abide by your license restrictions; if
you copy via ftp, your intent cannot likely have included being prepared
to pay $2000 just for the privilege of making the copy even without making
use of it, but just to read the license agreement!)
Note, also, that this clears the way for the 'user-does-the-link'
workaround. The scenario is: person A creates proprietary.o which
must be linked with a GPL-protected program to work.
Right off the bat, there's not really such a thing -- if it's a .o
file, it can be made to work with anything object-level compatible with
a GPL-protected program, whether PD or whatever. And it can be
disassembled. But the scenario is still interesting...
Person A then
distributes proprietary.o (presumably for some cost), with or without
GPLProgram.
For this to be legal, proprietary.o cannot have been derived from GPL-
protected code (which means it can't #include GPL-protected files --
copyright doesn't just protect the executable statements), and "with...
GPLProgram" can only mean "with...GPLProgram as an aggregation, not
GPLProgram.exe with proprietary.o already linked in". But that's
certainly all possible...
Person B obtains proprietary.o and GPLProgram. Assuming
person B has legally obtained proprietary.o and GPLProgram, there is
no legal way to bar person B from creating a derivative work _for
their own use_.
In fact, Person B can do whatever they want with the code, including
linking in proprietary.o, or somebody's proprietary library, or
whatever they want, as long as they don't distribute the results without
following the GPL (which they cannot for a linked-in proprietary.o).
So far, this seems pretty fine and dandy....
Person B, by both the GPL and (presumably)
proprietary.o's copyright and/or license, cannot redistribute their
derivative work, but they are legally entitled to use it themselves.
That part seems fine to me. It doesn't hurt anyone who doesn't want
to be hurt (i.e. B makes a decision to use proprietary software).
There is no way, under the GPL, to stop person A from distributing
GPLProgram along with proprietary.o (the mere aggregation clause).
So it would seem, but...
While RMS has indicated that such a technical workaround would be
viewed by the courts as an attempt to sidestep the intent of the GPL,
there really is no way for the court to rule in favor of the FSF.
Every step is _entirely_ legal. There is simply no room for a court to
make a ruling against person A -- they have broken no laws, and they
have never distributed a GPL-derivative work, so proprietary.o cannot
have fallen under the GPL. Person B has also acted in an entirely
legal way. Indeed, if the FSF did indeed sue either person A or B,
it's perfectly likely that a countersuit for interference would
succeed.
There are some conclusions above that I wouldn't bet money on (then
again, I don't gamble).
The only way for the FSF's viewpoint to succeed here is for the
internal interfaces of GPLProgram to be legally protected. Then
proprietary.o can be considered a derivative of those interfaces and
thus it would fall under the GPL. The GPL itself seems to indicate
that this might be the FSF's view. However, it seems implausible to me
that the FSF would really be in favor of a system where the interfaces
of a software program could be considered protected works. Such a
system would (for instance) have allowed AT&T to have copyrighted the
interface to UNIX and prevented the GNU project in the first place.
Again, this conclusion does not necessarily follow. I suspect the FSF
hasn't ever argued in favor of protecting interfaces. Rather, it
might have simply pointed out that a .o file evidencing significant
interfacing with the internals of a copyrighted work might be
evidence in favor of finding that the .o file effectively (if not
literally) derives from that work.
Disclaimer: I am not a lawyer. However, I've had some good contract
law classes and I've looked at a number of court decisions in this
area.
Okay, that sounds good. Let me pose some considerations that might
help clarify things, because I, too, am curious about this scenario
(and its various possible incarnations). Let me know if you know of
any court decisions that might clarify (or counter) some of these,
because you almost certainly have had more experience and education
than I in this area:
1. Suppose proprietary.o is simply a drop-in replacement for libraries
normal used by GNUProduct.exe (.exe just helps people know I mean the
executable) that themselves are LGPL, PD, part of the OS interface,
whatever. So proprietary.o has replacements for some routines so
they run faster, or offer additional features, or something.
Since proprietary.o is not derived from GPL code (that is, it need
not be at all derived from it), I don't see a problem with this
scenario at all. Customers can still change GNUProduct source code
and recompile. Bugs in proprietary.o? Customers get what they
deserve -- at least they know they were linking in a proprietary
object file (relatively) up front.
I agree that in this case the courts could hardly find against
the distributor of proprietary.o, and I doubt the FSF would ever
true and sue anyone like that. There's no reason to -- people still
get the benefit of the source code, even users of proprietary.o
2. Suppose proprietary.o replaces some significant numbers of GNUProduct's
routines, such that a customer changing GNUProduct source code,
recompiling, and relinking with proprietary.o is likely to end up
with a broken GNUProduct (as proprietary.o relies on various internal
structures, procedure names, and so on, being the same).
In this case, I agree that the FSF would have a problem with the
distributor of proprietary.o, because users making use of it would
be denied the benefits of source availability for GNUProduct.
I suspect the courts might also have a problem with the distributor,
because it could be shown that while the .o file didn't necessarily
derive _directly_ from GPL-protected (e.g. FSF-owned) code, it did
_substantially_ derive from it, to the extent that its utility was
largely limited to a particular incarnation of FSF-owned code and
that is purpose was clearly tailored to circumvent the GPL. Also, it
could be shown that it relied so extensively on knowledge internal
to GNUProduct code that it effectively _did_ derive from that code.
3. "Innocence in isolation" is not always a sufficient defence in a
court of law. If it can be shown that the comprehensive actions
of several individuals were designed to commit an illegal act, it
can be found that each individual is guilty of the whole act,
even when each individual in isolation is guilty of nothing.
Consider 4 people, A, B, C, and (the victim) D. A happens to
suggest to D that D stand right next to a wall at a particular
time. At the same time, B happens to use a carpenter's nail
driving gun to put a nail in the wall from the other side.
Unfortunately, C earlier happened to have adjusted the same gun
to maximum force. The result is that B kills D by firing a
nail through the wall into D's head. [This sort of thing has
happened by accident, FYI.]
Now, each action in isolation is basically innocent, at least of
murder. In fact, any combination of A, B, and C might have been
aware or unaware that the death of D would be the result of their
(isolalted-innocent) action in combination with the others' actions.
So, is a court completely unable to find anyone guilty in this case?
Of course not. If it can be shown that any one of them was aware
of the other preconditions (along with motive and other such things),
that person can be found guilty of murder for what appear to be
innocuous things, like asking someone to stand somewhere (which they
did voluntarily), or shooting a nail into a wall (which is a legitimate
thing for carpenters to do), or adjusting a nail gun (which is also
legitimate, though perhaps reckless if the operator wasn't
notified). E.g. if testimony shows that B didn't know D at all, but
A and C did and wanted D dead, that they conspired to commit
"innocent" acts knowing that, at a particular time based on knowing
B's job, B would fire the fatal nail. Or, B might know of C's
penchant for adjusting tools to full bore ("I've asked him over and over
again to stop doing that") and A's tendency to ask people to stand
against a wall to be photographed, and choose to fire when he overhears
D stand for a passport photo, and that, too, can be determined in court.
And, of course, A, B, and C could all be found guilty of murder.
Now, returning to our less macabre example involving GNU code, since
B is only the recipient of a distribution (and not a
distributor) and hence not even the "target" of GPL restrictions, it
is unlikely that the courts would see A and B as acting in concert to
get around the GPL. Nevertheless, I only realized this after typing in
this point, and I think it's too much fun to delete now, and anyway,
it does refute the notion that actions innocent in isolation are
necessarily innocent in combination and thus out of the reach of the
law.
4. Suppose A also reverse-engineers Lotus 1-2-3 and sells a proprietary
program that changes a 123.EXE (or whatever it's called) to provide
additional features, more speed, etc. Let's assume the reverse-
engineering itself was legal (or allowed because of shrink-wrap
problems :-). Now, if the proprietary program (FIX123.EXE) just
substitutes better library routines a la #1 above, could Lotus sue
for copyright infringement? I doubt it. But, if it makes significant
changes based on intimate knowledge of 1-2-3 internals, then even
though it might be strictly derived from no 1-2-3 code, I suspect
that a court might well rule that a copyright was nevertheless
infringed. But, again, this is the interesting point to me -- would
the courts allow people to essentially rewrite from scratch just a few
critical portions of a copyrighted work and offer those rewrites as
enhancements, such that those rewrites were usable only in conjunction
with the original work?
An example from other areas might be helpful. Suppose C publishes a
copyrighted (and expensive) photograph. A buys a single copy of the
photo, and develops a transparency with special coloring that, when
overlaid on a particular photo that is generally available (PD or
just cheap), causes the result to look just like the expensive photo.
B buys that transparency, gets the cheap photo, and does the overlay,
thus saving lots of money. Has C's copyright been violated? By whom?
A developed a transparency that, strictly speaking, might well be
underived from C's photo, but would the courts decide that it was
_effectively_ derived from C's photo since its only use was to create
an image of that photo in conjunction with some cheap photo? B would
seem completely innocent, not having distributed anything. What
about the copyright holder of the cheap photo -- is there a violation
of that person's rights? (I don't think so, since the photo was
paid for by all users and not redistributed.)
The whole issue comes down, again, to rms' assertion that the courts
don't work just like a computer. Just because there is no actual
GPL code from which proprietary.o is derived doesn't mean that that .o
file isn't in a legal sense derived from GPL code -- or maybe it does.
If it is possible that the two things are different, then it is a
question of degree. At what point does proprietary.o so depend on
the internal globals, types, and procedure names in GNUProduct that
it can be considered derived from GNUProduct code independent of whether
proprietary.o was compiled with any GNU code present on the compiling CPU?
If the answer is "never", then it should be legal for people to distribute,
say, programs that change "demo" .EXEs so they can function as full-
featured programs (e.g. reenable saving and printing). But I suspect it
is not.
Anyway, the only fatal flaw I know of in the GPL is that, in trying
to provide a reasonable way for reasonable people to have reasonable
access to reasonably good software, it has attracted a lot of
unreasonable attention. (Though I think that if the code it protected
didn't include some really good stuff, nobody'd really care.)
Tim Smith
>It raises a question, though: suppose the GPL _did_ say "no young black males
>may use this product". We could all agree that that's unethical. But why
>would it be illegal?
It's contrary to public policy.
--Tim Smith
John L. Coolidge
>I assume this person is unfamiliar with the Software Publisher's Association
>(SPA), which successfully sues (and collects damages from) companies
>that violate copyright when making copies of software in a manner that
>violate license agreements legally similar to the GPL (in terms of how
>they are "accepted" by users).
I could be wrong, but I don't recall the SPA suing anyone. Instead,
they use threats of lawsuits as a nuisance factor to coerce companies
into paying (often for software which they in fact legally own but
cannot conveniently prove ownership of). Most companies would rather
pay the SPA than risk losing a court action, even if they have a very
high probability of winning.
Note, also, that the SPA really only goes after damages which are
clear from copyright law (presence of copies of a program for which
sales receipts cannot be found). I don't recall them trying to enforce
shrinkwrap license agreements (probably because they don't really
believe they're enforcable either).
--John
By analysis of usenet source, the hardest part of C to use
is the comment.
-- Bill Davidsen (davi...@crd.ge.com)
John F Carr
bur...@apple-gunkies.gnu.ai.mit.edu (Craig Burley) writes:
>For this to be legal, proprietary.o cannot have been derived from GPL-
>protected code (which means it can't #include GPL-protected files --
>copyright doesn't just protect the executable statements)
Is there any legal precedent about #include of a file describing an
interface creating a derivative work of the header file? I asked about
this on misc.legal.computing a few weeks ago and didn't get a good answer.
>Again, this conclusion does not necessarily follow. I suspect the FSF
>hasn't ever argued in favor of protecting interfaces. Rather, it
>might have simply pointed out that a .o file evidencing significant
>interfacing with the internals of a copyrighted work might be
>evidence in favor of finding that the .o file effectively (if not
>literally) derives from that work.
I'd like to see some precedent for this also. I thought that trade secret
law and patent law, not copryright law, protected the internal workings of
programs.
I wonder if reverse engineering precedent can be applied directly to this
case? Most of the time you're looking at legally obtained copyrighted
binary code to determine the interfaces. In the case of GPL software
you're looking at legally obtained copyrighted source code to determine the
interfaces.
--
John Carr (j...@athena.mit.edu)
Tim Smith
>I could be wrong, but I don't recall the SPA suing anyone. Instead,
>they use threats of lawsuits as a nuisance factor to coerce companies
>into paying (often for software which they in fact legally own but
>cannot conveniently prove ownership of). Most companies would rather
Wouldn't it be easy for most companies to prove ownership? There should be
a purchase order in their files, and maybe something in their mail department
logging the receipt of the software.
--Tim Smith
Alan Braggins
> Clarificiation: Suppose I write a program with a "Racist Public License"
> (RPL). Someone distributes it to a black, and I sue for copyright violation.
> Now, the _license_ is against public policy. But remember, the rationale
> for being able to restrict how you can distribute GPL stuff, or RPL stuff, is
> not that the license _prohibits_ certain types of distribution (binary-only
> distribution in one case, distribution to blacks in another), but rather that
> the default is that you have no distribution rights at all and the license
> simply _allows_ distribution under more circumstances than "never".
> Therefore, if I sue someone who distributes RPL software for a black, I would
> not be suing them for violating the RPL. Rather I would be suing them because
> the default is that they have no redistribution rights at all. If the RPL
> violates public policy and is invalid, that doesn't mean they can now
> distribute to blacks; rather it means that since the RPL is invalid, there is
> no license at all, and they can't distribute to _anyone_, of any race. It
> would still be copyright infringement to distribute to blacks (as well as
> whites), and my suit would continue.
> --
Would the fact that you have actively been encouraging distribution to whites
in the meantime mean that you could be found to have given up your copyright?
--
Alan Braggins +44-223-316673 Shape Data, A Division of EDS-Scicon Ltd,
al...@sdl.ug.eds.com Regent St, Cambridge, CB2 1DB, U.K.
Mel Walker
It generally is easy to prove ownership of legal copies. What the SPA does
is count the number of copies existing on various media, count the purchase
records, and compare the numbers.
----------------------------------------------------------------
Mel Walker mwa...@novell.com
All opinions expressed are of the author.
Novell, Inc. is not responsible for the content of this article.
Craig Burley
bur...@apple-gunkies.gnu.ai.mit.edu (Craig Burley) writes:
>I assume this person is unfamiliar with the Software Publisher's Association
>(SPA), which successfully sues (and collects damages from) companies
>that violate copyright when making copies of software in a manner that
>violate license agreements legally similar to the GPL (in terms of how
>they are "accepted" by users).
They sued for software published as free? Go read my post again. Copyright
law gives you certain rights, sure. It does not, according to many lawyers,
allow you to place any kind of restriction you like. Your nice post have
failed to address my point about young black mails, for example. Indeed,
you failed to address any of the points.
It's "males", not "mails". And no, I did not fail to address it -- I
pointed out that the GPL has nowhere near the arbitrary kind of
restriction that shrink-wrap licenses have ("you cannot disassemble
the program") or anything like "no distribution by young black males".
(Or maybe I made that point in a separate post -- your post was so
silly that I didn't really say much about it.)
The GPL is not much different from many other license agreements for "free"
code that nevertheless is copyrighted -- for example, many include
restrictions on "distribution for commercial gain", which the GPL does
not (in the main).
The point is that the GPL isn't particularly unsual or out of line from
the point of view of looking at a wide variety of software
licenses that are widely used and have been unchallenged (or withstood
challenges) in the courts. The only things really interesting
about the GPL is that it protects some very high-quality and
desirable software, does not prevent distribution for commercial
gain, requires source distribution, and attracts the ire of people
who want to pillage the software it protects for their own uses.
In comp.lang.c++ this week, someone asked for a "free C++ compiler" and got
the answer "g++". No one bothered to correct that the software is NOT free.
A distribution (copy) of the software can be obtained for free, and
legally. That is what most people mean when they ask a question like
the above. Some people mean "public-domain", but unless an answer
claims g++ is public-domain, claiming it is free is not incorrect.
No court, unless ruled by an insane judge, would rule that attaching
the word "free" to something people could obtain for free but couldn't
redistribute copies of for free causes that thing to lose its copyright
status. Else that would mean that someone advertising "free books"
and especially giving books away would, unless all the copyright
holders of those books objected (sued), cause all the books to lose
their copyright protection. That's idiotic, obviously (so idiotic
that it causes me to think that anyone holding that view must be
an idiot, hence the hostility in my posts responding to yours -- sorry).
Even the _copyright holder_ of a book may give away free copies to
anyone and everyone, and advertise the book as "free", and the courts
won't take away that person's copyright! Why would software be any
different?
If the post has said "Borland C++" you bet your ass someone (probably
from Borland, too) would correct it. Such claims for "free software" about
the GNU software are common, and will most likely lead the
court NOT to accept the special binding of the GPL.
Wrong. If that were true, then the "Free Software Foundation", which
created and publishes the GPL, would have disappeared long ago, thanks
to various people who are more interested in pillaging FSF code than
writing their own.
Furthermore, the courts are hardly going to pay attention to whether
the FSF constantly rebuts claims that its software is "free" made on
forums like USENET.
Frequent violations
that go unnoticed (like posting diff's w/o reposting the GPL!) are another
serious problem if you ever going to sue for copyright/license violations.
Not really. diffs that have significant content from a copyright
perspective are rarely posted, as far as I can tell, and they are
applied to files with copyright notices. And, again, the FSF is
hardly responsible for noticing and objecting to (suing for) every
semi-public infringement of its copyrights on USENET, on BBSes
worldwide, etc. Copyright law doesn't rqeuire that. You must be
thinking of trademark law.
Sure, if someone widely and publicly gives away or sells more than
smallish chunks of GPL'ed products without the GPL, the FSF probably
would (and should) object and possibly sue them if they don't stop,
because they're violating the GPL. But the level of activity on
USENET and over ftp sites is, as far as I can tell, miniscule
compared to this possible example. And the FSF is not expected by
anyone (like the courts) to spend all its time going on search-and-
destroy missions trying to find copyright violations. On the whole,
copyright law doesn't take into account how tenacious the holder is
when it comes to preserving the copyright status.
>It's always funny watching people who know few facts post stuff like this,
>anyway. Good for a laugh. "GPL...not legally binding"? If it isn't,
>then copyright law itself probably isn't either. As if being freely
>distributed had anything to do with whether a copyright (and a license
>specifying terms of copying) were legally binding! What a laugh.
It is always funny to watch a reply that flames the original but provide
no counter details.
The GPL is legally binding. That is enough of a counter detail. Plenty
of lawyers in the USA of already been paid plenty of money (by any normal
person's standards anyway :-) and determined that the GPL is legally
binding. Otherwise there'd be widespread use of GCC and GNU EMACS source
code in proprietary products. Various organizations have threatened
to take this step if the FSF didn't give them a special license, and
they all, to my knowledge, backed down when the FSF didn't capitulate.
If I post something stupid like "1 plus 1 is not 2, as anyone can see",
would _you_ be able to provide "counter details"?? Now that _you've_
provided more detailed (and generally absurd) claims, I (and others)
can respond substantively, but your original post was laughable. I, too,
probably posted things like that when I was 14 years old and using
computers on a nearby campus (MIT), so it's a bit painful as well.
(Fortunately, my audience was not world-wide.)
Who said the copyright low isn't binding? I said it can not be stretched
as you wish, and that the GPL is probably stretching it too much.
Why is the GPL stretching it too much? My point is that if the GPL is
doing so, then many other licenses on copyrighted products (and
_especially_ shrink-wrap licenses, but probably most proprietary
products anyway) have done so as well. Many licenses restrict
_use_ (the GPL doesn't) in addition to redistribution (which the GPL also
does) -- for example, some restrict to "non-military" or "non-nuclear"
use (e.g. Apple Mac products -- do they still do this, I wonder?),
others prevent disassembly (the GPL doesn't). How is the GPL such
a standout among these other licenses? The word "free"? I've seen this
word applied to many copyrighted programs that are distributed for no
charge and have no license agreement -- which means they cannot be
redistributed without explicit permission of the author, which is
more restrictive than the GPL, so are all _those_ works going to lose
their copyright status too, because people make claims that they're
free (including the authors)? Answer: no. You don't know what you're
talking about.
That
the FSF but its actions and no-actions, have lost the copyrights law
protection. There are clear similar cases, e.g. Intel's 8088 (?) code & NEC;
it is not so hard to loose your rights.
It is very easy to "loose" your rights, but in this case, it is harder
to _lose_ your rights as a result of something like using the word "free"
to describe software that can be copied for free.
Please post case histories or reference numbers for the cases you mention,
or explain exactly how they managed to register their copyrights on
software and still lose the copyright because they used the word "free".
(Or were you kidding when you said "There are clear similar cases"?
They'd _better_ be "clear similar", which some of us who know English
might phrase as "clearly similar".)
ps. referring to the cost of the legal procedure, I meant sueing, not
the registration; I'm glad the FSF register their code. good for them.
You seem to have difficulty expressing what you mean, not to mention
spelling it correctly. If you'd told everyone last October that Bill
Clinton's economic plan would increase taxes for everyone above $36K/yr
income, I'd have guessed you were Dan Quayle!
As for lack of prosecution being a position of strength, thats only true
if people are not violating your rights in public. If they do and you take
no action, you may very well loose your rights!
The FSF has been reasonably effective in defending its copyrights, from
what I've seen, and remember that defending copyright does not require
the same kind of effort for the courts to permit retention of said
copyright as does, say, defending patents, or anywhere near what is
required for defending trademarks. GNU is not a trademark, by the way,
Maybe it should have been made one, and maybe it still could be, but
before you trademark something, you have to be willing to expend
significant resources defending that trademark, which the FSF probably
wasn't (and isn't) willing to do. (FYI, I believe the DoD trademarked
"Ada", and later decided it wasn't going to maintain that trademark.)
A few posts on USENET containing snippets of source code, to which the
FSF does not reply, no more causes the FSF to lose its copyright than
do the plethora of quotes in people's .signature's from copyrighted text
(to which the quotes' authors do not reply) cause the works containing
those quotes to lose copyright status.
Before you post drivel like this again, try talking to a lawyer who
understands copyright law, and make sure you get your facts straight
first. If the lawyer agrees with what you've said, be sure to post
his name and number so we can call and check with him, and determine
whether you gave him the facts, or whether he's simply incompetent.
And before you ask me whether I've done this, I'll answer:
I don't have to. The FSF has its own lawyer(s), and businesses that
have threatened to subvert the GPL have their lawyers, and simply
watching the behavior of the FSF and those businesses is sufficient
evidence for me to see that the courts would not do the foolish
things you seem to think they would.
Instead of convincing GNU supporters of your beliefs, why not convince
a major corporation's lawyers that they can and should distribute
proprietary versions of GNU code? You'd not only find a more
receptive audience, but you'd probably make money at it as well.
Craig Burley
Would the fact that you have actively been encouraging distribution to whites
in the meantime mean that you could be found to have given up your copyright?
Very funny.
Sigh. :-)
Craig Burley
bur...@apple-gunkies.gnu.ai.mit.edu (Craig Burley) writes:
>I assume this person is unfamiliar with the Software Publisher's Association
>(SPA), which successfully sues (and collects damages from) companies
>that violate copyright when making copies of software in a manner that
>violate license agreements legally similar to the GPL (in terms of how
>they are "accepted" by users).
I could be wrong, but I don't recall the SPA suing anyone. Instead,
they use threats of lawsuits as a nuisance factor to coerce companies
into paying (often for software which they in fact legally own but
cannot conveniently prove ownership of). Most companies would rather
pay the SPA than risk losing a court action, even if they have a very
high probability of winning.
Yes, you could be wrong, or I could be wrong. I seem to remember
reading about some serious judgements the SPA got in court against
some intransigent companies awhile back. Best to actually contact the
SPA and ask them, though.
Craig Burley
this is really off-topic, I decided to at least change the subject line.]
In article <1993Mar3.2...@blaze.cs.jhu.edu> arro...@jyusenkyou.cs.jhu.edu (Ken Arromdee) writes:
In article <1n33vd...@shelley.u.washington.edu> t...@stein2.u.washington.edu (Tim Smith) writes:
>>It raises a question, though: suppose the GPL _did_ say "no young black males
>>may use this product". We could all agree that that's unethical. But why
>>would it be illegal?
>It's contrary to public policy.
Clarificiation: Suppose I write a program with a "Racist Public License"
(RPL). Someone distributes it to a black, and I sue for copyright violation.
Now, the _license_ is against public policy. But remember, the rationale
for being able to restrict how you can distribute GPL stuff, or RPL stuff, is
not that the license _prohibits_ certain types of distribution (binary-only
distribution in one case, distribution to blacks in another), but rather that
the default is that you have no distribution rights at all and the license
simply _allows_ distribution under more circumstances than "never".
Therefore, if I sue someone who distributes RPL software for a black, I would
not be suing them for violating the RPL. Rather I would be suing them because
the default is that they have no redistribution rights at all. If the RPL
violates public policy and is invalid, that doesn't mean they can now
distribute to blacks; rather it means that since the RPL is invalid, there is
no license at all, and they can't distribute to _anyone_, of any race. It
would still be copyright infringement to distribute to blacks (as well as
whites), and my suit would continue.
It is an interesting question. It might fall into the same kind of
area as one or more of these:
- Whether you can sell your house to only someone of a particular
race (hard to actually prevent without preventing one from selling
only to a friend even though someone else offers more -- something the
law, to my knowledge, does not attempt -- and if you happen to
privately define "friend" by race, that is your business, not anyone
else's)
- Whether you can advertise that you'll sell your house only to
a SWM (easier to enforce preventing people doing this, since it's
published racism in some form)
- Whether you can advertise that your new condo development prohibits
any SWM from moving in
- Whether your college accepts SWMs (again, in effect, or publicly)
I see the point about the license _granting_ additional permissions,
but I'd be surprised if this point was persuasive in court. After
all, you _grant_ permission for someone to buy your house -- it's
not something you have to provide -- but that doesn't mean you can
advertise that you'll only sell to people of a certain race (or, if
it does, then maybe you can enforce an RPL).
Often forgotten in thinking about these things is the fundamental
distinction we have in our society between personal freedom and
trade freedom -- it's so implicit, it's easy to overlook. You can
almost certainly choose to never distribute a copy of something you
own to a Single White Male, if you want, because of your personal
freedom of association and such -- but if you put that thing into the
realm of trade, many more restrictions are possible, such as anti-
discrimination laws and truth-in-advertising laws.
What constitutes putting something into the realm of trade (aka commerce)?
Well, having each person you distribute to sign a contract whereby
they agree not to further distribute to a SWM might not, but it might.
Publicly distributing along with a license that is publicly readable
and has racist provisions probably does constitute entering the
realm of trade (even though you might not charge anything).
Understanding the trade aspect helps to decode messages you hear
in the media. For example, when an advertiser sued for false
advertising claims "freedom of speech", that is really code for "we
want to be able to lie in public about our products" -- if that were
purely a freedom-of-speech issue, then so would be paying for their product
with counterfeit currency -- after all, if it was accepted, what's
wrong with it, the currency is just a form of speech, right? But
they made those claims within the realm of trade, where many more
restrictions exist, and those restrictions change frequently.
Or when someone overhears a private conversation containing vicious
racial slurs and publicizes the conversation, claiming the same
kind of exception to "freedom of speech" that society makes for
yelling "Fire!" in a crowded theatre, that is really code for "we
want to prevent people thinking a certain way, and especially from
talking privately with others about their thoughts". Here, the
crime isn't the private conversation, it's the unauthorized
publicizing of it, putting it into the realm of trade, and victimizing
those who participated in the conversation (something that might
be sinful from a Biblical perspective, but should hardly be a crime).
I'm unsure whether the Marge Schott controversy falls into this
realm -- whether she had a private conversation that was (unjustly,
IMHO) publicized, or made a public or semi-public kind of like
Rev. Jackson's famous "hymietown" reference.
Or when someone claims that prohibitions against doctors performing
abortions or assisting suicides compromises an individual's personal
rights to reproductive freedom or to die, that is really code for
"we want doctors to feel okay about doing something that the Hipprocatic
Oath probably forbids, and not be legally at risk for doing it, so it
is less risky and perhaps less expensive for people to exercise their
rights". Here, it is fundamentally a different issue between a person's
individual rights and whether the government should sanction trade in
making exercising those rights easier (though it is often claimed that
economics make the two intertwined, that's pure subterfuge -- I have
a right to own a Ferrari, does that mean that federal funding should
be available for me to buy one?). That is, it's reasonable for there
to be laws requiring a 24-hour waiting period before a doctor may
perform an abortion or assist a suicide of a patient, because that's
a restriction on the doctor's trade (just as laws putting cooling-off
periods on home-mortgage loans are reasonable), but it is an unreasonable
infringement of a person's rights to require them to, say, have to
notify a gov't office and then wait 24 hours before killing themselves.
(Whether laws should exist that protect the unborn fetus at 1, 3, 6,
or 9 months is a tough issue in society -- me, I think fetuses at any
age deserve at least as much protection as any animal, but that's
not in line with the current Political Cleansing movement of course.
And if fetal transference were ever to become cheap and easy, I see
no reason the gov't should be permitted to restrict when and how it
happens as long as the life of the fetus isn't interrupted -- though,
again, laws might reaonably be made restricting how doctors perform
such services.)
Well, I've gotten completely off the track here, but discussing racism
with respect to the GPL probably started it. I plead collusion only. :-)
Ken Arromdee
> Would the fact that you have actively been encouraging distribution to whites
> in the meantime mean that you could be found to have given up your copyright?
>Very funny.
>Sigh. :-)
I don't think you've been following the discussion.
I made an analogy to a hypothetical "Racist Public License" that prevents
redistribution to blacks, instead of redistribution without source. The RPL
gets around limitations on shrink-wrap license the same way the GPL does:
if someone doesn't accept the license, then they cannot redistribute, not
because they violated the license, but rather because the license is the only
thing that gives any redistribution rights at all.
That being said, the point of the RPL analogy is, of course, to point out
certain ways the GPL doesn't work. If a court found the RPL invalid, it
could very well rule that the originator of the RPL gave up his copyright by
allowing distribution to whites; the fact that the RPL is the only thing
allowing redistribution at all is totally irrelevant. But the same could
happen with the GPL. If a court found the GPL illegal, the argument "well, the
GPL is the only thing allowing redistribution, and without it you cannot
redistribute at all" may very well be similarly invalid.
Ken Arromdee
>had much in common, since the offending aspects of shrink-wrap simply
>aren't present for the GPL, as described in excruciating detail in other
>posts. The GPL doesn't add post-receipt _restrictions_ beyond what
>people normally expect when obtaining a copyrighted item -- it adds
>post-receipt _freedoms_. It's hard to see that as a subterfuge,
>though not to a few people who've posted here before.
The user ends up allowed to do exactly the same things with the real version
'you may not redistribute without source because if you do there is no
contract' and the hypothetical version 'you may not redistribute without source
because we consider you bound to a contract that restricts you this way'
versions. It is quite likely a court would consider the former as a subterfuge
for the latter, since they both produce identical results with respect to
what the user is allowed to do.
>(Obviously, the logic "rms says the law will treat A as B even though
>A and B are different" cannot be universally applied, else the law will
>treat a tadpole as a convertible subordinated debenture. With the
>top down. :-)
But what about when A and B are different methods of achieving the same result?
Ken Arromdee
>but I'd be surprised if this point was persuasive in court.
That's the (or, a) point of the comparison of the RPL and the GPL. The GPL
isn't "really" a shrinkwrap license because it doesn't restrict permissions,
it grants them. Yet, we can clearly see that for the RPL "grant permission
to whites" is the same as "deny permission to non-whites".
Of course, the RPL is quite illegal; it's only an analogy to clarify similar
points about the GPL. If "grant permission to X" is a subterfuge for "deny
permission to !X" when X is whites, then it should also be so for other values
of X--and that makes the GPL a shrinkwrap license.
Joe Buck
>Ok, so now I've read this entire string, and I'm confused. Let's say
>I write a program which performs some unnatural act, and I compile it
>with gcc. Consider the following scenaria:
> 1.) I link it with the gcc libraries. Does the result of this
> mean I can't sell these particular binaries of my program,
> even if I distribute, or offer to distribute the gcc libraries?
If you use only -lgcc (which is required as a rule) there are absolutely
no restrictions at all. If you use the Gnu C library, libg++, or one of
the other libraries licensed under the LGPL, there are some rules that
you must follow, but you can still keep your own code proprietary if you
wish.
If you use bison, or if you take pieces of GPL'ed code and use them to
produce your application, then and only then must you license your whole
program under the GPL. (I think that the bison parser should be LGPL'ed,
but I can use byacc instead until the FSF folks realize this).
> 2.) I link it using the gnu linker, but without the gnu C library,
> and without any supporting libraries....
> 3.) I link the binaries produced by gcc with libraries which I
> write entirely myself....
In both cases, there are absolutely no restrictions at all. I won't even
bother answering point 4, and I've also deleted your insults.
>This started out as a serious note, but it's really hard for me to
>take all of this hair splitting seriously. I guess the moral is,
>if you want to sell your software, which is to say, if you think enough
>of yourself to want to make a living, have a house to sleep in,
>eat three squares a day, and have a family, then don't use gcc.
Ever use a NeXT? All code on that platform is built using a gcc-derived
compiler. And it can be sold. Also, you can make money distributing and
supporting GPL'ed software -- Cygnus folks look well-fed to me.
--
Joe Buck jb...@ohm.berkeley.edu
Bill White
Ok, so now I've read this entire string, and I'm confused. Let's say
I write a program which performs some unnatural act, and I compile it
with gcc. Consider the following scenaria:
1.) I link it with the gcc libraries. Does the result of this
mean I can't sell these particular binaries of my program,
even if I distribute, or offer to distribute the gcc libraries?
program if I offer to give the buyer the sources to gcc,
emacs, gzip, and a photograph of Richard Stallman?
3.) I link the binaries produced by gcc with libraries which I
offer to give the buyer the sources to gcc, emacs, gzip,
gdb, and a tape recording or sun .au file copy of Richard
singing the programmers's-don't-deserve-to-make-a-living
song?
4.) I write a C compiler using Microsoft C. I then compile my
C compiler using libraries I write myself. I then compile
my unnatural act program using my C compiler, and link my
unnatural act program with my own libraries. However, in
one line I use a variable which contains the letter 'G'.
Does Richard Stallman then own the rights to all movies made
of my life, the right to sleep under my front porch, the
right to tell me not to watch Dragnet on Nick at Night, and
the right to make me do Bulgarian sheep scaring dances?
This started out as a serious note, but it's really hard for me to
take all of this hair splitting seriously. I guess the moral is,
if you want to sell your software, which is to say, if you think enough
of yourself to want to make a living, have a house to sleep in,
eat three squares a day, and have a family, then don't use gcc.
Peace,
Bill White
Chris Waters
>In article <BURLEY.93...@apple-gunkies.gnu.ai.mit.edu> bur...@apple-gunkies.gnu.ai.mit.edu (Craig Burley) writes:
>> Would the fact that you have actively been encouraging distribution to whites
>> in the meantime mean that you could be found to have given up your copyright?
>>Very funny.
>>Sigh. :-)
>I don't think you've been following the discussion.
>I made an analogy to a hypothetical "Racist Public License" that prevents
>redistribution to blacks, instead of redistribution without source. The RPL
>gets around limitations on shrink-wrap license the same way the GPL does:
>if someone doesn't accept the license, then they cannot redistribute, not
>because they violated the license, but rather because the license is the only
>thing that gives any redistribution rights at all.
Yes, but I think that the main approach to challenging an RPL would be
through the fact that it violates certain rights pertaining to equal
protection under the law, regardless of race, creed, color, etc. (At
least in the US, though there are probably similar provisions in the
laws of many other countries.)
Attempts to overturn the hypothetical RPL on this basis might be
successful, and might not. I don't think that similar arguments can be
raised with respect to the GPL. Hell, an RPL might well work just fine,
though I doubt if I'd use any RPL'd code, even though I'm a white male. :-)
>That being said, the point of the RPL analogy is, of course, to point out
>certain ways the GPL doesn't work. If a court found the RPL invalid, it
>could very well rule that the originator of the RPL gave up his copyright by
>allowing distribution to whites; the fact that the RPL is the only thing
>allowing redistribution at all is totally irrelevant. But the same could
>happen with the GPL. If a court found the GPL illegal, the argument "well, the
>GPL is the only thing allowing redistribution, and without it you cannot
>redistribute at all" may very well be similarly invalid.
True.
But I suspect that challenges could be mounted against the RPL based on
violation of civil liberties. I don't think that you're going to get
very far claiming that the GPL is a violation of anyone's civil
liberties. So the analogy is flawed.
Thus, while all of what you say about "if a court found the {RPL|GPL}
illegal" is probably true, it only applies *IF* the license is illegal.
Which is questionable in both cases, and also not equivalent in both
cases. The RPL is, at least to my eyes, more likely to be declared as
illegal. But, in any case, *IF* the GPL is found to be illegal, then of
*course* it's provisions are not going to apply! But that's a *big* if!
'Course, I ain't a lawyer, so none of the above is anything but my own,
very possibly flawed, opinion.
--
Chris Waters | the insane don't | "Without gratuitous sex, we would be no
xt...@netcom.COM| need disclaimers | better than animals." --M. Swanwick
Craig Burley
In article <BURLEY.93...@apple-gunkies.gnu.ai.mit.edu> bur...@apple-gunkies.gnu.ai.mit.edu (Craig Burley) writes:
> Would the fact that you have actively been encouraging distribution to whites
> in the meantime mean that you could be found to have given up your copyright?
>Very funny.
>Sigh. :-)
I don't think you've been following the discussion.
Silly, I've been half the damn discussion practically!
[...]
That being said, the point of the RPL analogy is, of course, to point out
certain ways the GPL doesn't work. If a court found the RPL invalid, it
could very well rule that the originator of the RPL gave up his copyright by
allowing distribution to whites; the fact that the RPL is the only thing
allowing redistribution at all is totally irrelevant. But the same could
happen with the GPL. If a court found the GPL illegal, the argument "well, the
GPL is the only thing allowing redistribution, and without it you cannot
redistribute at all" may very well be similarly invalid.
The point is, the courts do not cancel copyright status on things just
because the copyright holder distributes the work to people. That would
be stupid. Nor do the courts cancel copyright status just because
the copyright holder behaves in a non-PC fashion. That would probably
bring armed insurrection, also pretty stupid. I really doubt an RPL
could ever cause someone to have copyright status forcibly canceled --
more likely, the courts might impose a short-term license agreement
that allowed restitutional distribution to the victims of the racist
policy. And, of course, even if something like that happened, it would
have _no_ impact on the legitimacy of the GPL, since the GPL does not
have anything nearly as weird as a racist policy, _unless_ (and this
keeps getting ignored, but it's a fact) many _other_ license agreements
currently in force also could fall victim in that way.
This whole idea that somehow the way one distributes a clearly copyrighted
item might result in the courts declaring the item public domain seems
pretty stupid to me, at least until someone posts a clear, verifiable
law or case that shows exactly this. As in "XYZ Corp. had a copyright
on document/program FOO, distributed it only to companies doing business
in South Africa, therefore FOO is now public domain". In all I've
read about copyright law, I've never come across anything like this that
I can recall. Obviously a case like "Adobe claimed the PostScript
language was public domain, then attempted to retract it, and the
courts ruled that, once declared PD, it is PD" is not applicable, since
the FSF has never claimed its GPL software is PD, and saying it is
free is not the same thing as saying it is PD (and every intelligent
person knows that -- especially lawyers and judges).
Stop posting suppositions, start posting facts. Go to a law library,
ask a lawyer, anything but wasting net.resources on this drivel until
you have some solid facts and/or cases to post. Then we'll all be
very interested. (I mean this. In fact, if y'all _don't_ come back
within a month or so either with definitive case histories supporting
your position or a "mea culpa", I think a bunch of us will just assume
you'd rather waste everyone's time with scare tactics than do some
actual helpful research. Then again, maybe everyone else has figured
this out, and that's why I'm basically the only one bothering to
respond to this stuff.)
Ken Arromdee
Let me get this correct, I'm supposed to consult a lawyer before arguing with
you? Just want to make sure.
Michael Golan
>Therefore, if I sue someone who distributes RPL software for a black, I would
>not be suing them for violating the RPL. Rather I would be suing them because
>the default is that they have no redistribution rights at all. If the RPL
>violates public policy and is invalid, that doesn't mean they can now
>distribute to blacks; rather it means that since the RPL is invalid, there is
>no license at all, and they can't distribute to _anyone_, of any race. It
>would still be copyright infringement to distribute to blacks (as well as
>whites), and my suit would continue.
No:
1. the RPL can not be "illegal". It is TOTALLY LEGAL to be racial in
this country. Try to join the KKK. Hack, their software might have an RPL.
2. the RPL will be invalid for the same reason the GPL is in valid - no
one signed it or agreed to it.
3. Give that you gave your software for free (ftp-able), and tried to
abuse the privileges the copyright law has provided - yes, most likely
the court will decide you have lost the copyright. Same with the GPL
as far as many lawyers are concerned.
-- Michael
Michael Golan
>I pointed out that the GPL has nowhere near the arbitrary kind of
>restriction that shrink-wrap licenses have ("you cannot disassemble
>the program") or anything like "no distribution by young black males".
>The GPL is not much different from many other license agreements for "free"
>code that nevertheless is copyrighted -- for example, many include
>restrictions on "distribution for commercial gain", which the GPL does
>not (in the main).
the GPL is like other licenses? you must be kidding! Every lawyer
I heard giving an opinion on it agrees it is very different. It demands
that I give up *MY* rights. My rights to my own copyright work.
>The point is that the GPL isn't particularly unsual or out of line from
>the point of view of looking at a wide variety of software
>licenses that are widely used and have been unchallenged (or withstood
>challenges) in the courts.
Which "withstood challenges"? Most copyright notice that claim "free but not
for commercial use" are probably invalid, too, indeed. Do you know of one
that the court has "validated"? you can not put arbitrary restrictions on
the protection the copyright gives you (and you give up, e.g. by making
the software ftp-able). Thats the opinion of most lawyers I talked to or
heard giving advice on this; no, I won't post names etc.
>The only things really interesting
>about the GPL is that it protects some very high-quality and
^^^^^^^^^^^^^^^^^ ya right...
>desirable software, does not prevent distribution for commercial
>gain, requires source distribution, and attracts the ire of people
^^^^^^^^^^^^^^^^^^^^^^^^^^^ you mean, actively take away my own copyr.
>who want to pillage the software it protects for their own uses.
>>In comp.lang.c++ this week, someone asked for a "free C++ compiler" and got
>>the answer "g++". No one bothered to correct that the software is NOT free.
>A distribution (copy) of the software can be obtained for free, and
>legally. That is what most people mean when they ask a question like
>the above. Some people mean "public-domain", but unless an answer
>claims g++ is public-domain, claiming it is free is not incorrect.
>No court, unless ruled by an insane judge, would rule that attaching
>the word "free" to something people could obtain for free but couldn't
>redistribute copies of for free causes that thing to lose its copyright
>status.
On the contrary.
lawyer: Do you swear to tell the whole truth and nothing but the truth?
rms: I do
lawyer: Isn't it true that your company advertise your software as free?
rms: well, not exactly, we have a GPL
lawter: please answer the question
rms: well, yes.
lawyer: Isn't your company's name is in fact, the FREE software foundation?
rms: yes
lawyer: didn't many people on many occasions referred to your software as
free, and your copmany didn't object?
rms: yes
lawyer: Isn't it in fact the whole idea of your not-for-profit org. to
make free software for everyone to use, for free?
rms: yes
this goes on and on; finally
lawyer: can you explain the term GNU
rms: well GNU is Gnu not Unix, which is gnu not unix not unix which is
gnu not unix not unix not unix...
Seriously though, with rms' social claims and the word free everywhere,
it will be reasonable for a judge to conclude the software IS in the public
domain, or that the "violators" had reason to believe it is because of
the way the FSF treated it etc, and dismiss the case.
> stuff deleted
>That's idiotic, obviously (so idiotic
>that it causes me to think that anyone holding that view must be
>an idiot, hence the hostility in my posts responding to yours -- sorry).
The most ugly apology I have seen yet. congrats! I have erased your other
rude comments. I'll let the netters decide who is the "idiot".
>Even the _copyright holder_ of a book may give away free copies to
>anyone and everyone, and advertise the book as "free", and the courts
>won't take away that person's copyright! Why would software be any
>different?
Said who? give away SOME copies - yes. give them to anyone and advertise it
and such? go ask a lawyer, which you suggest everyone else do. The lawyers
I have asked seem to think such a writer might have lost most of his
copyright rights in such a case.
>> If the post has said "Borland C++" you bet your ass someone (probably
>> from Borland, too) would correct it. Such claims for "free software" about
>> the GNU software are common, and will most likely lead the
>> court NOT to accept the special binding of the GPL.
>Wrong. If that were true, then the "Free Software Foundation", which
>created and publishes the GPL, would have disappeared long ago, thanks
>to various people who are more interested in pillaging FSF code than
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ can you give names? companies?
maybe you refer to me?
>writing their own.
>Furthermore, the courts are hardly going to pay attention to whether
>the FSF constantly rebuts claims that its software is "free" made on
>forums like USENET.
go ask a lawyer, again. You must ACTIVELY protect your rights! For example,
Princeton University, which OWNS a road near by, would loose
it if they didnt close it for public traffic every so often? The road
would become public property because ... they let the public use
it as if it WAS public property. Many lawyers seems to believe the same
holds for software. True, you don't have to protect ALL your right ALL
the time, but you must show some "concern" for it.
>The GPL is legally binding. That is enough of a counter detail. Plenty
>of lawyers in the USA of already been paid plenty of money (by any normal
>person's standards anyway :-) and determined that the GPL is legally
>binding. Otherwise there'd be widespread use of GCC and GNU EMACS source
>code in proprietary products. Various organizations have threatened
>to take this step if the FSF didn't give them a special license, and
>they all, to my knowledge, backed down when the FSF didn't capitulate.
Give names, not "various organizations". You sound like M$ came to
take the code away, or some other (large) "organization".
"Plenty of lawyers" determined that the GPL is not binding and the
copyright w/o it are very weak or non-existent. All said that unless
it goes to court, it is meaningless anyway.
If you have a legal opinion you paid for, from a known lawyer - post it!
Don't give us this "I know" "I have seen" "I have". You are talking from
the same position I do, yet you constantly claim I am a fool
and your *opinions* are the only truth.
>Why is the GPL stretching it too much?
I modify some GNU code. I give a bin copy to a friend. I explain to him
he won't get the source. Then he gives it away. Then I get sued for the
source. No other license make it appear I am liable ... (yes I can sue
my friend, I know. And you will claim my "private" contract with him was
illegal. maybe. but the GPL sure is a LOT different. Especially, e.g.
the Bison case).
>Please post case histories or reference numbers for the cases you mention,
>or explain exactly how they managed to register their copyrights on
>software and still lose the copyright because they used the word "free".
Please post a single case of code that was labeled free, got its
copyright challenged, and won. As far as I know, no such case exists,
so this is all lawyer talk. The cases I refer to are, e.g. Intel's loss
to NEC on 86' micro code. It seems quite easy to loose your copyrights ...
-- Michael
Michael Golan
>In <1993Mar4.1...@blaze.cs.jhu.edu> arro...@jyusenkyou.cs.jhu.edu (Ken Arromdee) writes:
>>In article <BURLEY.93...@apple-gunkies.gnu.ai.mit.edu> bur...@apple-gunkies.gnu.ai.mit.edu (Craig Burley) writes:
>>> Would the fact that you have actively been encouraging distribution to whites
>>> in the meantime mean that you could be found to have given up your copyright?
>>>Very funny.
>>>Sigh. :-)
>>I don't think you've been following the discussion.
>>I made an analogy to a hypothetical "Racist Public License" that prevents
>>redistribution to blacks, instead of redistribution without source. The RPL
>>gets around limitations on shrink-wrap license the same way the GPL does:
>>if someone doesn't accept the license, then they cannot redistribute, not
>>because they violated the license, but rather because the license is the only
>>thing that gives any redistribution rights at all.
>Yes, but I think that the main approach to challenging an RPL would be
>through the fact that it violates certain rights pertaining to equal
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
>protection under the law, regardless of race, creed, color, etc. (At
>least in the US, though there are probably similar provisions in the
>laws of many other countries.)
Ever heard of the KKK? This is a FREE (not like GNU's software :-) COUNTRY.
it is legal to be racial. There are only specific rules concerning
racial equality, usually related to government contracts, housing,
public services, etc. A "private" contract between us is otherwise legal,
even if racial. The real question is if it is VALID, since it relies only
on the copyright law for its binding.
I and others think the FSF/GPL try to force us into socialized software,
especially given rms opinions that prgrammers should be paid by the
government or that some of us make too much money. Which is why I think
it is invalid, since it is trying to use the copyright *priviledges*,
just like the RPL, to impose its own agenda, something the copyright law
never intended. (the GPL comments translate real well to RPL, btw, try
it sometimes ... we don't object that blacks use or write software ... we want
to protect you ... we believe ...)
There is a border, I think we agree, after which an XPL would be invalid,
because of its restrictions; The question then: is the GPL over this border?
if not, what would be? Can we agree the GPL is probably close to this border?
What about "only hackers may give away code"? "non-commerical use"? "only
if my name appear in red on the manual, which you must supply if requested"?
-- Michael
ps. I'm against software patents. It's a shame I can't send money to the LPF
given its members and thier reputation. Ya I know LPF!=FSF etc.
John L. Coolidge
>In article <1n4qi0...@shelley.u.washington.edu>,
>t...@hardy.u.washington.edu (Tim Smith) wrote:
>> cool...@apple.com writes:
>> >I could be wrong, but I don't recall the SPA suing anyone. Instead,
>> >they use threats of lawsuits as a nuisance factor to coerce companies
>> >into paying (often for software which they in fact legally own but
>> >cannot conveniently prove ownership of). Most companies would rather
>>
>> Wouldn't it be easy for most companies to prove ownership? There should be
>> a purchase order in their files, and maybe something in their mail department
>> logging the receipt of the software.
>It generally is easy to prove ownership of legal copies. What the SPA does
>is count the number of copies existing on various media, count the purchase
>records, and compare the numbers.
Sure, if you have purchase records. I've worked places (university
settings, mostly) where it was common practice to buy any low-cost
(sub-$200 or so) software packages off-budget on individual credit
cards and be reimbursed by the organization. The reimbursement forms
don't specify the program in question and no purchase orders ever
existed. If the manuals and disks get lost (as happened more than once
in office moves and, probably, theft), how in the world do we prove
ownership? Should the SPA (or a court, for that matter) be allowed to
conclude that the copies are illegal _simply because_ of the lack of
records?
It seems to me that "innocent until proven guilty" in this case
should require the body asserting illegality (SPA, prosecuter, etc)
to produce clear evidence of piracy (multiple copies with the same
serial number, a serial number registered to someone else, etc)
before any fines could be levied. It is not up to me, or my
organization, to prove that I _did not_ violate the law; it is up to
the SPA (or whoever) to prove that I did.
--John
May the forces of evil become confused on the way to your house.
-- George Carlin
Ken Arromdee
>was considered by the courts to have identical results compared with
>"must not distribute to young black males", and we'll have something.
>Then post a case showing how copyright status was lost due to overly
>restrictive requirements in a license. Until then, this thread about
>how the GPL is "fatally flawed" is fatally flawed.
Your argument is fatally flawed, since I haven't even used the words "fatally
flawed" (well, until now).
>>>(Obviously, the logic "rms says the law will treat A as B even though
>>>A and B are different" cannot be universally applied, else the law will
>>>treat a tadpole as a convertible subordinated debenture. With the
>>>top down. :-)
>>But what about when A and B are different methods of achieving the same
>>result?
>In what way are A and B achieving the same result if A is "can redistribute
>only if source provided to" and B is "can redistribute only to young
>black males"?
It's not. A is "can redistribute only if source provided (because of default
lack of permission without contract)", and B is "can redistribute only if
source provided (because it tries to hold you to a shrink-wrap contract)".
The don't-distribute-to-blacks case doesn't have the same result, of course. I
used it only to show that certain arguments apply to both it and the GPL, not
to claim identical results; at the risk of introducing more letters, comparing
an argument for X to an argument for Y is not the same as comparing X to Y.
Joe Buck
>Let me get this correct, I'm supposed to consult a lawyer before arguing with
>you? Just want to make sure.
Well, if you wish to continue making wild assertions about the GPL,
despite the fact that it was crafted by trained lawyers and you yourself
have no legal training, and companies like Motorola, NeXT, Data General,
Intel, Wind River Systems, etc have worked with gcc ports and treated the
GPL as valid ... you can continue to post, but I see no reason
why anyone should continue to read what you post because you're operating
beyond your level of competence.
For now, the bottom line is this: the GPL has not yet been tested in court
(but then, neither have the vast majority of legal documents in this country).
However, corporate America is treating it as valid, thus RMS's points
about it being effective even if it may not be perfect.
--
Joe Buck jb...@ohm.berkeley.edu
Rahul Dhesi
arro...@jyusenkyou.cs.jhu.edu (Ken Arromdee) writes:
>If "grant permission to X" is a subterfuge for "deny
>permission to !X" when X is whites, then it should also be so for other values
>of X--and that makes the GPL a shrinkwrap license.
Can we have a better, more universal definition of "shrinkwrap
license"? I always thought it meant "a license that is
shrink-wrapped", quite independent of the actual terms of the
license.
--
Rahul Dhesi <dh...@cirrus.com>
also: dh...@rahul.net
Craig Burley
I and others think the FSF/GPL try to force us into socialized software,
especially given rms opinions that prgrammers should be paid by the
government or that some of us make too much money. Which is why I think
it is invalid, since it is trying to use the copyright *priviledges*,
just like the RPL, to impose its own agenda, something the copyright law
never intended. (the GPL comments translate real well to RPL, btw, try
it sometimes ... we don't object that blacks use or write software ... we want
to protect you ... we believe ...)
This is the funniest bit of confused frothing-at-the-mouth I have
seen so far this year. Gets my vote for the "Hall of Shame" for USENET
'93, pending more posts from this guy.
Craig Burley
the GPL is like other licenses? you must be kidding! Every lawyer
I heard giving an opinion on it agrees it is very different. It demands
that I give up *MY* rights. My rights to my own copyright work.
No it doesn't. No wonder they "agree" with you -- you misstate what
the GPL does.
>The point is that the GPL isn't particularly unsual or out of line from
>the point of view of looking at a wide variety of software
>licenses that are widely used and have been unchallenged (or withstood
>challenges) in the courts.
Which "withstood challenges"? Most copyright notice that claim "free but not
for commercial use" are probably invalid, too, indeed. Do you know of one
that the court has "validated"? you can not put arbitrary restrictions on
the protection the copyright gives you (and you give up, e.g. by making
the software ftp-able). Thats the opinion of most lawyers I talked to or
heard giving advice on this; no, I won't post names etc.
I claim not a single lawyer with a proven case record in copyright law
would seriously claim that putting software up for ftp causes you to
lose copyright. No wonder you won't post names.
>The only things really interesting
>about the GPL is that it protects some very high-quality and
^^^^^^^^^^^^^^^^^ ya right...
Compared to code _you_ write, I assume? Sigh. Proof that most of
this anti-GPL attitude is mostly the hacker's equivalent of
"mine's bigger than his".
>desirable software, does not prevent distribution for commercial
>gain, requires source distribution, and attracts the ire of people
^^^^^^^^^^^^^^^^^^^^^^^^^^^ you mean, actively take away my own copyr.
False.
>who want to pillage the software it protects for their own uses.
>No court, unless ruled by an insane judge, would rule that attaching
>the word "free" to something people could obtain for free but couldn't
>redistribute copies of for free causes that thing to lose its copyright
>status.
On the contrary.
lawyer: Do you swear to tell the whole truth and nothing but the truth?
rms: I do
lawyer: Isn't it true that your company advertise your software as free?
rms: well, not exactly, we have a GPL
No, "rms: yes." There's no reason he has to hem or haw about this.
[...waste of space omitted...]
lawyer: Isn't it in fact the whole idea of your not-for-profit org. to
make free software for everyone to use, for free?
rms: yes
So what? That has no impact on whether copyright is preserved. If you
claim otherwise, you either show complete ignorance of copyright law,
or you lie through your teeth.
this goes on and on; finally
This anti-GPL frothing is the only thing that "goes on and on" around
here....
Seriously though, with rms' social claims and the word free everywhere,
it will be reasonable for a judge to conclude the software IS in the public
domain, or that the "violators" had reason to believe it is because of
the way the FSF treated it etc, and dismiss the case.
And then it will be reasonable for that judge to be removed from the
bench. There is _no_ legal precedent for a copyrighted work losing
its copyright status because it was advertised or sold as "free", at
least not since the Berne Convention was adopted.
I'll let the netters decide who is the "idiot".
I'm very comfortable with that, based on the email I've been getting,
they've already made the right choice. That is, the ones who've
already learned copyright law or have bothered to go to a library.
>Even the _copyright holder_ of a book may give away free copies to
>anyone and everyone, and advertise the book as "free", and the courts
>won't take away that person's copyright! Why would software be any
>different?
Said who? give away SOME copies - yes. give them to anyone and advertise it
and such? go ask a lawyer, which you suggest everyone else do. The lawyers
I have asked seem to think such a writer might have lost most of his
copyright rights in such a case.
I claim you lie, or misrepresent the lawyers' position. Especially since
many organizations, _including_ lawyer-happy ones like Apple, put their
own copyright code up for ftp and include license restrictions with it,
and probably would not do so if they feared the code would become
"virtual public domain". The fact is: you do not lose your rights to
the copyright of a thing simply by distributing it for free to all
comers (the essence of ftp).
>Wrong. If that were true, then the "Free Software Foundation", which
>created and publishes the GPL, would have disappeared long ago, thanks
>to various people who are more interested in pillaging FSF code than
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ can you give names? companies?
maybe you refer to me?
Somebody else just posted a list, I won't bother repeating it here.
Data General was one, for example.
>Furthermore, the courts are hardly going to pay attention to whether
>the FSF constantly rebuts claims that its software is "free" made on
>forums like USENET.
go ask a lawyer, again. You must ACTIVELY protect your rights!
No you mustn't, except in certain cases that have nothing to do with
making your _own_ work freely available to others for distribution and
redistribution-with-attribution.
For example,
Princeton University, which OWNS a road near by, would loose
it if they didnt close it for public traffic every so often? The road
would become public property because ... they let the public use
it as if it WAS public property.
You're claiming they own a _copyright_ on a road? Stick to the topic:
we're talking about copyright law, not property law and such. There
are plenty of cases illustrating your point about the road, indeed, but
_none_ of them have any impact on copyright law.
Does anyone else find it ironic that this guy whines about the GPL because
it causes the FSF to _give up_ copyright involuntarily, while in an
earlier rant, _another_ person with apparently _far_ more legal
experience (and certainly more eloquence), whined precisely because
he felt the FSF could someday decide to _change_ the GPL so free
distribution was no longer permitted, thus asserting its continuing
ownership of already-copyrighted-and-freely-distributed code by
stopping free distribution?
Many lawyers seems to believe the same
holds for software.
Name one lawyer experienced in copyright law who thinks software
is more like a road than like a book.
True, you don't have to protect ALL your right ALL
the time, but you must show some "concern" for it.
As the FSF does, witness simply these counterarguments rms and FSF
supporters (like myself) make in public forums contradicating blatantly
incorrect legal information posted by fools and liars.
Give names, not "various organizations". You sound like M$ came to
take the code away, or some other (large) "organization".
As I said, others have posted this info.
"Plenty of lawyers" determined that the GPL is not binding and the
copyright w/o it are very weak or non-existent. All said that unless
it goes to court, it is meaningless anyway.
Name one. You asked me (implicitly any FSF supporters) for names, and
they've been posted. Now it's your turn. I bet you turn your tail
and hide, without even having the courage to admit your mistake.
If you have a legal opinion you paid for, from a known lawyer - post it!
The FSF has a legal opinion it paid for, as have several companies
named in previous posts, as must be obvious given the current
behavior of the industry as regards FSF software (in that they honor
the GPL). Instead of attributing this to FUD, trying talking to a
lawyer for one of these organizations.
Don't give us this "I know" "I have seen" "I have". You are talking from
the same position I do, yet you constantly claim I am a fool
and your *opinions* are the only truth.
They aren't my opinions. They are prevailing fact in the industry.
I'm not sure copyright _should_ apply to software, but that's my
(hesitant) opinion. All I've been doing, ultimately, is pointing out
that making wild claims about the GPL being "worthless" is something
that's been going on for years but has _never been proven or even
effectively argued against prevailing practice_.
>Why is the GPL stretching it too much?
I modify some GNU code. I give a bin copy to a friend. I explain to him
he won't get the source. Then he gives it away. Then I get sued for the
source. No other license make it appear I am liable ... (yes I can sue
my friend, I know. And you will claim my "private" contract with him was
illegal. maybe. but the GPL sure is a LOT different. Especially, e.g.
the Bison case).
Some other licenses do make it appear you are liable. And how is the
GPL so vastly different here? If you make copies of (to use another
poster's concise example) Lord of the Rings for friends, but have added
300 of your own pages (hence making it a derived work in essentially
the same way as binaries compiled from GNU source), are you claiming
you won't/shouldn't be liable in a copyright infringement suit made
by the owner of the LotR copyright??
>Please post case histories or reference numbers for the cases you mention,
>or explain exactly how they managed to register their copyrights on
>software and still lose the copyright because they used the word "free".
Please post a single case of code that was labeled free, got its
copyright challenged, and won. As far as I know, no such case exists,
so this is all lawyer talk. The cases I refer to are, e.g. Intel's loss
to NEC on 86' micro code. It seems quite easy to loose your copyrights ...
Talk to the FSF lawyers about cases they've won out of court. Again,
I don't have to (and will certainly not bother to) post a case illustrating
current practice, any more than I feel a necessity to post a case showing
that it is just as illegal for person A to shoot person B to death when
they are standing on opposite sides of a county line as when they aren't.
But if you're going to make a wild claim countering prevailing behavior,
such as that it is entirely legal to shoot someone in a county different
from the one in which you're standing, it is entirely fair and reasonable
to ask you to post a case or clear and pertintent legal point of law
confirming your point.
Since you're incapable of doing that (e.g. you've refused to provide any
details on "Intel's loss to NEC on 86' micro code" -- is that pre- or post-
Berne Convention, I ask you? Is it code they'd registered? Etc.),
your arguments are a waste of time.
Craig Burley
In article <BURLEY.93...@apple-gunkies.gnu.ai.mit.edu> bur...@apple-gunkies.gnu.ai.mit.edu (Craig Burley) writes:
>Okay, please post a case that shows that "must distribute with source"
>was considered by the courts to have identical results compared with
>"must not distribute to young black males", and we'll have something.
>Then post a case showing how copyright status was lost due to overly
>restrictive requirements in a license. Until then, this thread about
>how the GPL is "fatally flawed" is fatally flawed.
Your argument is fatally flawed, since I haven't even used the words "fatally
flawed" (well, until now).
I know, but you joined the fray on the side of the person (who I think
has since turned tail and ran) who made that claim. I said "this thread",
not "your posts", for a reason.
>In what way are A and B achieving the same result if A is "can redistribute
>only if source provided to" and B is "can redistribute only to young
>black males"?
It's not. A is "can redistribute only if source provided (because of default
lack of permission without contract)", and B is "can redistribute only if
source provided (because it tries to hold you to a shrink-wrap contract)".
Yes, but the GPL is not a shrink-wrap license, nor is it a subterfuge
that would be considered the same as a shrink-wrap license by a judge.
An example of what probably _would_ be a subterfuge is claiming "it's
not a shrink-wrap license, it's a cardboard-box license, because it was
in a cardboard box that wasn't shrink-wrapped" when the license clearly
reaches the customer and limits him or her in the same manner as
classic shrink-wraps. The GPL isn't like this, however, as any intelligent
person realizes.
The don't-distribute-to-blacks case doesn't have the same result, of course. I
used it only to show that certain arguments apply to both it and the GPL, not
to claim identical results; at the risk of introducing more letters, comparing
an argument for X to an argument for Y is not the same as comparing X to Y.
Nevertheless, there never _was_ any persuasive logic or argument that
the GPL would be considered "subterfugually" equivalent to shrink-wrap
licenses, _or_ that copyright privileges are lost for products protected
by licenses found to be partially invalid, was there? Therefore the
whole discussion was pretty much a waste of time, as expected.
Craig Burley
Can we have a better, more universal definition of "shrinkwrap
license"? I always thought it meant "a license that is
shrink-wrapped", quite independent of the actual terms of the
license.
Here's what I would suggest, though it might well not fit the
prevailing definition used in the trade press:
A shrink-wrap license is a document that is provided with a copyrighted
work within the packaging of the work itself, such that the
recipient (purchaser) must somehow confirm agreement to the
license (such as by tearing it along with the envelope to access
the contents, which include the work itself) via a method that
is generally post-receipt, and that adds restrictions to the
recipient's use (other than commonly accepted restrictions on
redistributing copies) of the work.
The GPL shares the "packaged with the work itself" attribute, but
not the "must somehow confirm agreement to" attribute, and most
definitely not the "adds restrictions to use" attribute (the point
I think the courts probably found most unsupportable).
Note that, in a sense, the text in a book that says, after
"Copyright...", "All rights reserved. No part of this book may
be copied, ...", has at least as much in common with my definition
of a shrink-wrap license as does the GPL, and I'm pretty sure
that text would hold (and probably already has held) up in court.
What I'm most uncertain about, and omitted from my suggested
definition, is this item:
A shrink-wrap license is typically not available for the
recipient until after receipt, and the method the recipient
uses to refuse agreement to the license includes not making
any use of the work (e.g. not "reading" it) at all.
Again, the GPL and this-book-may-not-be-copied text are much more
similar to each other in this regard than either is similar to the
classic shrink-wrap license. And a license that comes with,
say, ftp'able software (such as on ftp.apple.com?) but disallows,
say, use in nuclear-weapons development, is more like a shrink-wrap
license than the GPL.
The reason I omitted the above in my first definition is that
I don't really have any idea how the courts' rulings on shrink-
wrap licenses would change if the license were clearly printed
on the packaging itself, especially given that that doesn't
help somebody receiving the package via mail order. (Such a
person still has the problem of starting to open the package,
discovering and reading the license agreement, deciding not to
abide by it, and thus being required by the agreement to not
open and use a work for which he or she has already paid and
must now pay shipping charges to return.)
Basically, shrink-wrap licenses are somewhat like buying a book,
getting it home, and discovering, in place of the usual
don't-copy-this-book text, a license (attached to a seal on the
book) saying:
If you break this seal, you accept this license, which
requires that you read the book from beginning to end,
without skipping forward or rereading already-read pages,
nor may you copy individual pages for your own use or
write or otherwise deface any page in this book or its cover.
I doubt the courts would allow any litigation trying to enforce
such a license to succeed, just as they apparently have not allowed
it to succeed for shrink-wrap licenses (the silly restrictions on
use above being roughly equivalent to disallowing making backup
copies and disassembling the code). But in neither case would the
courts rule that copyright had been entirely given up by virtue
of imposing a less-than-perfect license, obviously.
Craig Burley
>to various people who are more interested in pillaging FSF code than
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ can you give names? companies?
maybe you refer to me?
Somebody else just posted a list, I won't bother repeating it here.
Data General was one, for example.
Woops, I meant to say that Data General was an example of a company
that releases GPL software and (apparently) believes the GPL would
stand up in court, else they would simply hoard their many changes
to (I suspect) GCC to support DG machines and release such ports
of GCC as proprietary.
As far as people who are interested in pillaging FSF code, there are
several who have posted complaints to this newsgroup that the FSF
is hurting them (or others) because they can't take the code and
either "give it away as PD" or (the effect of the former) make
proprietary products out of it. I'll leave it to them to identify
themselves, if they still lurk on g.m.d. (boy, is _that_ asking
for trouble! :-).
Oh, and you could try asking rms which companies have threatened
to release GNU code in proprietary form and then backed down when
the FSF threatened a lawsuit. I seem to recall that Prime at
least explored whether the FSF would make exceptions to the GPL
for them, but don't know if they actual made a threat. Whether
rms wishes to answer you (and thus impugn the reputation of
the company, perhaps unfairly) is up to him, but he's probably
documented this sort of thing in the past. He might only state
that such a thing _has_ happened at least once. It is up to you
whether you want to then admit your mistake, or add to your
list of them by calling such a statement by rms a lie.
Michael Golan
>Well, you have the right idea, but I think you still miss the point. :-)
>The GPL is written in such a way that it doesn't "really" bind people who dont
>agree to it; it restricts people who don't agree not by forcing them to follow
>the contract, but by depending on the default lack of rights to copy at all.
As pointed out before, the court may rule that, since the only use of GNU
software was with a GPL, and the GPL is invalid, the software is in the
public domain now, rather than "everyone is now in violation".
I admit I don't know of a specific case details where copyr. was "taken away"
by publishing. One exmaple given to me was a book on bridge designs. Someone
used a design from the book to build a REAL bridge (e.g. for cars, not for
ethernet packets :-), and the author sued for copyr. violations.
The court decided that by publishing the book, the author gave up the rights
for the copyrighted design itself - anyone can buy the book and build
bridges based on the design. (Note - we are not talking "patent" or
"algorithm" here, but an actual design, with full details.) I understand all
other "design books" fall into this category, even though they have not
given up their copyr. explicitly. Imagine if you bought a book with
home furnishing suggestions and was later sued for furnishing your home in
the way shown.
One may claim that that the GNU software, being "free" has given up its
rights in the same way; the "profits" from the distribution are the
modified code sent back, fame, etc. This is what I understood that
lawyers (I talked to or heard) claim.
Last comment: supposingly, many "organizations" tried to make FSF software
into proprietary product, but were scared away by FSF/GPL lawyers. I have
plenty of suggestions on how to circumvent the GPL, which I think anyone
serious would attempt to do. For example, create a new language D. modify
gcc to compile it; modify, in a non-reversable way, Emacs source into D.
sell emacs binaries and a (useless) D source (no one said u need to
give away a compiler!). This isn't in violation as far as I can tell.
I don't think it is sabotage either -- could be a sensible way to proceed
with any source (and it has been done!). There are many other ways, e.g.
DESTROY all sources! compile into Asm. and claim you do all your work at
that level (let the FSF prove you lie). Or, when the FSF does sue, sell
the source to a 3rd party and go bankrupt. etc etc. While all of these
could be attacked (I thought them up in 5 minutes), a good combination
is likely to make it very hard, or impossible, for the FSF to sue successfully.
-- Michael
Christopher Davis
MG> As pointed out before, the court may rule that, since the only use of
MG> GNU software was with a GPL, and the GPL is invalid, the software is
MG> in the public domain now, rather than "everyone is now in violation".
The GPL does not restrict use. It only *allows*, and what it allows is
distribution. Quoting from v2:
GPL> Activities other than copying, distribution and modification are not
GPL> covered by this License; they are outside its scope. The act of
GPL> running the Program is not restricted, and the output from the
GPL> Program is covered only if its contents constitute a work based on
GPL> the Program (independent of having been made by running the Program).
GPL> Whether that is true depends on what the Program does.
Have you *read* it?
And, yes, granted that the court *may* rule that way. Sillier rulings have
been made.
--
* Christopher Davis * <c...@eff.org> * <c...@kei.com> * [CKD1] * MIME * RIPEM *
226 Transfer complete. 17512509 bytes received in 5.2e+02 seconds (33 Kbytes/s)
Chris Waters
>As far as people who are interested in pillaging FSF code, there are
>several who have posted complaints to this newsgroup that the FSF
>is hurting them (or others) because they can't take the code and
>either "give it away as PD" or (the effect of the former) make
>proprietary products out of it.
Actually, this reminds me of one of the few objections to the GPL I've
seen that I was actually somewhat sympathetic with. Someone who was
developing software under a government grant, who *had* to make the
results of his work PD (since it was developed with public funds).
OC, this person was no more limited wrt FSF code than he was wrt to any
other copyrighted code, and his argument that he *should* be able to use
FSF code since it is "free" was clearly specious, but I did have a
momentary pang of sympathy before reason reasserted itself. What the
guy didn't realize is that he was trying to help the *government* steal
FSF code, which is really not much different than trying to steal it for
your company.
And, after all, he could still use the FSF tools to develop his
software, just like the rest of us, and the only limitation was that he
couldn't incorporate GPL'd code, just like all of us who do commercial
development. And, like the rest of us, this is unlikely to be a problem
except, perhaps, in the case of Bison output. (When is that stuff going
to be LGLP'd?)
--
Chris Waters | the insane don't | "Pfui!" -- Nero Wolfe
xt...@netcom.COM| need disclaimers |
Paul Crowley
>I think my original point also still applies: rms says that the law doesn't
>work like a machine and can consider intent, so it's still illegal to
>circumvent the GPL with "legal" separate distribution tricks. But if rms is
>right, the law would work similarly with the GPL itself, so that the fact that
>the GPL doesn't fit the definition of a shrinkwrap license doesn't mean that
>the law might not treat it as one anyway.
As I understand it, the courts ruled that SWLs are illegal because you
have to agree to it before you read it. It is because they are
different *on the legal point which failed SWLs* that the GPL will
probably stand up.
If the GPL doesn't work, the upshot is that no-one has any right to give
their friends copies of GNU software at all, not that you can copy and
modify it freely without regard to the terms of the GPL, since the GPL
is the only thing that gives you the right to copy it.
On this "hidden derivative" bit, can I throw in another hypothetical?
Supposing I modify ghostview so that it uses the Motif widget set, can I
redistrubute that? Now suppose I'm OSF (who own Motif), can I still
redistribute it? If I want to get around the GPL, could I put together
a library of useful functions and sell it, and later modify a GNU
product to use that library?
Since the courts aren't a machine, I'd guess it might come down to
whether the court rules that my primary intent is that the library be
bought so that my modifications to the GNU software can be used. It
doesn't seem to me too hard to obscure that intent with the technique I
describe.
Why does this discussion seem to come up every couple of months, anyway?
__ _____
\/ o\ Paul Crowley p...@dcs.ed.ac.uk \\ //
/\__/ Trust me. I know what I'm doing. \X/ Fold a fish for Jesus!
Russell Nelson
In article <BURLEY.93...@apple-gunkies.gnu.ai.mit.edu> bur...@apple-gunkies.gnu.ai.mit.edu (Craig Burley) writes:
>Stop posting suppositions, start posting facts. Go to a law library,
>ask a lawyer, anything but wasting net.resources on this drivel until
>you have some solid facts and/or cases to post.
Let me get this correct, I'm supposed to consult a lawyer before arguing with
you? Just want to make sure.
No. The problem, Ken, is that it doesn't really matter what you
think. Your opinion is meaningless in this context. A court is not
a democracy. The only opinion that matters is the judge's.
-russ <nel...@crynwr.com> What canst *thou* say?
Crynwr Software Crynwr Software sells packet driver support.
11 Grant St. 315-268-1925 Voice | LPF member - ask me about
Potsdam, NY 13676 315-268-9201 FAX | the harm software patents do.
Ian Lance Taylor
>the GPL is like other licenses? you must be kidding! Every lawyer
>I heard giving an opinion on it agrees it is very different. It demands
>that I give up *MY* rights. My rights to my own copyright work.
In what way does the GPL make you give up your rights to your own
copyright work? Keep in mind that if you disassemble Microsoft Word
and modify it a bit, you will still not be permitted to sell it or
even to redistribute it in any way whatsoever.
>go ask a lawyer, again. You must ACTIVELY protect your rights! For example,
>Princeton University, which OWNS a road near by, would loose
>it if they didnt close it for public traffic every so often? The road
>would become public property because ... they let the public use
>it as if it WAS public property. Many lawyers seems to believe the same
>holds for software. True, you don't have to protect ALL your right ALL
>the time, but you must show some "concern" for it.
I haven't heard of the Princeton case before, but I do know of a
similar case at Harvard University. Harvard closes the gates to
Harvard Yard on a complex schedule to avoid continuous public access
to the yard. However, if they did not do this the Yard would not
become public property. What would happen is that the public would
gain a right of way: Harvard would lose the right to deny public
access to the Yard. (This was explained to me as a consequence of old
English common law).
In any case whether you have to protect your rights depends upon the
rights in question. You do not have to protect patent rights. You do
have to protect trademark rights. I actually don't know whether you
have to protect copyright rights. However, I do know that to defend
trademark rights all you have to do is send a cease and desist letter
to anybody using the trademark without proper credit. You do not have
to go about asserting the trademark yourself, once you have registered
it.
--
Ian Taylor | i...@airs.com | First to identify quote wins free e-mail message:
``Intuition told him that the vast ineptitude of the venture would serve as
ample proof that no fraud was afoot, since an impostor would hardly have
overlooked such flagrant discrepancies.''
Paul Crowley
>Whether laws should exist that protect the unborn fetus at 1, 3, 6,
>or 9 months is a tough issue in society -- me, I think fetuses at any
>age deserve at least as much protection as any animal, but that's
>not in line with the current Political Cleansing movement of course.
Can we talk about gun control and Nazi Germany now?
Tim Smith
--Tim Smith
Michael I Bushnell
1. the RPL can not be "illegal". It is TOTALLY LEGAL to be racial in
this country. Try to join the KKK. Hack, their software might have an RPL.
It is legal to be racial (which is a word that means "a member of a
race"). To be racist (which is what I think you mean) is legal, but
not completely. As regards speech, racism is legal. As regards trade
practices or contracts, racism is illegal. It is not legal to
discriminate on the basis of race in the provision of public
accomodations (restaurants, retail stores, housing), nor are most
racist contracts valid. Such contracts are termed "contrary to public
policy" and not upheld by courts. The hypothetical RPL would be
contrary to public policy. Whether it would be overturned is a
different question--because whether it is a contract is disputable.
2. the RPL will be invalid for the same reason the GPL is in valid - no
one signed it or agreed to it.
It is certainly disputable whether the hypothetical RPL or the GPL is
a contract. But that does not mean it doesn't bind. Many things
validly constrain behavior, only one form of which is called a
"contract". In the case of the GPL, if you don't agree to it, then
you are quite right: it doesn't bind you. But, in that case, you are
prohibited by the law itself from copying GNU software, because the
copyright still holds. Their is precedent on the books that copyright
is not lost merely by choosing to use a non-profit oriented means of
distribution.
--
Michael I. Bushnell | Why are you so full of heaviness, O my soul?
+1 617 628 6197 (H) -+- and why are you so disquieted within me?
+1 617 253 8568 (W) | Put your trust in God;
14 Paulina Street | for I will yet give thanks to him,
Somerville, MA 02144 | who is the help of my countenance, and my God.
Michael I Bushnell
1. the RPL can not be "illegal". It is TOTALLY LEGAL to be racial in
this country. Try to join the KKK. Hack, their software might have an RPL.
It is legal to be racial (which is a word that means "a member of a
race"). To be racist (which is what I think you mean) is legal, but
not completely. As regards speech, racism is legal. As regards trade
practices or contracts, racism is illegal. It is not legal to
discriminate on the basis of race in the provision of public
accomodations (restaurants, retail stores, housing), nor are most
racist contracts valid. Such contracts are termed "contrary to public
policy" and not upheld by courts. The hypothetical RPL would be
contrary to public policy. Whether it would be overturned is a
different question--because whether it is a contract is disputable.
2. the RPL will be invalid for the same reason the GPL is in valid - no
one signed it or agreed to it.
It is certainly disputable whether the hypothetical RPL or the GPL is
a contract. But that does not mean it doesn't bind. Many things
validly constrain behavior, only one form of which is called a
"contract". In the case of the GPL, if you don't agree to it, then
you are quite right: it doesn't bind you. But, in that case, you are
prohibited by the law itself from copying GNU software, because the
copyright still holds. There is precedent on the books that copyright
Michael I Bushnell
Ever heard of the KKK? This is a FREE (not like GNU's software :-) COUNTRY.
it is legal to be racial. There are only specific rules concerning
racial equality, usually related to government contracts, housing,
public services, etc. A "private" contract between us is otherwise legal,
even if racial. The real question is if it is VALID, since it relies only
on the copyright law for its binding.
That depends on just what you mean by a private contract. I saw the
following example recently:
Mr Doe wishes to purchase a car from Mr Smith. Mr Smith agrees to
sell the car, but only on condition that Mr Doe promise not to have
any black friends. (Mr Smith does not attach a restrictive covenant
to the use of the car itself; that would invalidate the contract, or
be separated and invalited.) Mr Doe completes the sale. Later, Mr
Smith finds out that Mr Doe has black friends, and sues for the
cancellation of the contract and the reversal of the sale. The
provision that Mr Doe hav no black friends was separated from the
contract and invalidated; the remainder of the contract was retained.
Here, then, a "private contract" was judged contrary to public policy,
and invalidated.
Craig Burley
In article <BURLEY.93...@apple-gunkies.gnu.ai.mit.edu> bur...@apple-gunkies.gnu.ai.mit.edu (Craig Burley) writes:
> A shrink-wrap license is a document that is provided with a copyrighted
> work within the packaging of the work itself, such that the
> recipient (purchaser) must somehow confirm agreement to the
> license (such as by tearing it along with the envelope to access
> the contents, which include the work itself) via a method that
> is generally post-receipt, and that adds restrictions to the
> recipient's use (other than commonly accepted restrictions on
> redistributing copies) of the work.
>The GPL shares the "packaged with the work itself" attribute, but
>not the "must somehow confirm agreement to" attribute, and most
>definitely not the "adds restrictions to use" attribute (the point
>I think the courts probably found most unsupportable).
It doesn't share the "must somehow confirm agreement to" attribute, literally
speaking. But its use of "if you don't follow the contract, the default is
that you can't redistribute" as a substitution for "you're bound by a contract
that says you can't redistribute" has the same _effect_; the user ends up
permitted to do identical things in both cases, even though the means of
acheieving this differ.
The same for the "adds restrictions to use" attribute.
The former, maybe; but the GPL does not add restrictions to use. Shrink-
wrap licenses do. The GPL only adds permissions to redistribution,
no restrictions, above and beyond what is typical for copyrighted works
such as books and non-licensed-but-proprietary software.
Is this really so confusing for people to understand?
Craig Burley
As pointed out before, the court may rule that, since the only use of GNU
software was with a GPL, and the GPL is invalid, the software is in the
public domain now, rather than "everyone is now in violation".
I doubt any court will ever rule the GPL invalid, and, again, if it
did, it would not rule that the sw was PD.
I admit I don't know of a specific case details where copyr. was "taken away"
by publishing. One exmaple given to me was a book on bridge designs. Someone
used a design from the book to build a REAL bridge (e.g. for cars, not for
ethernet packets :-), and the author sued for copyr. violations.
I think the equivalent thing in the GNU world would be a GNU author (such
as the FSF) suing someone for reading GNU code, then writing their own
implementation from scratch, i.e. where independent recreation happened
and thus the resulting code was _not_ derived from the original by
copyright standards. The courts would be right to throw that one out, but
that wouldn't hurt the GPL or GNU copyrights.
The court decided that by publishing the book, the author gave up the rights
for the copyrighted design itself - anyone can buy the book and build
bridges based on the design. (Note - we are not talking "patent" or
"algorithm" here, but an actual design, with full details.) I understand all
other "design books" fall into this category, even though they have not
given up their copyr. explicitly. Imagine if you bought a book with
home furnishing suggestions and was later sued for furnishing your home in
the way shown.
Now, there's probably a lesson here for everyone: did the author of the
book give of the rights for the _book_ itself, because the courts
ruled that he didn't have rights over any implementation of designs
described in the book? Answer: No. The same would go for the GPL and GNU,
I believe.
One may claim that that the GNU software, being "free" has given up its
rights in the same way; the "profits" from the distribution are the
modified code sent back, fame, etc. This is what I understood that
lawyers (I talked to or heard) claim.
GNU software has "given up" its rights to own the algorithms it contains
both by design and by fiat, indeed. But the issue is whether the
code itself, or derivations thereof, can be copied freely. The answer is:
no, in the same way that those books on building bridges cannot now
be freely and legally copied and redistributed by non-copyright-
holders, even if material is added and/or deleted (i.e. derived works).
Nor can digital copies of the scanned texts (a written-word
equivalent of compiling C code into assembler and distributing the
assembler) be legally distributed by anyone.
Last comment: supposingly, many "organizations" tried to make FSF software
into proprietary product, but were scared away by FSF/GPL lawyers. I have
plenty of suggestions on how to circumvent the GPL, which I think anyone
serious would attempt to do.
The serious ones hired the good lawyers. They told them that these
"suggestions" will not work.
For example, create a new language D. modify
gcc to compile it; modify, in a non-reversable way, Emacs source into D.
sell emacs binaries and a (useless) D source (no one said u need to
give away a compiler!). This isn't in violation as far as I can tell.
Read the license -- it is in violation, since you wouldn't be
distributing what the GPL defines as "source code".
Others have already responded to the other suggestions.
Funny how all these discussions about the supposed invalidity of the
GPL or the horribleness/illegality of the FSF being called the
"Free Software Foundation" seem to devolve into people trying to
come up with scenarios (they implicitly or explicitly threaten to
try) legally undermining all the work the FSF and its supporters are
trying to do. And most of those scenarios require more work than
simply writing equivalent PD programs from scratch (at least, more
work for talented programmers -- admittedly untalented ones might
do translations and write translators faster than they might write
PD versions of GNU EMACS, GCC, and GNU utilities).
The only person I can remember honestly disagreeing, in the past, with
the GPL but not resolving the discussion by saying "well, then, supposed
I did <insert-threat-here>" has since become a convert. Maybe there
are others, but I think the way these discussions go show that the
anti-GPL people have agendas other than simply being "concerned" that
the GPL is "unenforcable". And it often turns out that it is those
other agendas, not any legal knowledge or understanding, that drives
these "GPL unenforcable" tirades.
I suggest that all of us (or those few of us left) who bother to
publicly rebut these uninformed claims simply change to saying "why,
you might be right, in fact you probably are, why not release
GNU code as PD or proprietary and earn yourself a lot of money,
you couldn't possibly lose a suit" and then sit back and enjoy the
show as the FSF sues their butts off.
Then again, that's probably not a smart idea, either. FSF resources
are better spent doing other things (judgements would take a while,
and the defendants probably don't have much money, else they wouldn't
bother pestering the FSF).
Craig Burley
Actually, this reminds me of one of the few objections to the GPL I've
seen that I was actually somewhat sympathetic with. Someone who was
developing software under a government grant, who *had* to make the
results of his work PD (since it was developed with public funds).
[clarifications of sympathy omitted]
I feel somewhat the same way. I think the best solution would be for
the government to recognize that the GPL is a reasonable alternative
for accomplishing the same goals as releasing stuff as PD. Basically
any license that allows commercial use of the code and places no
restrictions on the number of copies made by a user should be
acceptable to the government. (This is different from what
universities usually want, I realize -- the gov't actually wants
its efforts to be directly beneficial to businesses, but that goal
wouldn't be given up by distributing some stuff under the GPL.)
Of course if the gov't is just using the PD-only approach to avoid
any and all legal hassle, then I guess no license is a good
alternative. I'm just guessing as to why they distribute only as PD.
Craig Burley
Should Nazi fetuses be allowed to own guns?
Guns don't kill people. Nazi fetuses and bullets do. :-)
Per Abrahamsen
>>>>> On 8 Mar 93 11:57:34, bur...@apple-gunkies.gnu.ai.mit.edu (Craig
>>>>> Burley) said:
In article <1nfkve...@shelley.u.washington.edu>
t...@stein2.u.washington.edu (Tim Smith) writes:
> Should Nazi fetuses be allowed to own guns?
Craig> Guns don't kill people. Nazi fetuses and bullets do. :-)
If Gnus are Outlawed, Only the Outlaws will have Gnus.
(Will the NRA help defend the GPL in court?)
mathew
> In article <1993Mar5.0...@Princeton.EDU> mg@elan (Michael Golan)
> writes:
>
> I and others think the FSF/GPL try to force us into socialized software,
> especially given rms opinions that prgrammers should be paid by the
> government or that some of us make too much money. Which is why I think
> it is invalid, since it is trying to use the copyright *priviledges*,
> just like the RPL, to impose its own agenda, something the copyright law
> never intended. (the GPL comments translate real well to RPL, btw, try
> it sometimes ... we don't object that blacks use or write software ... we
> want to protect you ... we believe ...)
>
> This is the funniest bit of confused frothing-at-the-mouth I have
> seen so far this year. Gets my vote for the "Hall of Shame" for USENET
> '93, pending more posts from this guy.
His first sentence is obviously true; the bits about "paid by the government"
and "make too much money" are straight from the GNU Manifesto.
The second sentence is also obviously true, in that the first part is a
statement of his opinion, and it is certainly the case that the copyright law
was not intended to enable people to impose licenses on those who use their
material in order to achieve political ends.
That just leaves the final parenthesized comment, which is a simple comment
on prose style. So which bit do you think is "confused"?
mathew
mathew
> As I understand it, the courts ruled that SWLs are illegal because you
> have to agree to it before you read it. It is because they are
> different *on the legal point which failed SWLs* that the GPL will
> probably stand up.
Several times I've copied software and only discovered that it's under the
GPL once I've unpacked it and found the license agreement inside. In fact, I
can't think of a single occasion where I've been asked to agree to the GPL
before being allowed to copy some GNU software. So it seems to me that the
GPL is exactly like a SWL on this point -- maybe not in theory, but in
practice, which is what's important.
I mean, if I had to sign agreement to the GPL before being allowed to copy
Emacs, you'd have a point. But I don't. In that respect, it's just like a
SWL. I don't find out about it until after I've done something which I need
to have agreed to it in order to do.
mathew
Craig Burley
> Should Nazi fetuses be allowed to own guns?
Craig> Guns don't kill people. Nazi fetuses and bullets do. :-)
If Gnus are Outlawed, Only the Outlaws will have Gnus.
(Will the NRA help defend the GPL in court?)
New bumper sticker:
"I'm an NRA Member -- support your right to bear a rms!"
Craig Burley
bur...@apple-gunkies.gnu.ai.mit.edu (Craig Burley) writes:
> In article <1993Mar5.0...@Princeton.EDU> mg@elan (Michael Golan)
> writes:
>
> I and others think the FSF/GPL try to force us into socialized software,
> especially given rms opinions that prgrammers should be paid by the
> government or that some of us make too much money. Which is why I think
> it is invalid, since it is trying to use the copyright *priviledges*,
> just like the RPL, to impose its own agenda, something the copyright law
> never intended. (the GPL comments translate real well to RPL, btw, try
> it sometimes ... we don't object that blacks use or write software ... we
> want to protect you ... we believe ...)
>
> This is the funniest bit of confused frothing-at-the-mouth I have
> seen so far this year. Gets my vote for the "Hall of Shame" for USENET
> '93, pending more posts from this guy.
His first sentence is obviously true; the bits about "paid by the government"
and "make too much money" are straight from the GNU Manifesto.
No, the first sentence is false. First, even if it is "given" that
rms thinks the gov't should fund software (and I don't remember
that being the context of the comments in the GNU MAnifesto) and
that programmers make too much money, it is entirely foolish to
conclude that the FSF/GPL "try to force us into socialized
software" and is a wonderful example of frothing at the mouth, just
as is those who claim the FSF/GPL "are going to put lots of
programmers out of work" (i.e. aren't socialistic _enough_).
Craig Burley
I mean, if I had to sign agreement to the GPL before being allowed to copy
Emacs, you'd have a point. But I don't. In that respect, it's just like a
SWL. I don't find out about it until after I've done something which I need
to have agreed to it in order to do.
No, the GPL does not require you to agree to it before making a copy
of GPL-protected code for yourself or for your own use. It only does
so before you distribute the code to others. Read the license again.
John L. Coolidge
>In article <1993Mar5.1...@times.aux.apple.com>,
>cool...@sirius.aux.apple.com (John L. Coolidge) wrote:
>> The reimbursement forms
>> don't specify the program in question and no purchase orders ever
>> existed. If the manuals and disks get lost (as happened more than once
>> in office moves and, probably, theft), how in the world do we prove
>> ownership? Should the SPA (or a court, for that matter) be allowed to
>> conclude that the copies are illegal _simply because_ of the lack of
>> records?
>Of course not. However, finding a place with more copies than can be
>accounted for is like finding a man with a smoking gun -- you can't prove
>he's a murderer, but it's a good place to start looking [...]
Sure, I agree here. It's a very good place to _start_ looking.
However, the SPA seems to think it's a good place to _stop_ looking --
they seem to believe that anyone holding a smoking gun is,
automatically, a murderer. I'd like to see them at least show a corpse
first (i.e. show duplicate serial numbers, copies that are registered
to someone else, etc).
>That's what the audits are for. What bothers me personally about the SPA is
>not any question of illegality, but the lawsuit-bullying-type attitude that
>appears to accompany them. Of course, I haven't been involved with them
>personally.
I don't think their actions are illegal, per se. I do have a lot of
trouble with their bullying attitude; I'd really like to someone stand
up to them and require proof of violations before paying. However, in
our settle-out-of-court-to-avoid-trouble society, that's not likely.
--John
I think it is a little premature to attribute the failures
of American foreign policy to Carl Kadie.
-- Mike Godwin (found in soc.feminism)
+++John L. Coolidge++++++++...@apple.com+++++++++++++++++++++++
I speak for myself, not for Apple Computer. Copyright 1993 John L. Coolidge.
Copying allowed only if attributed, and if all copies may be further copied.
John L. Coolidge
>In article <1993Mar5.1...@Princeton.EDU> mg@elan (Michael Golan) writes:
> For example, create a new language D. modify
> gcc to compile it; modify, in a non-reversable way, Emacs source into D.
> sell emacs binaries and a (useless) D source (no one said u need to
> give away a compiler!). This isn't in violation as far as I can tell.
>Read the license -- it is in violation, since you wouldn't be
>distributing what the GPL defines as "source code".
Not so (or at least not according to _my_ interpretation of the GPL.
Here's the relevant section:
*The source code for a work means the preferred form of the work for
*making modifications to it. For an executable work, complete source
*code means all the source code for all modules it contains, plus any
*associated interface definition files, plus the scripts used to
*control compilation and installation of the executable. However, as a
*special exception, the source code distributed need not include
*anything that is normally distributed (in either source or binary
*form) with the major components (compiler, kernel, and so on) of the
*operating system on which the executable runs, unless that component
*itself accompanies the executable.
This is a pretty complicated paragraph and many terms are not
defined. However, let's start at the top and see where we can go.
Assume first that someone (X) has done what Michael Golan suggests
and translated Emacs into D. They then do all their editting and
maintenance in D; they do not ever go back to the C source, except to
translate Emacs updates from C into D and then integrate them. It
seems to me that D is then the 'preferred form of the work for making
modifications to it' (a phrase which is not defined). Certainly C
cannot be, since X does not do any work in C and has done at least
some work in D.
Given this, X would seem to be allowed by the GPL to distribute their
version of Emacs in D. Assuming they also send along D-format
interfaces, their Makefiles, and the like, I think they're covered.
They _might_ also have to distribute the C->D translator, since they
probably cannot distribute the system include files, and the include
files they use are not the system include files per se (they've been
translated first). However, if their zipped-up gcc can parse C include
files as well as D, they're covered on that point too.
The GPL mentions compilers at this point, but it _does not_ (IMHO)
require distributing a compiler which will compile the program. The
only mention of compilers seems to be to help define 'major
components', which is then used merely to indicate what included
source code need not be distributed.
If this section of the GPL does require the distribution of a
compiler, I believe that I know of at least one company that would be
in violation here (they sell GNU binaries and will happily give out
source, but you must pay money for a binary of a compiler and
libraries which will produce the binaries in question, or so I
understand their license to indicate. I could be wrong).
--John
It wasn't lies, Jenny, it was acting.
-- Nevil Sinclair, _The_Rocketeer_
Craig Burley
This is where I'd interpret the GPL differently, but I can see that
it is difficult to be sure how the courts would interpret it.
To me, the "preferred form for modifications" is fairly inclusive,
applying not just to the person distributing the code, but the
people receiving it as well. Else, why not encrypt the source code,
use editors that automatically do the encryption/decryption, and
use an internal, top-secret (or artifically very high-priced)
compiler that does the decryption on the fly? The encrypted form
would clearly be "preferred" by the author(s), but useless to the
receipients.
(A good indication of whether this'd work is whether it would be
valid for a firm to release under the GPL documentation whose source
can be manipulated only by their proprietary WP or page-layout program,
and/or in conjunction with a data-dictionary kind of thing that
is kept as a trade secret by the firm. Again, it'd be that firms'
"preferred form", but nobody else could use it, they'd be getting
the equivalent of binaries.)
I'd like to think that if such a thing happened, the courts would
see that the letter of the GPL, not being perfectly clear, defers
to the spirit in that recipients of the code should be able to
read, edit, and recompile it using "preferred" tools. A "D"
source file clearly cannot be a "preferred form" for users
who have no way of obtaining a D compiler.
Otherwise, why can't someone claim that they edit all their
stuff at the executable, object, or assembler level, and that
that is their "source", aka "preferred form", and anything they
use to produce it is just an editor (even if _we'd_ call it
a compiler)? It would be somewhat difficult, from a
practical perspective, to prove that someone intent enough on
fooling the court was actually using gcc as a compiler rather
than an editor of sorts (i.e. by using -S and hand-editing the
output to tune it, whatever). Though the court battle would
be fun to watch (having the plaintiff, e.g. the FSF, force the
defendant to prove his/her ability to make substantial changes
to the code in his/her "preferred form", and contrast that to
the speed and reliability of making such changes in C or
whatever the FSF claims is the "preferred form").
So rather than try and use this phrase to cleverly opt out of the
GPL (and certainly such a method would be argued as a subterfuge
in court), I suggest people not attempt to redefine "preferred
form of the work for making modifications to it" as "_my_
preferred form of the work (which nobody else can possibly
understand)".
If this section of the GPL does require the distribution of a
compiler, I believe that I know of at least one company that would be
in violation here (they sell GNU binaries and will happily give out
source, but you must pay money for a binary of a compiler and
libraries which will produce the binaries in question, or so I
understand their license to indicate. I could be wrong).
It'd be interesting to know more about this case. If they claim
to distribute source code in, say, C, but working binaries
cannot be produced without purchasing their particular C compiler
(and that compiler isn't normally provided with such a system),
I suspect they're in violation of the GPL. Again, it's simply too
easy to defeat (in the form of subterfuge) by providing a
very-expensive proprietary "C compiler" that detects one particular
kind of special form and does something very "secret" with it,
required to make the compiled program work, or run as fast as
the distributed binary, etc. Personally, I don't think perfect
binary production (i.e. the user being able to create the exact
same .EXE) as being a requirement, necessarily, as long as the
complete functioning of the binary can be recreated from scratch
using the source and an available compiler.
How the courts would rule on all this, I don't know. Surely it
is clear that such a means for getting around the GPL could be
very effective at preventing the author of copyrighted code
from maintaining (and thus retaining useful ownership) of a
derivation of his/her code, so from that standpoint alone, it
is likely that a thoughtful court would force the distributor
to release their compiler (or whatever) for a nominal fee. (They'd
still retain copyright, naturally.)
But this is all, IMHO, much fuzzier than whether an author
retains copyright of a work he/she makes available for
retrieval via anonymous ftp, advertises as "free", etc.
And it goes back to the practical issue: why would someone capable
of creating a language, D, sufficiently distinct from C that
mechanical translation cannot be easily reverse-engineered and
sufficiently robust that a reliable converter of C->D can be
written, and capable of implementing a compiler, debugger, and
run-time library for D, ever waste his/her time doing all that
just to steal GNU code and risk civil (or perhaps criminal?)
penalties?
Such a person, if really competent enough to do all this, would be
just like the guy who carefully painted forged $20 bills or
something, was arrested, convicted, died in jail, meanwhile his
other works (that he figured weren't worth much) sold for thousands
of dollars. (And the short-term rewards for writing good software
are vast compared to those for making good paintings -- one mustn't
die for one's software to increase substantially in value. In fact,
its value tends to decrease upon death, I suspect. :-) In other
words, this painter could have made more money by selling his
paintings than by painting counterfeit money! A similar competent
programmer would do far better than that, of course.
Such a person should instead spend his/her time writing _better_
stuff than yet another UNIX clone (GNU) and designed better than
a mechanical C->D translation would provide (given how much GNU code
is, well, not necessarily designed in the best fashion). E.g., ever
see Fortran code written in C or PL/1 or Pascal? That's what all
that GNU code would look like in D -- or D is just too similar to C
for the technique to work (it is too easy for hackers like me to figure
it out and make a free compiler for it, thus defeating the whole
purpose of the exercise).
My sneaking suspicion is that anyone really thinking about this
approach realizes they are not sufficiently competent to re-engineer
and re-design a GNU clone (not protected by the GPL) from scratch
in the timeframe they'd like, but they somehow think that they're
competent enough to do the comparatively much more difficult work
(though admittedly more focused) of creating D and making it usable
as a target for a C translator in that same timeframe. And I'm
sure that they'll prove themselves wrong, either by never really
getting D or C->D to work adequately, or doing such a trivial D that
the law (if not the FSF supporters) will have no problem catching
up with them (the law by judging against them, the FSF by making
a free D compiler, and/or C->D converter, available).
In case it isn't already clear, I do recognize that _you_,
Mr. Coolidge, were not in the least suggesting any of this.
It has been proposed by others, who exhibit far less ability
to read and understand the GPL than you, and who make enough
typos in _their_ posts that I'd be surprised if they could
succeed at _spelling_ "D" correctly often enough to even begin
their proposed project! :-)
Rahul Dhesi
<mat...@mantis.co.uk> writes:
>p...@dcs.ed.ac.uk (Paul Crowley) writes:
>> As I understand it, the courts ruled that SWLs are illegal because you
>> have to agree to it before you read it. It is because they are
>> different *on the legal point which failed SWLs* that the GPL will
>> probably stand up.
>Several times I've copied software and only discovered that it's under the
>GPL once I've unpacked it and found the license agreement inside. In fact, I
>can't think of a single occasion where I've been asked to agree to the GPL
>before being allowed to copy some GNU software. So it seems to me that the
>GPL is exactly like a SWL on this point -- maybe not in theory, but in
>practice, which is what's important.
No, no, you all have it wrong. Here is the critical difference between
the GPL and the typical shrink-wrapped license, based on US law.
- When you buy software from a store, you pay $x and get back software
product y. You have *entered into a contract* and for the price you
paid, you now *own* a copy of product y, which you may use as allowed
by US copyright law. Any license statement inside a shrink-wrap (or
even one readable before you buy the package) is totally, utterly,
irrelevant (unless it grants you rights that you don't already have
under copyright law, and this is a rare case that I have never
encountered).
- When you get GNU software by anonymous ftp, *there is no contract*
and you have no legal right to use it. You are granted rights by the
GPL that you did not have, but these are not legal rights, because
you cannot enter into a binding contract without consideration. So
although you are permitted to use the software, you have no
enforceable legal right to do so.
- When you order a tape from the FSF, the situation is unclear.
If your payment is clearly only a handling free, then any contract
that exists is solely for the purpose of the tape being shipped to
you, and it does not govern the contents of the tape. If the payment
is actually for the software itself, then there is a contract that
affects how you use it.
--
Rahul Dhesi <dh...@cirrus.com>
also: dh...@rahul.net
Alan Braggins
> - When you get GNU software by anonymous ftp, *there is no contract*
> and you have no legal right to use it. You are granted rights by the
> GPL that you did not have, but these are not legal rights, because
> you cannot enter into a binding contract without consideration. So
> although you are permitted to use the software, you have no
> enforceable legal right to do so.
The GPL grants/controls rights to distribute, not rights to use.
--
Alan Braggins +44-223-316673 Shape Data, A Division of EDS-Scicon Ltd,
al...@sdl.ug.eds.com Regent St, Cambridge, CB2 1DB, U.K.