copyleft licence compatible with apache licence

[Pierre Gambarotto]

Hello everyone

I'm searching for a copyleft licence (to prevent abusive commercial 
reuse) compatible

with the apache licence.

From what I've understood, the GPL is not compatible with the Apache 
licence, because
of the clause of the Apache licence that force to advertise the Apache 
foundation.


Some questions I have :
_ is it possible to use the GPL with a modification, like : this code 
is distributed in GPL, but

can be used with any apache licenced code.
_ is there an other licence that would suit this kind of need ?

Sorry if this kind of discussion has already been adressed on this 
list, I had some problems

accessing the archives.

Any pointer would be much appreciated

Pierre

Re: Bug#221709: ITP: at76c503a-source -- at76c503a driver source

[Joachim Breitner]

Hi,

Am Di, den 25.11.2003 schrieb Anthony DeRobertis um 21:27:
> On Nov 25, 2003, at 13:54, Joachim Breitner wrote:
> No, it has nothing to do with whether a promise is kept, but it does 
> have something to do with if a legal promise ( = contract ) is made. A 
> contract (at least in many places) needs an offering ("consideration") 
> from both sides; and if you paid nothing, I doubt you've made one.
>
> That's not always the case --- for example, if its adware, then being 
> subjected to ads is it. Or at least I'd argue.
> 
> But if I just hand you a floppy disk with two files on it, program.exe 
> and COPYING, and ask nothing in return, what contract could of been 
> formed?
Ok, your point seems valid here, especially in private business and in
isolated products (yet another text editor). But in this case, we are
talking about something I need to make the hardware run. And though I
didn't pay for the firmware, I payed for the hardware (here is our
contract) and thus need the firmware. And along with the firmware they
hand out to me because I bought their hardware, they give me the promise
(the GPL).

Assume you alter your example: I sell you my old computer. I say, if you
want, you can get an operating system for that computer for free. You
take it. I give you the disk "as part of the system" (though not
necessarily the same time) and then all promises made along this (it
works, you can get a second disk if your dogs eats this one, you can get
the source) are binding. 

> > And does it matter whether they advertised the GPL? I also think this
> > does not change any facts. As said before (not checked by me), the GPL
> > is noted in the firmware file - could be enough of an "advertisement"
> > for someone really looking for Free firmware.
> Maybe, but I suspect even if that were the case, all they could do is 
> demand their money back (for the card purchase).
Yes, I guess this is always the company's alternative.

nomeata

(Sorry I cound't stop :-))
-- 
Joachim "nomeata" Breitner
  e-Mail: [EMAIL PROTECTED] | Homepage: http://www.joachim-breitner.de
  JID: [EMAIL PROTECTED] | GPG-Keyid: 4743206C | ICQ#: 74513189
  Geekcode: GCS/IT/S d-- s++:- a--- C++ UL+++ P+++ !E W+++ N-- !W O? M?>+ V?
PS++ PE PGP++ t? 5? X- R+ tv- b++ DI+ D+ G e+>* h! z?
Bitte senden Sie mir keine Word- oder PowerPoint-Anhänge.
Siehe http://www.fsf.org/philosophy/no-word-attachments.de.html


signature.asc
Description: Dies ist ein digital signierter Nachrichtenteil

Re: Bug#221709: ITP: at76c503a-source -- at76c503a driver source

[Walter Landry]

Alexander Cherepanov <[EMAIL PROTECTED]> wrote:
> 19-Nov-03 13:25 Don Armstrong wrote:
> > On Wed, 19 Nov 2003, Oliver Kurth wrote:
> >> Sigh. So if Atmel says these files are no longer GPL'ed, but are just
> >> freely distributable, it could at least go to non-free?
> 
> > Yes.
> 
> >> Sounds ridiculous. (Law is too complicated to me, so I stick to
> >> programming ;-) )
> 
> > Thats part and parcel of the GPL... if the company doesn't include the
> > prefered form for modification, no one else can distribute it.
> 
> Sorry for the intrusion, but is there a consensus on this issue?
> I.e. why binaries can not be distributed under section 2 of the GPL?
> 
> In the thread starting from
> http://lists.debian.org/debian-legal/2002/debian-legal-200212/msg00161.html
> opinions seem to be divided:
> 
> In
> http://lists.debian.org/debian-legal/2002/debian-legal-200212/msg00202.html
> Walter Landry wrote:
> 
>   Richard Braakman <[EMAIL PROTECTED]> wrote:
>   > It's clear that our basic disagreement is here.  I see nothing in
>   > section 2 that would limit it only to source code.
> 
>   Correct.  Compiling is a form of modification.  But are you able to
>   distribute everything in the object file (including the libraries)
>   under the terms of the GPL?  If not (which is the case most of the
>   time for compiled languages on non-free platforms), then the GPL
>   allows a special exemption: Section 3.

I take it back.  Section 2 says

  You may modify your copy or copies of the Program or any portion of
  it, thus forming a work based on the Program, and copy and
  distribute such modifications or work under the terms of Section 1

and Section 1 requires source code.

Regards,
Walter Landry
[EMAIL PROTECTED]

Re: Preparation of Debian GNU/Linux 3.0r2 (II)

[Osamu Aoki]

Hi,

On Fri, Nov 21, 2003 at 02:09:15PM +, Andrew Suffield wrote:
> On Fri, Nov 21, 2003 at 09:04:32PM +0900, GOTO Masanori wrote:
> > At Fri, 21 Nov 2003 08:35:10 +,
> > Andrew Suffield wrote:
> > > [1  ]
> > > On Fri, Nov 21, 2003 at 09:52:01AM +0900, GOTO Masanori wrote:
> > > > At Thu, 20 Nov 2003 22:36:40 +0100,
> > > > Osamu Aoki wrote:
> > > > > > > One of "More-clearly-free alternative scalable Japanese fonts" is
> > > > > > > kochi-mincho/kochi-gothic in sid/sarge. Many Japanese use this
> > > > > > > font rather than Watanabe font.
> > > > > > 
> > > > > > If this alternative contains the necessary glyphs, then I do not see
> > > > > > that much of a problem with removing the Hitachi fonts.
> > > > > 
> > > > > Exactly.  We just has to make sure HITACHI's claim was not the primary
> > > > > reason to do so.  HITACHI is just a noise.
> > > > 
> > > > So you just ignore original font author's claim.  Is it good attitude?

What claim?  The fact that "they are so called 'stolen' from the word
processor"?  I know their claim of "font history".  I pretty much
believe it as is although they did not do a good job convincing me.  I
am not arguing on that.  The evaluation of the history is only required
when the "ownership" issues is resolved to Hitachi's favor.

My problem is their claim of exclusive "ownership" which streches to
vectorized fonts.  Their claim of "ownership" requires due action on
their side.  If we bend over to their claim of this extensive
"ownership" under such an loose claim, we set ourselves a bad
precidence.  That is the real issue.  This is not just Japanese issue.

Attitude?  Let's not deviate from real discussion.

> > > If their claim was bogus? Yup, it is. Paying attention to bogus claims
> > > isn't just silly, it sets a very bad precendent.
   Yes.

> > Yeah, if we recognize it's just bogus, then we don't discuss seriously
> > and don't consume our precious time.
> > 
> > Original author (Hitachi, who were infringed), and kochi upstream
> > author (who infringed without knowing) already discussed and their
> > conclusion was that it was not just bogus.

The "inflingiment" is very doubtful considering Hitachi does not have
any right to start with.

> Erm, when asking the question of whether or not they are right, their
> own statement that they are right is not useful.

True.  The same goes with the fact of their "ownership".  That needs to
be evaluated.   Or kill these fonts out of archive with tottaly
different reason.  Then we loose reason to argue.

Osamu

Re: Bug#221709: ITP: at76c503a-source -- at76c503a driver source

[Anthony DeRobertis]

On Nov 25, 2003, at 18:27, Joachim Breitner wrote:




Ok, your point seems valid here, especially in private business and in
isolated products (yet another text editor). But in this case, we are
talking about something I need to make the hardware run. And though I
didn't pay for the firmware, I payed for the hardware (here is our
contract) and thus need the firmware. And along with the firmware they
hand out to me because I bought their hardware, they give me the 
promise

(the GPL).


You probably bought a box marked as having Windows 98/ME/2000/XP 
support and maybe Mac support as well. I very much doubt it mentioned 
having GPL'd Linux drivers. If you weren't aware of it, its hard to see 
it being part of a contract, at least in any sane jurisdiction.


Now, you may of been aware of it from their web site, and made a 
purchase decision based on it. In that case, you're probably right: It 
would be considered part of the contract, and a court would award 
reasonable damages. I'd suspect that a court would award you your money 
back (plus possibly expenses). The furthest I can see a court going is 
ordering them not to refer to the drivers as GPL'd on their website any 
longer.


Assume you alter your example: I sell you my old computer. I say, if 
you

want, you can get an operating system for that computer for free. You
take it. I give you the disk "as part of the system" (though not
necessarily the same time) and then all promises made along this (it
works, you can get a second disk if your dogs eats this one, you can 
get

the source) are binding.


If you agreed to do that as part of the original purchase, then 
certainly. If not, it gets more murky.

Re: Binaries under GPL(2)

[Anthony DeRobertis]

On Nov 26, 2003, at 21:59, Alexander Cherepanov wrote:

Sorry, it's not yet clear to me why Section 2 is not applicable to
binaries.


Section 2 says you may do so "under the terms of Section 1 above." 
Section 1 grants rights to "copy and distribute verbatim copies of the 
Program's source code"

Re: Bug#221709: ITP: at76c503a-source -- at76c503a driver source

[Henning Makholm]

Scripsit Anthony DeRobertis <[EMAIL PROTECTED]>

> No, it has nothing to do with whether a promise is kept, but it does
> have something to do with if a legal promise ( = contract ) is made. A
> contract (at least in many places) needs an offering ("consideration")
> from both sides;

That is true in common-law jurisdictions. In most continental European
legal systems, unilateral promises are legally valid and can be
enforced by the beneficiary. (Though, in the case of disagreement
about exactly what was promised, the beneficiary will often have
to bear the entire burden of proof).

-- 
Henning Makholm "Fuck Lone."

Re: Binaries under GPL(2) (was: Re: Bug#221709: ITP: at76c503a-source -- at76c503a driver source)

[Henning Makholm]

Scripsit Don Armstrong <[EMAIL PROTECTED]>
> On Wed, 26 Nov 2003, Alexander Cherepanov wrote:
> > 24-Nov-03 22:02 Don Armstrong wrote:

> >> in order to redistribute under the terms of the GPL, you need to be
> >> able to provide source (the prefered form for modification.)

> > Section 2 of the GPL doesn't require to provide source. It doesn't
> > talk about source at all.

> Section 3 is the critical point here, as it covers distribution in
> object or executable formats, which is what we would be distributing.

I think Alexander's point may have merit. If you distribute whatever
precise bits it was that the copyright holder waved a copy of the GPL
over, those bits must be assumed to be "the Program", and as such GPL
#2 gives you right to distribute a modified version of the bits.

The fact that GPL #3 *also* gives you the right to distribute the bits
under somewhat different conditions should not affect the right given
in #2.

However,

 1. This is clearly a wording oversight in the GPL. I'm not sure that
Debian should base its decisions to distrtibute things on such
loopholes.

 2. Some kind of modification to the bits is required for #2 to be
applicable. Then one needs to add prominent notices of the
modification to the file. An ELF binary could conveivably
be relinked to contain such a notice in an unused section,
but it's harder do see how to do this for firmware object code
in a closed format.

 3. Opaque bits GPL-licensed in this way are not GPL-compatible (!) -
if they are combined with object code produced from GPLed *source*
the resulting work is undistributable.

 4. (And, of course, opaque bits cannot be DFSG-free, GPL or no GPL,
due to DFSG #2).

-- 
Henning Makholm"I ... I have to return some videos."

Re: Preparation of Debian GNU/Linux 3.0r2 (II)

[Anthony DeRobertis]

On Nov 24, 2003, at 11:15, GOTO Masanori wrote:


So it's hard to make Japanese characters which have beautiful shape
and unified baseline because each form is complex, and there are a lot
of such complicated characters.


Well, at the risk of starting a flame war, that says more about how 
screwed up the language is than about the copyrightability (in the US) 
of a work. The amount of work doesn't really matter; the amount of 
creativity does. See, e.g., Feist v. Rural Telephone.



This is because we don't decide
easily that Japanese fonts are no creativity in even 32 dot bitmap
resolution.  They are just art.


There is certainly some creativity in any font, otherwise all fonts 
would look the same (or would be random). I'm sure neither is the case.



So it's hard to make Japanese characters which have beautiful shape
and unified baseline because each form is complex, and there are a lot
of such complicated characters.  This is because we don't decide
easily that Japanese fonts are no creativity in even 32 dot bitmap
resolution.  They are just art.


They are, as a matter of law in the United State, not copyrightable. 
For example, this paragraph comes from the House Report on the 1976 
Copyright Act:


A "typeface" can be defined as a set of letters, numbers or
other symbolic characters, whose forms are related by repeating
design elements consistently applied in a notational system and
are intended to be embodied in articles whose intrinsic
utilitarian function is for use in composing text or other
cognizable combinations or characters. The Committee does not
regard the design or typeface, as thus defined, to be a copyright
able "pictorial, graphic, or sculptural work" within the meaning
or this bill and the application of the dividing line in
section 101. [H.R. Rep. No. 1478, 94th Cong., 2d Sess. 55 (1976)].

In fact, before the 1976 act, typefaces were copyrightable in the 
United States. Congress chose to change that.


See also 37 CFR Sec. 202.1, which reads (in part):
The following are examples of works not subject to copyright
and applications for registration of such works cannot be
entertained:

(e) Typeface as typeface.

There are other copyright office rulings saying that bitmap fonts can't 
be copyrighted.




BTW: For the benefit of list readers who can't see that glyph, here it 
is in PNG format w/ 96.0pt Osaka. Or, at least this is what my Mac 
displays it as.



No, I have no idea what it means

Re: Binaries under GPL(2) (was: Re: Bug#221709: ITP: at76c503a-source -- at76c503a driver source)

[Henning Makholm]

Scripsit Don Armstrong <[EMAIL PROTECTED]>
> On Wed, 26 Nov 2003, Henning Makholm wrote:

> > I think Alexander's point may have merit. If you distribute whatever
> > precise bits it was that the copyright holder waved a copy of the GPL
> > over, those bits must be assumed to be "the Program", and as such GPL
> > #2 gives you right to distribute a modified version of the bits.

> At least, the way I read the GPL, 2 gives you the right to distribute
> a modified version of the Program Source (which is what 1 covers).

Are we reading different versions of the license? The one I'm looking
at (/usr/share/emacs/20.7/etc/COPYING from woody's emacs20 package)
does not mention the word "source" in its section 2 at all.

#1 explicitly applies only to "the Program's source code", but #2
speaks generally about "the Program", source code or not.

-- 
Henning Makholm "Al lykken er i ét ord: Overvægtig!"

Re: Preparation of Debian GNU/Linux 3.0r2 (II)

[Anthony DeRobertis]

On Nov 25, 2003, at 18:04, Don Armstrong wrote:


I'm not sure that Fiest v Rural Telephone can lead us down this road.


It doesn't, not at all.


Assuming the font is a work of authorship (which many large or
relatively large bitmap and TT fonts are) claiming a copyright on it
is an entirely reasonable thing to do.


No it isn't. In what is probably one of them few times Congress said 
"no", it did so in the 1976 Copyright Act.

Re: Binaries under GPL(2) (was: Re: Bug#221709: ITP: at76c503a-source -- at76c503a driver source)

[Don Armstrong]

On Wed, 26 Nov 2003, Henning Makholm wrote:
> #1 explicitly applies only to "the Program's source code", but #2
> speaks generally about "the Program", source code or not.

Sure, but the only type of distribution allowed under #2 is
distribution under #1 (You may modify your copy or copies of the
Program [...] and copy and distribute such modifications or work under
the terms of Section 1 above) which is distribution of "the Program's
source code," with the additional permision of being able to
distribute modified versions of that source code. 

[We come right back to the question of whether or not the binary is
source. I claim it isn't in this case, and since what what would
normally be source under the GPL is extant, I'd imagine a court would
agree.] 


Don Armstrong

-- 
Miracles had become relative common-places since the advent of
entheogens; it now took very unusual circumstances to attract public
attention to sightings of supernatural entities. The latest miracle
had raised the ante on the supernatural: the Virgin Mary had
manifested herself to two children, a dog, and a Public Telepresence
Point.

-- Bruce Sterling, _Holy Fire_ p228

http://www.donarmstrong.com
http://www.anylevel.com
http://rzlab.ucr.edu


signature.asc
Description: Digital signature

Re: copyleft licence compatible with apache licence

[Wouter Verhelst]

On Thu, Nov 27, 2003 at 07:42:23AM +0100, Pierre Gambarotto wrote:
> Some questions I have :
> _ is it possible to use the GPL with a modification, like : this code 
> is distributed in GPL, but
> can be used with any apache licenced code.

Yes, you can do that. You should add a paragraph to your copyright
notice that says something like

"As a special exception, the author grants you the right to link the
Program with any software licensed under "

Obviously, you can only do this legally if you are, in fact, the author
of that license.

-- 
Wouter Verhelst
Debian GNU/Linux -- http://www.debian.org
Nederlandstalige Linux-documentatie -- http://nl.linux.org
"Stop breathing down my neck." "My breathing is merely a simulation."
"So is my neck, stop it anyway!"
  -- Voyager's EMH versus the Prometheus' EMH, stardate 51462.


signature.asc
Description: Digital signature

Re: Bug#221709: ITP: at76c503a-source -- at76c503a driver source

[Patrick Herzig]

On Wed, 2003-11-26 at 00:27, Joachim Breitner wrote:
> Hi,
(...)
> > But if I just hand you a floppy disk with two files on it, program.exe 
> > and COPYING, and ask nothing in return, what contract could of been 
> > formed?
> Ok, your point seems valid here, especially in private business and in
> isolated products (yet another text editor). But in this case, we are
> talking about something I need to make the hardware run. And though I
> didn't pay for the firmware, I payed for the hardware (here is our
> contract) and thus need the firmware. And along with the firmware they
> hand out to me because I bought their hardware, they give me the promise
> (the GPL).
(...)

1. The GPL is not a promise to receive the source code. The GPL only
talks about such a promise.

2. The GPL does not cover what rights you have against the one giving
something to you. It is concerned only with the rights you have in
using/modifying/distributing to others.

If someone gives you a software binary under the GPL but refuses to hand
over the source code it makes his act of distribution a violation of
copyright law and the copyright holder can go after him. The fact that
he is in violation of copyright law does not give you any entitlements
unless you are the copyright holder.

Re: Binaries under GPL(2)

[Walter Landry]

Alexander Cherepanov <[EMAIL PROTECTED]> wrote:
> 24-Nov-03 22:02 Don Armstrong wrote:
> > On Tue, 25 Nov 2003, Alexander Cherepanov wrote:
> >> Sorry for the intrusion, but is there a consensus on this issue? I.e.
> >> why binaries can not be distributed under section 2 of the GPL?
> 
> > When binaries are not the prefered form for modification, as in the
> > case where there is still source code extant[1], in order to
> > redistribute under the terms of the GPL, you need to be able to
> > provide source (the prefered form for modification.)
> 
> Why? Section 2 of the GPL doesn't require to provide source. It
> doesn't talk about source at all.

Section 2 references Section 1, which does talk about source.

> > 1: In my opinion anyway, it is not enough that source is not available
> > to the secondary distributor.
> 
> What is source according to the GPL is an interesting question, but
> that's another question. And answer is only needed when you want to
> distribute something under section 3 of the GPL. So it's not directly
> relevant in case of Debian. Let me quote Walter Landry again
> (http://lists.debian.org/debian-legal/2002/debian-legal-200212/msg00182.html):
> 
>   Section 3 gives you rights in addition to section 2.  Section 3 lets
>   you distribute a particular kind of modification that is not allowed
>   in Section 2 (a modification that incorporates things that can not be
>   licensed under the GPL).  But Debian is not doing that, so there is no
>   need to resort to section 3.

I have since changed my mind.  I wasn't reading Section 2 closely
enough.  Only Section 3 gives you permission to distribute non-source.
Section 1 lets you distribute source, and Section 2 lets you
distribute modified source.

Regards,
Walter Landry
[EMAIL PROTECTED]

Re: copyleft licence compatible with apache licence

[Wouter Verhelst]

On Thu, Nov 27, 2003 at 01:00:17PM +0100, Wouter Verhelst wrote:
> Obviously, you can only do this legally if you are, in fact, the author
> of that license.

Uh. the author of the Program, not the license :)

-- 
Wouter Verhelst
Debian GNU/Linux -- http://www.debian.org
Nederlandstalige Linux-documentatie -- http://nl.linux.org
"Stop breathing down my neck." "My breathing is merely a simulation."
"So is my neck, stop it anyway!"
  -- Voyager's EMH versus the Prometheus' EMH, stardate 51462.


signature.asc
Description: Digital signature

Re: Bug#221709: ITP: at76c503a-source -- at76c503a driver source

[Arnoud Engelfriet]

Joachim Breitner wrote:
> Now compare Atmel: They give me the binary with a note (called GPL),
> that I can get the source code from then, the next 2 years at the
> expense of the copying (or something like that). If they don't do that,
> they are misleading the customer.

The GPL nowhere says that the original author of the code will
give the source to anyone. It only says that recipients who
distribute binaries, must also distribute source (or offer to
distribute source). 

If someone writes software under GPL and distributes binaries
only, he makes it impossible for everyone else to redistribute
those binaries. I do not think that that in itself is sufficient
to _demand_ source from that person. There has to be something
else, like a statement that this software can be freely modified
or may be copied by anyone.

Arnoud

-- 
Arnoud Engelfriet, Dutch patent attorney - Speaking only for myself
Patents, copyright and IPR explained for techies: http://www.iusmentis.com/

Re: Binaries under GPL(2)

[Henning Makholm]

Scripsit Don Armstrong <[EMAIL PROTECTED]>

> If you are distributing an executable or object code, that
> distribution is subject to Section 3. No other section of the GPL
> gives you rights to distribute executable or object code.

Remember that we're talking about the non-standard situation where the
object code *itself* is "the Program" to which the copyright holder
applied the GPL. Then I as a recipient has a choice of distributing
it under section 2 or section 3, because in that case either section
offera me a permission to distribute.

-- 
Henning Makholm  "Guldnålen er hvis man har en *bror* som er *datalog*."

Re: Bug#221709: ITP: at76c503a-source -- at76c503a driver source

[Henning Makholm]

Scripsit Joachim Breitner <[EMAIL PROTECTED]>

> Now compare Atmel: They give me the binary with a note (called GPL),
> that I can get the source code from then,

No, that's not what the note says. The note says that you can
distribute the binary to your neighbour if you also give your
neighbour the source. Where you get the source is immaterial to
this offer.

The GPL is all about what *you* have to do if *you* distribute. It
does not in any way enable you to demand things from *others* who
distribute, unless you happen to hold a copyright on the thing they
distribute.

> What if I chose their product because their firmware was GPL, and the
> competitor's product wasn't? Then the competitor could sue them for what
> is called "unlauterer Wettbewerb" in Germany (unfair competition
> according to "translate").

Perhaps, but would be because their marketing created wrong
expectations for you. The lawsuit will only fly because those
expectations *are*, in fact, wrong.

-- 
Henning Makholm   "Hi! I'm an Ellen Jamesian. Do
you know what an Ellen Jamesian is?"

Re: simplest copyleft license for a wiki

[Henning Makholm]

Scripsit Edmund GRIMLEY EVANS <[EMAIL PROTECTED]>
> Alex Schroeder <[EMAIL PROTECTED]>:

> >1. You have the right to copy, modify, and/or distribute the work.

> I don't know what "and/or" means, but I find it hard to imagine a
> definition of "and/or" which would make this sentence mean that I have
> clear and explicit permission to distribute modified copies.

Um, what have you been smoking? That is as clear and explicit as it is
possible to make it. "And/or" is a pretty standard way of phrasing an
inclusive or: You have the right to do one. You have the right to do
the other.  You have to the right to do both. If you do both, then,
naturally, what you distribute is the result of the modifications.

Cf. the pine/UW attack which interpreted "right to modify and
distribute" as: You have the right to modify. You have the right to
distribute. You *don't* have the right to do both at once. This is
exactly what the "and/or" wording seeks to defuse.

> >2. You must grant recipients the same rights.

> Same as what?

Possibly as in clause 1. There's definitely room for improved wording
here.

> I'm not sure that the term "copyleft" is sufficiently well known to be
> used like this.

I agree.

-- 
Henning Makholm  "Underlige Ugle vågner midt om natten."

Re: Binaries under GPL(2) (was: Re: Bug#221709: ITP: at76c503a-source -- at76c503a driver source)

[Henning Makholm]

Scripsit Don Armstrong <[EMAIL PROTECTED]>
> On Wed, 26 Nov 2003, Henning Makholm wrote:

> > #1 explicitly applies only to "the Program's source code", but #2
> > speaks generally about "the Program", source code or not.
> 
> Sure, but the only type of distribution allowed under #2 is
> distribution under #1 (You may modify your copy or copies of the
> Program [...] and copy and distribute such modifications or work under
> the terms of Section 1 above) which is distribution of "the Program's
> source code,"

The reference to #1 obviously targets only the part after "provided
that". There is a completely similar reference from #3, and #3
unambigously does not speak about source.

> [We come right back to the question of whether or not the binary is
> source.

No. The binary in question is clearly not source.

-- 
Henning Makholm"What a hideous colour khaki is."

Re: Bug#221709: ITP: at76c503a-source -- at76c503a driver source

[Joachim Breitner]

Hi,

I think this is a reasonable conclusion that I fully agree.

If one is mean, he could try is to mail Atmel, saying them that for a
pretty large setup of WLAN-Cards, his company says that the complete
(i.e. with the firmware) drivers have to be under the GPL, and if they
answer "Yes, get ours, they are under the GPL", you buy the cards (or
just one) and a week later request the source for the firmware. But this
would piss them and other vedords off, so let's not do it :-)

nomeata

Am Do, den 27.11.2003 schrieb Anthony DeRobertis um 10:37:
> On Nov 25, 2003, at 18:27, Joachim Breitner wrote:
> 
> >>
> > Ok, your point seems valid here, especially in private business and in
> > isolated products (yet another text editor). But in this case, we are
> > talking about something I need to make the hardware run. And though I
> > didn't pay for the firmware, I payed for the hardware (here is our
> > contract) and thus need the firmware. And along with the firmware they
> > hand out to me because I bought their hardware, they give me the 
> > promise
> > (the GPL).
> 
> You probably bought a box marked as having Windows 98/ME/2000/XP 
> support and maybe Mac support as well. I very much doubt it mentioned 
> having GPL'd Linux drivers. If you weren't aware of it, its hard to see 
> it being part of a contract, at least in any sane jurisdiction.
> 
> Now, you may of been aware of it from their web site, and made a 
> purchase decision based on it. In that case, you're probably right: It 
> would be considered part of the contract, and a court would award 
> reasonable damages. I'd suspect that a court would award you your money 
> back (plus possibly expenses). The furthest I can see a court going is 
> ordering them not to refer to the drivers as GPL'd on their website any 
> longer.
> 
> > Assume you alter your example: I sell you my old computer. I say, if 
> > you
> > want, you can get an operating system for that computer for free. You
> > take it. I give you the disk "as part of the system" (though not
> > necessarily the same time) and then all promises made along this (it
> > works, you can get a second disk if your dogs eats this one, you can 
> > get
> > the source) are binding.
> 
> If you agreed to do that as part of the original purchase, then 
> certainly. If not, it gets more murky.
-- 
Joachim "nomeata" Breitner
  e-Mail: [EMAIL PROTECTED] | Homepage: http://www.joachim-breitner.de
  JID: [EMAIL PROTECTED] | GPG-Keyid: 4743206C | ICQ#: 74513189
  Geekcode: GCS/IT/S d-- s++:- a--- C++ UL+++ P+++ !E W+++ N-- !W O? M?>+ V?
PS++ PE PGP++ t? 5? X- R+ tv- b++ DI+ D+ G e+>* h! z?
Bitte senden Sie mir keine Word- oder PowerPoint-Anhänge.
Siehe http://www.fsf.org/philosophy/no-word-attachments.de.html


signature.asc
Description: Dies ist ein digital signierter Nachrichtenteil

Re: simplest copyleft license for a wiki

[Edmund GRIMLEY EVANS]

Henning Makholm <[EMAIL PROTECTED]>:

> > >1. You have the right to copy, modify, and/or distribute the work.
> 
> > I don't know what "and/or" means, but I find it hard to imagine a
> > definition of "and/or" which would make this sentence mean that I have
> > clear and explicit permission to distribute modified copies.
> 
> Um, what have you been smoking? That is as clear and explicit as it is
> possible to make it.

I find section 2 of the GPL to be much more clear and explicit.

> "And/or" is a pretty standard way of phrasing an
> inclusive or: You have the right to do one. You have the right to do
> the other.  You have to the right to do both.

The semantics of "or" in English has almost nothing to do with the
"inclusive or" and "exclusive or" of logic. It has more to do with
quantification and has many of the same ambiguities. In general, when
you write "or" you should specify at whose option. For example, in a
contract where ten euros are to be exchanged for "a burger, chips and
cola or lemonade" you'd expect the choice of beverage to be at the
purchaser's option, but grammatically, without further context, it
could mean at the vendor's option.

> If you do both, then,
> naturally, what you distribute is the result of the modifications.

It is possible for me to do both by modifing a program and also
distributing the unmodified program. The best way of giving permission
to distribute modified copies is to say so. "Preparation of
deriviative works" is perhaps even better.

The expression "and/or" is a monstrosity that should be avoided in all
contexts, especially when one is trying to be precise, IMHO.

> Cf. the pine/UW attack which interpreted "right to modify and
> distribute" as: You have the right to modify. You have the right to
> distribute. You *don't* have the right to do both at once.

So, you're saying that "and" on its own doesn't allow both at once. It
must be an XAND :-)

> This is
> exactly what the "and/or" wording seeks to defuse.

But fails absolutely to do so ...

Re: Preparation of Debian GNU/Linux 3.0r2 (II)

[Osamu Aoki]

On Thu, Nov 27, 2003 at 05:24:47AM -0500, Anthony DeRobertis wrote:
> On Nov 24, 2003, at 11:15, GOTO Masanori wrote:
> 
> >So it's hard to make Japanese characters which have beautiful shape
> >and unified baseline because each form is complex, and there are a lot
> >of such complicated characters.
> 
> Well, at the risk of starting a flame war, that says more about how 
> screwed up the language is 

What? I wonder your intent for this OT statement :-( 

When you discuss things with non-native audience, safer choice of
language reduces frictions.  This statement caused me to doubts the
intent of the speaker from my modern day multi-cultural values.  Sorry.

Anyway, the idiosyncrasies are, I think, essential part the of culture.
Do not you love totally inconsistent spelling rule of English while most
other languages in Europe have rationalized their spelling rules in
modern days during the 17th-20th century.  So I should think this
statement of "screwed up" as a compliment to the Japanese culture,
maybe :-)

> than about the copyrightability (in the US) of a work. 

You seem to know and very intelligent on this subject of copyright.
...
> They are, as a matter of law in the United State, not copyrightable. 
> For example, this paragraph comes from the House Report on the 1976 
> Copyright Act:
...
> There are other copyright office rulings saying that bitmap fonts can't 
> be copyrighted.

But US is not the only country in the world.

Anyway, the updated request for removal/update for these packages, I
hope, shall not contain reference to the "copyright" issues.  Then the
result of these legal assessments will be irrelevant for the request.

Osamu

Re: Bug#221709: ITP: at76c503a-source -- at76c503a driver source

[Joachim Breitner]

Hi,

Am Do, den 27.11.2003 schrieb Henning Makholm um 09:55:
> Scripsit Joachim Breitner <[EMAIL PROTECTED]>
> The GPL is all about what *you* have to do if *you* distribute. It
> does not in any way enable you to demand things from *others* who
> distribute, unless you happen to hold a copyright on the thing they
> distribute.

Now I'm starting to get it. Guess you just made me feel bad about
posting a lot of messages based upon a wrong assumption. Sorry everybody
:-)

But this new view leads me to other interesting effects: (assuming we
are talking about a close source driver, coming in binary form, but
under the GPL, distributed by the copyright holder)

 * The driver is under the GPL, and since I want to use it (not
distribute it), I do not violate the GPL.
 * Therefore when linking with the kernel, the module should have full
access to all functions and not taint the kernel.
 * This way, proprietary drivers can use the full kernel functionality,
when downloaded directly from the copyright holder.

Sure, the GPL does prohibit linking with GPL-incompatible code. But the
driver is _under_ the GPL - there is just nobody that I can demand the
source code from.

I guess this is not right, but I wonder at what point I missed
something.

nomeata


-- 
Joachim "nomeata" Breitner
  e-Mail: [EMAIL PROTECTED] | Homepage: http://www.joachim-breitner.de
  JID: [EMAIL PROTECTED] | GPG-Keyid: 4743206C | ICQ#: 74513189
  Geekcode: GCS/IT/S d-- s++:- a--- C++ UL+++ P+++ !E W+++ N-- !W O? M?>+ V?
PS++ PE PGP++ t? 5? X- R+ tv- b++ DI+ D+ G e+>* h! z?
Bitte senden Sie mir keine Word- oder PowerPoint-Anhänge.
Siehe http://www.fsf.org/philosophy/no-word-attachments.de.html


signature.asc
Description: Dies ist ein digital signierter Nachrichtenteil

Re: simplest copyleft license for a wiki

[Henning Makholm]

Scripsit Edmund GRIMLEY EVANS <[EMAIL PROTECTED]>

> > Cf. the pine/UW attack which interpreted "right to modify and
> > distribute" as: You have the right to modify. You have the right to
> > distribute. You *don't* have the right to do both at once.

> So, you're saying that "and" on its own doesn't allow both at once. It
> must be an XAND :-)

No, I'm saying that "and" in this position is ambguous. "And/or" is
not.

> > This is exactly what the "and/or" wording seeks to defuse.

> But fails absolutely to do so ...

Rubbish.

-- 
Henning Makholm "... and that Greek, Thucydides"

Re: Bug#221709: ITP: at76c503a-source -- at76c503a driver source

[Henning Makholm]

Scripsit Joachim Breitner <[EMAIL PROTECTED]>

> But this new view leads me to other interesting effects: (assuming we
> are talking about a close source driver, coming in binary form, but
> under the GPL, distributed by the copyright holder)

>  * The driver is under the GPL, and since I want to use it (not
> distribute it), I do not violate the GPL.
>  * Therefore when linking with the kernel, the module should have full
> access to all functions and not taint the kernel.
>  * This way, proprietary drivers can use the full kernel functionality,
> when downloaded directly from the copyright holder.

There's not complete agreement about this. Some would claim that just
by being compiled using headers from the kernel sources, the binary
driver is legally derived from the kernel sources, and that the vendor
must therefore distribute with source.

However, it seems that the upstream Linux kernel maintainers' position
is that loadable modules in .o form are not considered derivates of
the kernel source, and so do not need to be GPLed. The kernel sources
even contains infrastructure for this (i.e., modules can declare which
license they are under, and 2.4 kernels explicitly make some but not
all kernel symbols available to non-GPL modules).



If one links such modules statically into a kernel image of one's own,
the natural consequence of this would be that the image cannot be
*distributed*, but it sould be perfectly legal to *use* it.

> Sure, the GPL does prohibit linking with GPL-incompatible code.

Not really. It just effectively prohibits distribution of the
resulting binary. As long as one just uses it for oneself, there
should be no problem.

-- 
Henning Makholm  "Den nyttige hjemmedatamat er og forbliver en myte.
Generelt kan der ikke peges på databehandlingsopgaver af
  en sådan størrelsesorden og af en karaktér, som berettiger
  forestillingerne om den nye hjemme- og husholdningsteknologi."