Knuth statement on renaming cm files and Licence violation.

[David Carlisle]

Having been away I just have just (rather depressingly) caught up with
the August archive of latex licence discussion.

There seems to be the usual quota of misinformation concerning the tex
licence, but one particular point was raised.

As has been mentioned the "rename clause" of computer modern licence was
invoked to cause linux distributions to stop distributing modified cm
fonts (ie tfms different from Knuth's, not just as Thomas Bushnell has
claimed, the .mf source files).

The text of Knuth's statement appears to have been removed from all the
obvious places where it was posted at the time, but I turned up a copy
at this address:

http://groups.google.com/groups?q=knuth+romlig.mf&hl=en&lr=&ie=UTF-8&selm=5nd91y6eix.fsf%40sun.dante.de&rnum=6

It is prefixed (In German, which I can't read, sorry) by a statement
from the president of German TeX users group, but it contains the full
text (in English) of Knuth's statement asserting that he believed that
the distribution of the modified tfms was

"in direct violation of my stipulation on the copyright page of Computers
& Typesetting, Volume E."


Further on he goes on to say

Dear friends, I decided to put these fonts into the public domain rather
than to make them proprietary; all I have asked is that nobody change them,
UNLESS THE NAME IS CHANGED, so that every user can obtain equivalent results
on all computer systems, now and 50 years from now. I went to enormous
efforts to make TeX and METAFONT systems equivalent on hundreds of different
computers, and to make them archival as no commercial software has ever
been. If you want to improve the fonts, go ahead, but DON'T GIVE YOUR FONTS
THE SAME NAME AS MINE. I insist that every font named cmr10 have the same
font metrics, so that TeX will choose exactly the same line breaks and page
breaks on every computer system in the world. This compatibility must be
enforced by peer pressure (boycotts, bad publicity , etc.), to anybody who
breaks the rules. The TeX Users Group is now deciding how best to condemn
this action and to keep the cancer from infecting too many systems.



David

_
This message has been checked for all known viruses by Star Internet
delivered through the MessageLabs Virus Scanning Service. For further
information visit http://www.star.net.uk/stats.asp or alternatively call
Star Internet for details on the Virus Scanning Service.

Re: Knuth statement on renaming cm files and Licence violation.

[Thomas Bushnell, BSG]

David Carlisle <[EMAIL PROTECTED]> writes:

> The text of Knuth's statement appears to have been removed from all the
> obvious places where it was posted at the time, but I turned up a copy
> at this address:

And note that it begins with "I decided to put these fonts into the
public domain; all I have asked is that ..."

Note that if this means *anything* at all, the request is a mere
request and not legally binding.

Re: Knuth statement on renaming cm files and Licence violation.

[David Carlisle]

> And note that it begins with "I decided to put these fonts into the
> public domain; all I have asked is that ..."

As has been stated many times, the conditions on Knuth's programs and
fonts are scattered over many places, the copyright pages in books, and
comments in source code and readme files in the distributions.
The wording (especially the use of "public domain") is often confusing
and arguably contradictory.

However the _intent_ of the TeX conditions is clear (and stated in all
caps in the text I quoted); Change whatever you like, so long as you
change your name (including names of relevant files).

This is the basis of the claim we have often stated (and you've often
denied) that the design of the LPPL was primarily intended to
clarify a situation in which latex could be distributed in the same
manner as TeX, but with clearer conditions, all stated in one place.
I think the LPPL succeeded in that. It is precisely because the
conditions on latex are so much easier to find than the conditions on
TeX that we (as opposed to Knuth) get the initial comments about the
renaming clause.

> Note that if this means *anything* at all, the request is a mere
> request and not legally binding.

It is not phrased as a request it is stated that this is the intended
interpretation of the licence by the copyright holder. (Yes I know he
also mentions PD in the same statement...)


"in direct violation of my stipulation on the copyright page of Computers
& Typesetting, Volume E."

We do have a clear statement that Debian wouldn't try to use legal
technicalities to weasel out of honouring such a statement, don't we?

David

_
This message has been checked for all known viruses by Star Internet
delivered through the MessageLabs Virus Scanning Service. For further
information visit http://www.star.net.uk/stats.asp or alternatively call
Star Internet for details on the Virus Scanning Service.

Re: Knuth statement on renaming cm files and Licence violation.

[Brian Sniffen]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1


> On Wed, 4 Sep 2002 09:36:31 +0100, David Carlisle <[EMAIL PROTECTED]> 
> said:

>> And note that it begins with "I decided to put these fonts into the
>> public domain; all I have asked is that ..."

> As has been stated many times, the conditions on Knuth's programs and
> fonts are scattered over many places, the copyright pages in books, and
> comments in source code and readme files in the distributions.
> The wording (especially the use of "public domain") is often confusing
> and arguably contradictory.

> However the _intent_ of the TeX conditions is clear (and stated in all
> caps in the text I quoted); Change whatever you like, so long as you
> change your name (including names of relevant files).

> This is the basis of the claim we have often stated (and you've often
> denied) that the design of the LPPL was primarily intended to
> clarify a situation in which latex could be distributed in the same
> manner as TeX, but with clearer conditions, all stated in one place.
> I think the LPPL succeeded in that. It is precisely because the
> conditions on latex are so much easier to find than the conditions on
> TeX that we (as opposed to Knuth) get the initial comments about the
> renaming clause.

>> Note that if this means *anything* at all, the request is a mere
>> request and not legally binding.

> It is not phrased as a request it is stated that this is the intended
> interpretation of the licence by the copyright holder. (Yes I know he
> also mentions PD in the same statement...)

The reference to enforcement by shunning and community effort seems to
indicate otherwise.  I've been trying to read that statement from
every angle I can think of, but I just can't find a consistent meaning
other than that Knuth has put this in the public domain, but makes strong
requests of the community of users of the code he wrote.

> "in direct violation of my stipulation on the copyright page of Computers
> & Typesetting, Volume E."

Sadly, I don't own a copy of Computers & Typesetting.  Can you quote
the full copyright page, and give a general indication of the contents
of Volume E?

> We do have a clear statement that Debian wouldn't try to use legal
> technicalities to weasel out of honouring such a statement, don't we?

Certainly.  If anything is to be weaseled, I suspect it will be the
DFSG: it's enough to have legal permission to distribute modified code
(say, because it's public domain) that we don't think we'll likely
ever exercise (out of politeness and respect).

- -Brian

- -- 
Brian Sniffen   [EMAIL PROTECTED]
http://www.evenmere.org/~bts/

-BEGIN PGP SIGNATURE-
Version: GnuPG v1.0.7 (GNU/Linux)

iD8DBQE9dh9/03mlJHngJfERApkWAJ9adMXakOuYcnOeFE9xFDgrPlaUdACfT0X7
FZNshUGGbAsfWxsBc0PR1jc=
=SWQ7
-END PGP SIGNATURE-

Re: Knuth statement on renaming cm files and Licence violation.

[David Carlisle]

> The reference to enforcement by shunning and community effort seems to
> indicate otherwise.  I've been trying to read that statement from
> every angle I can think of, but I just can't find a consistent meaning
> other than that Knuth has put this in the public domain, but makes strong
> requests of the community of users of the code he wrote.

I don't think so. He states clearly that this is in violation of the
copyright notice, the bit about community pressure is an indication that
he thinks that is an effective way (as opposed to the courts) of
enforcing the rules. (It seemed to work in that case:-)

Sadly, I don't own a copy of Computers & Typesetting vol E either.

Frank's quoted parts of it here:
http://lists.debian.org/debian-legal/2002/debian-legal-200208/msg9.html
No doubt he could quote the full text of the notice if it's thought
important.

David


_
This message has been checked for all known viruses by Star Internet
delivered through the MessageLabs Virus Scanning Service. For further
information visit http://www.star.net.uk/stats.asp or alternatively call
Star Internet for details on the Virus Scanning Service.

Re: Knuth statement on renaming cm files and Licence violation.

[Martin Schröder]

On 2002-09-04 16:13:24 +0100, David Carlisle wrote:
> Sadly, I don't own a copy of Computers & Typesetting vol E either.

The millenium edition has this:

"The programs for Computer Modern are in the public domain,
and readers may freely generate and hand-tune their own fonts
using the algorithms of this book. However, use of the names
is restricted: Any font whose names cmr10 or cmbx12 or ...
are identical to the standard font names of this book should
be fully compatible with the fonts defined here; i.e., fonts
with the same names are supposed to have precisely the same
character coding schemes and precisely the same font metric
files."

Best regards
Martin
-- 
   Martin Schröder, [EMAIL PROTECTED]
  ArtCom GmbH, Grazer Straße 8, D-28359 Bremen
  Voice +49 421 20419-44 / Fax +49 421 20419-10

Re: Knuth statement on renaming cm files and Licence violation.

[Brian Sniffen]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1


> On Wed, 4 Sep 2002 16:13:24 +0100, David Carlisle <[EMAIL PROTECTED]> 
> said:

>> The reference to enforcement by shunning and community effort seems to
>> indicate otherwise.  I've been trying to read that statement from
>> every angle I can think of, but I just can't find a consistent meaning
>> other than that Knuth has put this in the public domain, but makes strong
>> requests of the community of users of the code he wrote.

> I don't think so. He states clearly that this is in violation of the
> copyright notice, the bit about community pressure is an indication that
> he thinks that is an effective way (as opposed to the courts) of
> enforcing the rules. (It seemed to work in that case:-)

I don't see that as clearly as you do.  He says it's in violation of
the stipulation on the copyright page of C&T:E, which isn't quite the
same thing.  TeX, METAFONT, and the CM fonts certainly were under his
copyright at some point in the past, and there are copyright notices
- From that era.  Perhaps this should be taken to mean that even though
they're now in the public domain, the wishes expressed in those
copyright notices should still be followed?

> Sadly, I don't own a copy of Computers & Typesetting vol E either.

> Frank's quoted parts of it here:
> http://lists.debian.org/debian-legal/2002/debian-legal-200208/msg9.html
> No doubt he could quote the full text of the notice if it's thought
> important.

I was trying to avoid that message in particular:  the last message
has a grave misunderstanding of the DFSG, and the quotation from C&T:E
is only indirect.  Also very serious is the interpretation of
"software systems that agree exactly with the program presented here."
Frank seems to believe that this refers to the text of the source
code... and I disagree.  I think it's much more likely that this
refers to an agreement at the system level, that the output is
correct for all input (which of course prohibits the introduction of
new features, but not more efficient algorithms which produce the same
output, or translated comments).

- -Brian

- -- 
Brian Sniffen   [EMAIL PROTECTED]
http://www.evenmere.org/~bts/


-BEGIN PGP SIGNATURE-
Version: GnuPG v1.0.7 (GNU/Linux)

iD8DBQE9disW03mlJHngJfERAmuPAKCA0k8JetATksjuGPGtTyQsXJ1VwgCfc1LP
UqT1f2qXeD9lzs3vKOE0tU8=
=iWG3
-END PGP SIGNATURE-

Re: Knuth statement on renaming cm files and Licence violation.

[David Carlisle]

> Perhaps this should be taken to mean that even though
> they're now in the public domain, the wishes expressed in those
> copyright notices should still be followed?

I don't think the status of the released  cm fonts has ever changed has
it? From the beginning the statements about copyright and PD have been
there (for good or bad:-)

David



_
This message has been checked for all known viruses by Star Internet
delivered through the MessageLabs Virus Scanning Service. For further
information visit http://www.star.net.uk/stats.asp or alternatively call
Star Internet for details on the Virus Scanning Service.

Re: Knuth statement on renaming cm files and Licence violation.

[David Carlisle]

> TeX, METAFONT, and the CM fonts certainly were under his
> copyright at some point in the past, and there are copyright notices
> - From that era.  

I see Martin has posted the text from vol E (ME).

Note that the "millenium edition" is fairly new (as its name suggests)
a combined set of all of the volumes, from 2000 or thereabouts.

David

_
This message has been checked for all known viruses by Star Internet
delivered through the MessageLabs Virus Scanning Service. For further
information visit http://www.star.net.uk/stats.asp or alternatively call
Star Internet for details on the Virus Scanning Service.

Re: Knuth statement on renaming cm files and Licence violation.

[Martin Schröder]

On 2002-09-04 17:22:06 +0100, David Carlisle wrote:
> Note that the "millenium edition" is fairly new (as its name suggests)
> a combined set of all of the volumes, from 2000 or thereabouts.

And it is by all intentions the _last_ edition. :-)

Best regards
Martin
-- 
   Martin Schröder, [EMAIL PROTECTED]
  ArtCom GmbH, Grazer Straße 8, D-28359 Bremen
  Voice +49 421 20419-44 / Fax +49 421 20419-10

Re: Knuth statement on renaming cm files and Licence violation.

[Thomas Bushnell, BSG]

David Carlisle <[EMAIL PROTECTED]> writes:

> However the _intent_ of the TeX conditions is clear (and stated in all
> caps in the text I quoted); Change whatever you like, so long as you
> change your name (including names of relevant files).

The name of, exactly, what?  That's the problem.

Knuth maddeningly doesn't bother distinguishing between legal
requirements and earnest encouragements.  The FSF thinks you should
only run GNU Emacs on a free operating system, but it certainly
doesn't make this a legal requirement.  And that's crucial.

> This is the basis of the claim we have often stated (and you've often
> denied) that the design of the LPPL was primarily intended to
> clarify a situation in which latex could be distributed in the same
> manner as TeX, but with clearer conditions, all stated in one place.
> I think the LPPL succeeded in that. It is precisely because the
> conditions on latex are so much easier to find than the conditions on
> TeX that we (as opposed to Knuth) get the initial comments about the
> renaming clause.

TeX has *no* "renaming clause" at all, it has a totally different set
of requirements.

> It is not phrased as a request it is stated that this is the intended
> interpretation of the licence by the copyright holder. (Yes I know he
> also mentions PD in the same statement...)

Then he must be giving his earnest encouragements, and not a license,
since the work is public domain, right?

> "in direct violation of my stipulation on the copyright page of Computers
> & Typesetting, Volume E."
> 
> We do have a clear statement that Debian wouldn't try to use legal
> technicalities to weasel out of honouring such a statement, don't we?

Debian is surely quite happy to agree with Knuth's urgent
encouragements and not just the license.

This is not about what *Debian* will do, but about what freedoms we
want to make sure our users have.

Re: Knuth statement on renaming cm files and Licence violation.

[Thomas Bushnell, BSG]

Martin Schröder <[EMAIL PROTECTED]> writes:

> On 2002-09-04 16:13:24 +0100, David Carlisle wrote:
> > Sadly, I don't own a copy of Computers & Typesetting vol E either.
> 
> The millenium edition has this:
> 
> "The programs for Computer Modern are in the public domain,
> and readers may freely generate and hand-tune their own fonts
> using the algorithms of this book. However, use of the names
> is restricted: Any font whose names cmr10 or cmbx12 or ...
> are identical to the standard font names of this book should
> be fully compatible with the fonts defined here; i.e., fonts
> with the same names are supposed to have precisely the same
> character coding schemes and precisely the same font metric
> files."

Which, note carefully, is not an "if you change this, you must rename
it" requirement.  Rather, it requires renaming for *certain* kinds of
changes, rather similarly to the tex.web rules, and rather differently
from the LAPL.

Re: Knuth statement on renaming cm files and Licence violation.

[Russ Allbery]

Thomas Bushnell, BSG <[EMAIL PROTECTED]> writes:
> David Carlisle <[EMAIL PROTECTED]> writes:

>> The text of Knuth's statement appears to have been removed from all the
>> obvious places where it was posted at the time, but I turned up a copy
>> at this address:

> And note that it begins with "I decided to put these fonts into the
> public domain; all I have asked is that ..."

> Note that if this means *anything* at all, the request is a mere
> request and not legally binding.

Surely you're not arguing that makes any difference.  Surely you're not
claiming that Debian should ignore a very reasonable request by Donald
Knuth because you think you've found a legal loophole in his statement.
Surely you're just saying this to make an obscure pedantic point, not to
indicate that it should have anything whatsoever to do with what Debian
does.

-- 
Russ Allbery ([EMAIL PROTECTED]) 

Re: Knuth statement on renaming cm files and Licence violation.

[Thomas Bushnell, BSG]

Russ Allbery <[EMAIL PROTECTED]> writes:

> Thomas Bushnell, BSG <[EMAIL PROTECTED]> writes:
> > David Carlisle <[EMAIL PROTECTED]> writes:
> 
> >> The text of Knuth's statement appears to have been removed from all the
> >> obvious places where it was posted at the time, but I turned up a copy
> >> at this address:
> 
> > And note that it begins with "I decided to put these fonts into the
> > public domain; all I have asked is that ..."
> 
> > Note that if this means *anything* at all, the request is a mere
> > request and not legally binding.
> 
> Surely you're not arguing that makes any difference.  Surely you're not
> claiming that Debian should ignore a very reasonable request by Donald
> Knuth because you think you've found a legal loophole in his statement.
> Surely you're just saying this to make an obscure pedantic point, not to
> indicate that it should have anything whatsoever to do with what Debian
> does.

I fear that you have lost track of the context of the (rather complex)
discussion.

Debian basically just distributes tetex with no significant
modifications.  The issue is not at all about what Debian will do.

The issue is that Debian needs to be sure that our users have certain
freedoms, even if we think it would be foolish to exercise them, and
even if we have no intention of exercising them ourselves.

Re: Knuth statement on renaming cm files and Licence violation.

[Russ Allbery]

Thomas Bushnell, BSG <[EMAIL PROTECTED]> writes:
> Russ Allbery <[EMAIL PROTECTED]> writes:
>> Thomas Bushnell, BSG <[EMAIL PROTECTED]> writes:

>>> And note that it begins with "I decided to put these fonts into the
>>> public domain; all I have asked is that ..."

>>> Note that if this means *anything* at all, the request is a mere
>>> request and not legally binding.

>> Surely you're not arguing that makes any difference.  Surely you're not
>> claiming that Debian should ignore a very reasonable request by Donald
>> Knuth because you think you've found a legal loophole in his statement.
>> Surely you're just saying this to make an obscure pedantic point, not
>> to indicate that it should have anything whatsoever to do with what
>> Debian does.

> I fear that you have lost track of the context of the (rather complex)
> discussion.

Perhaps in part, but your line of reasoning above makes me extremely
uncomfortable.  When the author of a package has clearly stated their
expectations and requirements for redistribution, it seems like that
should be treated as the license, even if they didn't cross the
appropriate legal "t".  To do anything else seems... well, extremely
rude.  Even if Debian is not violating the intended license directly, to
base a stance on the viewpoint that the license is legally uninforcable
and therefore irrelevant seems rather disconcerting.

(I'm probably representing this position as stronger than you're actually
intending to argue.)

In other words, that doesn't strike me as the right way to resolve a
licensing problem.  It seems much better to me to view Knuth's statements
as the license and then try to figure out if that really does violate the
DFSG, and if so, whether that means the license really is incompatible
with Debian's underlying principals or if the DFSG didn't anticipate this
case.

> The issue is that Debian needs to be sure that our users have certain
> freedoms, even if we think it would be foolish to exercise them, and
> even if we have no intention of exercising them ourselves.

I must admit that I fail to see where Knuth's license poses any sort of
issue for the DFSG.  The Artistic License has a very similar provision,
and point 4 of the DFSG specifically states:

The license may require derived works to carry a different name or
version number from the original software.

-- 
Russ Allbery ([EMAIL PROTECTED]) 

Re: Knuth statement on renaming cm files and Licence violation.

[Thomas Bushnell, BSG]

Russ Allbery <[EMAIL PROTECTED]> writes:

> Perhaps in part, but your line of reasoning above makes me extremely
> uncomfortable.  When the author of a package has clearly stated their
> expectations and requirements for redistribution, it seems like that
> should be treated as the license, even if they didn't cross the
> appropriate legal "t".  To do anything else seems... well, extremely
> rude.  Even if Debian is not violating the intended license directly, to
> base a stance on the viewpoint that the license is legally uninforcable
> and therefore irrelevant seems rather disconcerting.

Are you worried that we would be violating it indirectly?

> I must admit that I fail to see where Knuth's license poses any sort of
> issue for the DFSG.  The Artistic License has a very similar provision,
> and point 4 of the DFSG specifically states:
> 
> The license may require derived works to carry a different name or
> version number from the original software.

Knuth's doesn't depending on what it means.  The LAPL (as previously
posted here) certainly does cause a problem.

Note that DFSG-4 does not allow requiring he changing of *file names*,
but rather only the name of the *work*, which is a different
matter--and essentially so--since the name of the work is not a
functional element.  

The LAPL advocates were previously saying that the whole goal of the
renaming requirement was specifically to restrict functional elements,
and that's ipso facto a violation of freeness.

The LAPL people were consistently misstating exactly what Knuth's
actual licenses said, and being incredibly simplistic about them; it
is in that background that I stress a careful legalistic reading of
Knuth's words, as a counterweight to the carefree "he means whatever
we say he means" attitude of some.

You seem to be suddenly entering this as if you were totally unaware
of the preceding controversy.  I would suggest you go to the mailing
list archives and review the discussion.

Re: Knuth statement on renaming cm files and Licence violation.

[Russ Allbery]

Thomas Bushnell, BSG <[EMAIL PROTECTED]> writes:
> Russ Allbery <[EMAIL PROTECTED]> writes:

>> Even if Debian is not violating the intended license directly, to base
>> a stance on the viewpoint that the license is legally uninforcable and
>> therefore irrelevant seems rather disconcerting.

> Are you worried that we would be violating it indirectly?

I'm worried about the basis for decision-making that would be used, in a
broader sense, and what potential slippery slopes it may lead into.  I
understand that legal exactness has to be used as a basis for legal
decision-making.  It just seems to me like when it comes to a question of
policy, using exact legal readings as a way of working around a
problematic licensing clause whose intent doesn't coincide with the exact
legal reading is a bad approach.

That's most of what I was trying to say.  The questions of whether the
license are acceptable are a separate issue to me; the main point that I
feel strongly about is to evaluate the license on its own terms, rather
than on the basis of whether it's worded so as to give people a legal out.

> Note that DFSG-4 does not allow requiring he changing of *file names*,
> but rather only the name of the *work*, which is a different matter--and
> essentially so--since the name of the work is not a functional element.

Yes.  This is the fundamental question, namely whether the DFSG only
allows requirements of changing names as a way to handle what are in
essence marketing issues (making sure the user knows they have a modified
package), or whether it's permissable to protect APIs this way.

(In essence, that's what we're talking about with TeX fonts and with LaTeX
packages.  The name of the package and the name of the font are an element
in the TeX API, and if you change them, you have to change the document to
reflect that change, exactly as if you changed an API in a library.)

Knuth clearly is attempting to protect an API.  The whole reason why he
feels so strongly about this is because he wants to be able to take a TeX
document and process it with a TeX system 50 years from now and get the
exact same result down to the word wrapping, and in order to do that, he
has to be sure that when one uses the standard API for the TeX system, the
results will be identical as when one used that standard API twenty years
earlier on a completely different system.

My personal belief, for whatever it's worth, is that this is so borderline
as to almost require the nature of the language be taken into account; if
changing the calls to fit a new API requires a lot of work, that's
different than changing a line in the prelude of a document.  I can both
understand Knuth's concern and understand the concern from the direction
of supporting modification and redistribution, and it's hard to find the
right balance, particularly when we're talking about reproducibility of
written work rather than software.  Written works don't change in the same
ways and for the same reasons as software packages.

But my basic point here is not to so much get into all of that as it is to
say that I think that fundamental question needs to be addressed according
to the intentions of the original author, regardless of the precise
wording of the license.

> You seem to be suddenly entering this as if you were totally unaware of
> the preceding controversy.  I would suggest you go to the mailing list
> archives and review the discussion.

I'm aware of the basic issues and have followed this exact argument a few
times through from different angles; what you're probably picking up on is
that I didn't follow the immediately previous Debian discussion before
the recent threads.

-- 
Russ Allbery ([EMAIL PROTECTED]) 

Re: Knuth statement on renaming cm files and Licence violation.

[Branden Robinson]

On Wed, Sep 04, 2002 at 05:24:30PM -0700, Russ Allbery wrote:
> Yes.  This is the fundamental question, namely whether the DFSG only
> allows requirements of changing names as a way to handle what are in
> essence marketing issues (making sure the user knows they have a modified
> package), or whether it's permissable to protect APIs this way.

The former.

> Knuth clearly is attempting to protect an API.

It's not clear to me, but that doesn't really matter.  If the entire TeX
community is going to rise up and call Debian a bunch of degenerates for
saying that something that's been placed in the public domain is in the
public domain, then maybe Debian doesn't need TeX or its community.
(That said, perhaps you don't represent the entire TeX community -- just
a didactic, moralizing fragment.)

The DFSG is about preserving freedom.  What people do with their
freedoms can be good or bad.

The DFSG is founded on the premise that APIs don't need "protection"
from their users.  If you hold copyright in software and feel that its
APIs do require this sort of protection, you're welcome to compete with
Adobe, Apple, and Microsoft in the proprietary software arena.

But Debian has no use for your software in that event.

-- 
G. Branden Robinson| If you have the slightest bit of
Debian GNU/Linux   | intellectual integrity you cannot
[EMAIL PROTECTED] | support the government.
http://people.debian.org/~branden/ | -- anonymous


pgplJ6uiU6xhQ.pgp
Description: PGP signature

Re: Knuth statement on renaming cm files and Licence violation.

[Steve Langasek]

On Wed, Sep 04, 2002 at 05:24:30PM -0700, Russ Allbery wrote:
> Thomas Bushnell, BSG <[EMAIL PROTECTED]> writes:
> > Russ Allbery <[EMAIL PROTECTED]> writes:

> >> Even if Debian is not violating the intended license directly, to base
> >> a stance on the viewpoint that the license is legally uninforcable and
> >> therefore irrelevant seems rather disconcerting.

> > Are you worried that we would be violating it indirectly?

> That's most of what I was trying to say.  The questions of whether the
> license are acceptable are a separate issue to me; the main point that I
> feel strongly about is to evaluate the license on its own terms, rather
> than on the basis of whether it's worded so as to give people a legal out.

What does it mean to evaluate a legally-binding document "on its own
terms", if not to evaluate in terms of the legal requirements?

You say that it would be rude for someone to ignore Knuth's request.  But
the aim of the DFSG isn't to protect our users from being perceived as
rude; it's to protect them from civil suits and criminal prosecution. 
Granted, there are other, loftier goals that the DFSG aspire to, but I
don't think anyone could argue with a straight face that the framers of
the DFSG held "never be rude to upstream developers" as one of these
goals.

In cases where Debian's legal interpretation of a license differed from
that of the copyright holder, we have in the past removed packages from
the archive because of the possibility of a lawsuit.  However, if Knuth
really meant "public domain", then in this case there is no copyright
holder at all.

Steve Langasek
postmodern programmer


pgpGw0kwUsFQo.pgp
Description: PGP signature

Re: Knuth statement on renaming cm files and Licence violation.

[Russ Allbery]

Branden Robinson <[EMAIL PROTECTED]> writes:

> It's not clear to me, but that doesn't really matter.  If the entire TeX
> community is going to rise up and call Debian a bunch of degenerates for
> saying that something that's been placed in the public domain is in the
> public domain, then maybe Debian doesn't need TeX or its community.
> (That said, perhaps you don't represent the entire TeX community -- just
> a didactic, moralizing fragment.)

I don't represent any part of the TeX community at all; I don't even use
it any more.  Please don't make assumptions.

I don't find your argument particularly persuasive; it seems to be very
strong on emotion without a lot of logic to back it up, or without any
real discussion of what you're trying to defend and why.

-- 
Russ Allbery ([EMAIL PROTECTED]) 

Re: Knuth statement on renaming cm files and Licence violation.

[Brian Sniffen]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1



Bear in mind, Russ, nobody is questioning whether TeX (or LaTeX) are
*good* software, or *useful* software, or even *open source* software.
The question is whether they are free software.  A restricted API,
which you call a protected API, is not a free API.

Even if LaTeX is found by the Debian Project to be non-free, many
Debian users will continue using it.  There'll probably be an attempt
at a fork from the last known free license, or a pre-translation of
the API to avoid the need for further work by end-users, and the
results may someday end up as Debian packages.  

As has been said here before, the right way to resolve this appears to
involve contacting the potential copyright holder, Knuth.  It wouldn't
be unreasonable to include a bit of background on the current
discussion, a copy of the DFSG, and a short series of questions as to
whether TeX, MF, and CM are in the public domain or copyrighted, and
if the latter, what are the licenses.

- -Brian
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.0.7 (GNU/Linux)

iD8DBQE9ds2i03mlJHngJfERAkOGAJ9ztmFjUS3rkfpxvnCOoTIpRTnbIQCePuzB
G/iG0T/4Sj8DKs1+yWbFasA=
=Adsv
-END PGP SIGNATURE-

Re: Knuth statement on renaming cm files and Licence violation.

[Chris Lawrence]

On Sep 04, Russ Allbery wrote:
> I don't find your argument particularly persuasive; it seems to be very
> strong on emotion without a lot of logic to back it up, or without any
> real discussion of what you're trying to defend and why.

The Debian Project has a philosophical commitment to protecting the
freedoms of the users of software that it calls "free".  These
freedoms are spelled out in the Debian Social Contract and Debian Free
Software Guidelines.

You can argue whether the freedom to rename some particular file is
important or not, but that's largely beside the point as far as Debian
is concerned; it is possible for reasonable people to disagree about
the relative importance of that (or any other) freedom.  However, we
believe that irrespective of whether we intend to exercise the
particular rights in question, possessing them (and, more importantly,
ensuring our users possess them) is important.

For example, the DFSG has a paragraph about non-discrimination.
Debian has no intention of setting up a nuclear power plant, but a
license that restricts people who own nuclear power plants from using
our software [licenses like this do, in fact, exist] is unacceptable.
Similarly, Debian has no intention of violating Prof. Knuth's request
that the Computer Modern fonts not be replaced without renaming them,
yet we are unwilling to call them "free" unless our end users have the
freedom to do so.  (Leave aside whether Prof. Knuth's request is
legally binding; his statement that the fonts are "in the public
domain" suggests that no request of his regarding the fonts is legally
binding, although his wishes should not be lightly disregarded.)


Chris
-- 
Chris Lawrence <[EMAIL PROTECTED]> - http://www.lordsutch.com/chris/

Computer Systems Manager, Physics and Astronomy, Univ. of Mississippi
125B Lewis Hall - 662-915-5765

Re: Knuth statement on renaming cm files and Licence violation.

[Russ Allbery]

Brian Sniffen <[EMAIL PROTECTED]> writes:

> Bear in mind, Russ, nobody is questioning whether TeX (or LaTeX) are
> *good* software, or *useful* software, or even *open source* software.

I understand.

> The question is whether they are free software.  A restricted API, which
> you call a protected API, is not a free API.

I understand that viewpoint and even agree with it to a degree.

On the other hand, this has been the copying policy for TeX fonts since
pretty much the beginning of TeX.  This isn't new.  Which makes me
confused when I see comments like:

> There'll probably be an attempt at a fork from the last known free
> license,

because it seems to me like they misunderstand the nature of the
situation.  So far as I know, this is very basic to the nature of TeX;
according to Knuth's writings that I'm familiar with, it was part of his
goal in how he wrote the TeX system from very early on.  There is no "last
known free license" that I've heard of in the sense that you mean.

RMS considered TeX part of the GNU System from the writings that I'm
familiar with since very early on in the development of that system, so
apparently, at least from that, did not have a problem with the copying
policy.  I suppose it's possible that he was unaware of it, but that seems
very much unlike his normal extreme care with such things.

Now, if the new LaTeX license is going beyond that, that's another matter,
and I want to be careful to point out here that I'm not commenting
directly on that, only specifically on the issue of the CM fonts, because
I have more of a grasp on that.  I also think that authors of LaTeX
packages should really think about whether they actually need the
guarantee that Knuth was striving for, and whether they've done all of the
*other* things that Knuth did to achieve his goals.  Most of them,
frankly, have not, and this provision does nothing for them and is really
rather pointless and they would be much better served by some other
license.

What brought me into this argument, though, is basically a question of
principal, namely that licenses should be evaluated as if they were
enforcible even if written in a way that may make them legally
unenforcible for some reason and even if the person doesn't intend to
enforce them.  And actually, I doubt that people really disagree with me
on that.

If Knuth's actual position is that the legal license on his fonts is
public domain and that he has no interest in forcing anyone to use
different names, only to get upset at them if they do, and if that
satisfies the legal requirements of Debian wrt the DFSG, then that's all
fine and good.  But I personally still find it important that it's *not*
morally okay to modify the CM fonts and redistribute them under the same
names.

To me, one of the most important parts of the gift culture is to respect
the terms of the gift, if the giver felt like they wanted to put terms on
it for whatever reason of their own.  Even if they don't intend to enforce
those terms with lawyers.

-- 
Russ Allbery ([EMAIL PROTECTED]) 

Re: Knuth statement on renaming cm files and Licence violation.

[Thomas Bushnell, BSG]

Russ Allbery <[EMAIL PROTECTED]> writes:

> RMS considered TeX part of the GNU System from the writings that I'm
> familiar with since very early on in the development of that system, so
> apparently, at least from that, did not have a problem with the copying
> policy.  I suppose it's possible that he was unaware of it, but that seems
> very much unlike his normal extreme care with such things.

TeX doesn't *have* a renaming requirement, actually.  You can call the
files *anything you want*, but you can't call the system "TeX" unless
you meet certain conditions.  Naming requirements on packages and such
are harmless, naming requirements on *files* are a much different
question. 

> What brought me into this argument, though, is basically a question of
> principal, namely that licenses should be evaluated as if they were
> enforcible even if written in a way that may make them legally
> unenforcible for some reason and even if the person doesn't intend to
> enforce them.  And actually, I doubt that people really disagree with me
> on that.

It depends.  If the question is "what will Debian do" then we act
liberally as you suggest we should.

If the question is "what can our users do", we might well decide that
we want to interpret things differently, with the goal of arguing (on
behalf of our users) for the broadest possible freedoms.

> If Knuth's actual position is that the legal license on his fonts is
> public domain and that he has no interest in forcing anyone to use
> different names, only to get upset at them if they do, and if that
> satisfies the legal requirements of Debian wrt the DFSG, then that's all
> fine and good.  But I personally still find it important that it's *not*
> morally okay to modify the CM fonts and redistribute them under the same
> names.

Fine, but nobody in Debian is considering doing that.  

Re: Knuth statement on renaming cm files and Licence violation.

[Russ Allbery]

Thomas Bushnell, BSG <[EMAIL PROTECTED]> writes:
> Russ Allbery <[EMAIL PROTECTED]> writes:

>> RMS considered TeX part of the GNU System from the writings that I'm
>> familiar with since very early on in the development of that system, so
>> apparently, at least from that, did not have a problem with the copying
>> policy.  I suppose it's possible that he was unaware of it, but that
>> seems very much unlike his normal extreme care with such things.

> TeX doesn't *have* a renaming requirement, actually.  You can call the
> files *anything you want*, but you can't call the system "TeX" unless
> you meet certain conditions.  Naming requirements on packages and such
> are harmless, naming requirements on *files* are a much different
> question.

The CM fonts are generally considered to be part of the TeX system, since
they're the default fonts, and I believe they've had this renaming
requirement at least for quite some time.  I certainly remember it clearly
from when I first started using TeX in the early 1990s and had the
impression that it was fairly old then.

> If the question is "what can our users do", we might well decide that we
> want to interpret things differently, with the goal of arguing (on
> behalf of our users) for the broadest possible freedoms.

Right, I understand your distinction between what Debian does and what its
users do, and it makes sense to me.  I just still would never say that
it's okay to break this kind of request, although I might say that it's
legally permissible.

-- 
Russ Allbery ([EMAIL PROTECTED]) 

Re: Knuth statement on renaming cm files and Licence violation.

[Thomas Bushnell, BSG]

Russ Allbery <[EMAIL PROTECTED]> writes:

> The CM fonts are generally considered to be part of the TeX system, since
> they're the default fonts, and I believe they've had this renaming
> requirement at least for quite some time.  I certainly remember it clearly
> from when I first started using TeX in the early 1990s and had the
> impression that it was fairly old then.

See, this is where precision is really necessary.

First, "the TeX system" is a vague term, and best avoided entirely
when talking about legal things, since it doesn't enter into any legal
documents in question.

Second, the CM fonts contain, individually for each source file, a
requirement that if the source file is changed, it must be renamed.
There is nowhere a requirement placed on what the compiled .tfm can be
called.  

> Right, I understand your distinction between what Debian does and what its
> users do, and it makes sense to me.  I just still would never say that
> it's okay to break this kind of request, although I might say that it's
> legally permissible.

That's fine.  What Debian does is what's *okay*, which is only a subset
of what is legally permissible.  But what we insist on for our users
is a matter of legal permissibility.  That is, we insist on their
right to be an asshole, even if we have no intention of being such
ourselves.

Thomas

Re: Knuth statement on renaming cm files and Licence violation.

[Russ Allbery]

Chris Lawrence <[EMAIL PROTECTED]> writes:

> The Debian Project has a philosophical commitment to protecting the
> freedoms of the users of software that it calls "free".  These freedoms
> are spelled out in the Debian Social Contract and Debian Free Software
> Guidelines.

> You can argue whether the freedom to rename some particular file is
> important or not, but that's largely beside the point as far as Debian
> is concerned; it is possible for reasonable people to disagree about the
> relative importance of that (or any other) freedom.  However, we believe
> that irrespective of whether we intend to exercise the particular rights
> in question, possessing them (and, more importantly, ensuring our users
> possess them) is important.

I understand that you're trying to simplify the argument to try to explain
it to me, but you're simplifying so far that you're not usefully
representing the discussion.

The DFSG consider many freedoms to be unimportant.  The freedom to
distribute modified software directly rather than in the form of patches
is not considered important, for example; see point four.  The freedom to
incorporate software into a proprietary product is not considered
important; the freedom to always have modifiable software is considered
overriding.  And so forth.

So while I understand and appreciate your point of view, we wouldn't be
having this discussion if it were as simple as all that.  One cannot
simply say "we support freedom and that's all there is to it" because
that's not what the DFSG actually say.  They say that Debian supports some
specific freedoms and considers other ones to be less important, and the
question is whether a given software package fits into those guidelines or
not.

The arguments that have been presented that say that requiring file
renaming is an infringment on the freedoms guaranteed by the DFSG are
certainly reasonable ones and I can find much in them to agree with, but
the DFSG really *aren't* clear on this point, and a ruling on the subject
does not just obviously fall out of what's already there.

(As an aside, once Debian reaches some sort of general conclusion on this,
it would be really nice to add that to either the DFSG or some supporting
material, since this has come up repeatedly for years and this exact
argument happens every time.)

-- 
Russ Allbery ([EMAIL PROTECTED]) 

Re: Knuth statement on renaming cm files and Licence violation.

[Thomas Bushnell, BSG]

Russ Allbery <[EMAIL PROTECTED]> writes:

> The arguments that have been presented that say that requiring file
> renaming is an infringment on the freedoms guaranteed by the DFSG are
> certainly reasonable ones and I can find much in them to agree with, but
> the DFSG really *aren't* clear on this point, and a ruling on the subject
> does not just obviously fall out of what's already there.

DFSG says that you have to permit modification.  (By patches or
directly.)  That is violated by a rule like "if you modify this, you
must chant the kama sutra" or "if you modify this, you cannot name the
output file foo.bar".  The reason the latter is crucial is because it
is an *operational* matter for the software, changing such things as
APIs is exactly why we want the right to modify files.

> (As an aside, once Debian reaches some sort of general conclusion on this,
> it would be really nice to add that to either the DFSG or some supporting
> material, since this has come up repeatedly for years and this exact
> argument happens every time.)

Eww, no.  The current method is actually better.  It takes time and
patience, but the result is an ever-growing cadre of people who have
thought it out and talked about it, and so much more deeply understand
it than if they just read it in some document.

Re: Knuth statement on renaming cm files and Licence violation.

[Jeff Licquia]

On Wed, 2002-09-04 at 18:40, Russ Allbery wrote:
> Thomas Bushnell, BSG <[EMAIL PROTECTED]> writes:
> > Russ Allbery <[EMAIL PROTECTED]> writes:
> >> Thomas Bushnell, BSG <[EMAIL PROTECTED]> writes:
> 
> >>> And note that it begins with "I decided to put these fonts into the
> >>> public domain; all I have asked is that ..."
> 
> >>> Note that if this means *anything* at all, the request is a mere
> >>> request and not legally binding.
> 
> >> Surely you're not arguing that makes any difference.  Surely you're not
> >> claiming that Debian should ignore a very reasonable request by Donald
> >> Knuth because you think you've found a legal loophole in his statement.
> >> Surely you're just saying this to make an obscure pedantic point, not
> >> to indicate that it should have anything whatsoever to do with what
> >> Debian does.
> 
> > I fear that you have lost track of the context of the (rather complex)
> > discussion.
> 
> Perhaps in part, but your line of reasoning above makes me extremely
> uncomfortable.  When the author of a package has clearly stated their
> expectations and requirements for redistribution, it seems like that
> should be treated as the license, even if they didn't cross the
> appropriate legal "t".  To do anything else seems... well, extremely
> rude.  Even if Debian is not violating the intended license directly, to
> base a stance on the viewpoint that the license is legally uninforcable
> and therefore irrelevant seems rather disconcerting.

To repeat, there's a lot of context you're missing.

Debian distributes TeX in main.  When we distribute something in main,
we are saying to our users, "You may do these things with TeX and not
worry about violating the law."  Is it wrong to ask ourselves if this
statement is accurate?

Re: Knuth statement on renaming cm files and Licence violation.

[Branden Robinson]

On Wed, Sep 04, 2002 at 08:09:55PM -0700, Russ Allbery wrote:
> I don't find your argument particularly persuasive; it seems to be very
> strong on emotion without a lot of logic to back it up, or without any
> real discussion of what you're trying to defend and why.

That's because some of us have grown tired of reiterating the same
points over and over, without any substantive -- logical -- rebuttal.

If you want logical arguments, catch up on 2 month's worth of list
archives:

http://lists.debian.org/debian-legal/2002/debian-legal-200207/threads.html
http://lists.debian.org/debian-legal/2002/debian-legal-200208/threads.html

Refusing to countenance the logical arguments that have been presented
ad nauseum for the past couple of months and just griping about
non-logical ones is a good way to get yourself ignored.

If you want a logical argument, get yourself up to speed and engage us
on that basis.

If, on the other hand, all you want to do is to burst in, say "well, you
guys are wrong and you're exploiting the great Don Knuth and what's
wrong with you folks and oh well I can't be troubled to see what you
have to say about this matter list archives? what are those you know
it's just completely inconceivable to me that anyone's has anything
substantive to say about this subject before I showed up and in any
event I'm just convinced you're wrong and I think I'll babble on a
little more about how misguided and exploitative you are and my aren't
you guys a little tetchy when I do so has it ever occured to you to use
logic?", then please go away.

Because that's just what you sound like.

If you want logic, establish your premises with the aid of 2 months'
worth of context.

If you just want to bitch, consider yourself heard.  Now move along.

-- 
G. Branden Robinson|I've made up my mind.  Don't try to
Debian GNU/Linux   |confuse me with the facts.
[EMAIL PROTECTED] |-- Indiana Senator Earl Landgrebe
http://people.debian.org/~branden/ |


pgpUY8wX1Kvir.pgp
Description: PGP signature

Re: Knuth statement on renaming cm files and Licence violation.

[Branden Robinson]

On Wed, Sep 04, 2002 at 08:48:16PM -0700, Russ Allbery wrote:
> Right, I understand your distinction between what Debian does and what its
> users do, and it makes sense to me.  I just still would never say that
> it's okay to break this kind of request, although I might say that it's
> legally permissible.

You're entitled to your opinion, and to make all the moral judgements
you can squeeze into your waking hours.

Your emotions are, however, utterly and completely irrelevant to the
case at issue.  We're talking about intellectual property laws.  You're
talking about how people can hold hands and render proper respect unto
Professor Knuth and love and share and make the world a better place.

Such things are beyond the scope of debian-legal.  The DFSG is an
instrument in service of Debian's Social Contract, which is a document
more oriented towards the universal peace and harmony issues you're
concerned about.  The proper forum for discussion of the Debian Social
Contract is the debian-project mailing list.

Debian-legal is happy to stipulate that the freedoms the DFSG seeks to
guarantee won't necessarily be exercised to friendly and polite ends in
all situations.

Now, please, move along, for your observations are off-topic.

-- 
G. Branden Robinson| No math genius, eh?  Then perhaps
Debian GNU/Linux   | you could explain to me where you
[EMAIL PROTECTED] | got these...   PENROSE TILES!
http://people.debian.org/~branden/ | -- Stephen R. Notley


pgp1ufMOHi3bc.pgp
Description: PGP signature

Re: Knuth statement on renaming cm files and Licence violation.

[David Carlisle]

> Then he must be giving his earnest encouragements, and not a license,
> since the work is public domain, right?

I don't know.

I can understand the principle that if something is PD then you can't
claim licence conditions on it, however some who've indicated that they
have a better grasp of the law than I do have indicated that if you try
to simultaneously say something is PD and assert conditions then the law
will err on the side of caution and the work is in fact not PD even if
you said it was. The conditions as stated in the book (and comments in
the files) are not worded as "encouragements" they are worded as
constraints that must be followed.

David




_
This message has been checked for all known viruses by Star Internet
delivered through the MessageLabs Virus Scanning Service. For further
information visit http://www.star.net.uk/stats.asp or alternatively call
Star Internet for details on the Virus Scanning Service.

Re: Knuth statement on renaming cm files and Licence violation.

[Brian Sniffen]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1


> On Wed, 04 Sep 2002 20:36:23 -0700, Russ Allbery <[EMAIL PROTECTED]> said:

> On the other hand, this has been the copying policy for TeX fonts since
> pretty much the beginning of TeX.  This isn't new.  Which makes me
> confused when I see comments like:

>> There'll probably be an attempt at a fork from the last known free
>> license,

> because it seems to me like they misunderstand the nature of the
> situation.  So far as I know, this is very basic to the nature of TeX;
> according to Knuth's writings that I'm familiar with, it was part of his
> goal in how he wrote the TeX system from very early on.  There is no "last
> known free license" that I've heard of in the sense that you mean.

And if you'd taken that sentence in the context of its paragraph,
you'd have seen that it referred explicitly to LaTeX.  Please check
the debian-legal archives for the past summer to gain context in the
recent LPPL debate.  It's been fascinating to watch.

As far as the CM fonts go, if the end resolution of this is that their
license prohibits using the same .mf or .tfm file names for derivative
works as for the modified works (which seems unlikely), then I'd
expect one of two things to happen:

1. At a cost of great effort, new fonts with the same names as the CM
   fonts are built.  They are probably not as good, but they are free
   and thus continually improve.  Sufficient noise is made about this
   in the free software community that tetex and many Linux
   distributions start using them by default.

2. The file names are changed in the Debian tetex bundle, and all the
   APIs updated.  Now it's possible to make arbitrary changes without
   having to change any extra functional parts, though there's still an
   annoying license about what names you may change *to*.

Fortunately, as Thomas has repeatedly explained, the license on the CM
fonts appears to require changing only the .mf file name, which is not
a functional part.

- -Brian

- -- 
Brian Sniffen [EMAIL PROTECTED]
Security Engineer day: (617) 444-2642cel: (617) 721-0927
Akamai Technologies   eve: (617) 354-1526 pi: (314) 159-2654 
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.0.7 (GNU/Linux)

iD8DBQE9d1TVJ6fLHGswGokRAkrDAKDLprD6AVVi2YEb5hXMtgHqrNjaxwCbB9lN
5W6sKSDeXiRHLPq6MBthtC0=
=3rT0
-END PGP SIGNATURE-

Re: Knuth statement on renaming cm files and Licence violation.

[Jeremy Hankins]

I don't really have anything to say that hasn't been said by others in
the past, but since some folks are (understandably, really) tired of
the issue, I thought I'd try to explain the standard debian-legal take
on it.  Someone correct me if I'm inaccurate.

Russ Allbery <[EMAIL PROTECTED]> writes:

> What brought me into this argument, though, is basically a question of
> principal, namely that licenses should be evaluated as if they were
> enforcible even if written in a way that may make them legally
> unenforcible for some reason and even if the person doesn't intend to
> enforce them.  And actually, I doubt that people really disagree with me
> on that.

No, I don't think anyone disagrees with that.  I think the problems
comes up because your scenario is a bit ambiguous; it describes some
very different situations:

A) Someone releases software under a license that debian-legal thinks
is free (quite self-evidently, perhaps -- the pine license is a
real-world example of this, IIRC).  But the licensor comes to us and
tells us that our interpretation is all wrong, that the license does
not permit X, Y, or Z, which would clearly make it non-free.
Debian-legal will (and has) decide that this makes the license
non-free, even though the same license in use by someone else would
likely be considered free.

B) Someone releases software under a license and claims it is free,
but debian-legal looks at it and doesn't think that the license
permits the freedoms that the licensor says it does.  This puts us in
a awkward spot, we don't know whether the licensor doesn't understand
the freedoms or the license.  Generally someone tries to talk to the
licensor and figure out what's going on before deciding for sure
whether the license is free (and hopefully gets the ambiguous
requirements cleared up in the process).

C) Someone releases software under a free license, but with addenda
that make requests that are not legally binding.  These addenda would
perhaps make the license non-free if they were legally binding, but
since they are not the license is considered free.  The author may do
this because he thinks that the law is too blunt an instrument to
accomplish his purposes, and so will use more social mechanisms
(public condemnation, boycott, etc.) to further his ends.  My
understanding is that most folks on debian-legal are quite happy with
this approach, and in fact will often recommend it in cases where the
author has a particular concern with respect to the software.

Respect for the author's opinion means a different thing in all three
of these situations, and it isn't always clear which scenario is
active in any given situation.  The question in the current situation
is which of these scenarios applies.  If it's A (the license is
ambiguous and/or Knuth claims it places restrictions that aren't DFSG
free, even though debian-legal doesn't see that in the license text)
then the license may well be non-free, and further discussion
(preferably with the copyright holder, which would presumably be
Knuth) is required.  If it's C (which to my eye seems as likely, if
not more so, as a reading of the text) then there's no problem.

To be honest, given that there's such disagreement on the issue, I
don't really see how it can be resolved without contacting Knuth,
explaining the issues to him, and getting his verdict.


Note, though, that I think everyone here agrees that the extra
requirements in scenario C are *morally* as important to follow as the
legal ones.  But nonetheless there is an important distinction between
a request (however strongly worded and important) and a legal
requirement.

-- 
Jeremy Hankins <[EMAIL PROTECTED]>
PGP fingerprint: 748F 4D16 538E 75D6 8333  9E10 D212 B5ED 37D0 0A03

Re: Knuth statement on renaming cm files and Licence violation.

[Martin Schröder]

On 2002-09-05 08:57:57 -0400, Brian Sniffen wrote:
> Fortunately, as Thomas has repeatedly explained, the license on the CM
> fonts appears to require changing only the .mf file name, which is not
> a functional part.

IBTD. The common interpretation in the TeX community IIRC is that
a file cmr10.tfm must contain the same metrices and encodings as
Knuth's. You may change the font (i.e. the appearance of the
glpyhs), but the metrices must be Knuth's.

Best regards
Martin
-- 
   Martin Schröder, [EMAIL PROTECTED]
  ArtCom GmbH, Grazer Straße 8, D-28359 Bremen
  Voice +49 421 20419-44 / Fax +49 421 20419-10

Re: Knuth statement on renaming cm files and Licence violation.

[Russ Allbery]

Brian Sniffen <[EMAIL PROTECTED]> writes:

> And if you'd taken that sentence in the context of its paragraph, you'd
> have seen that it referred explicitly to LaTeX.

You're right; I'm sorry.  My mind was going faster than my eyes, and I
apologize for the misunderstanding.

Anyway, thank you for putting up with the intrusion; I think I've
explained what I was trying to explain and I now have a much better
understanding of where other people are coming from.  I appreciate the
discussion.

-- 
Russ Allbery ([EMAIL PROTECTED]) 

Re: Knuth statement on renaming cm files and Licence violation.

[Thomas Bushnell, BSG]

Martin Schröder <[EMAIL PROTECTED]> writes:

> IBTD. The common interpretation in the TeX community IIRC is that
> a file cmr10.tfm must contain the same metrices and encodings as
> Knuth's. You may change the font (i.e. the appearance of the
> glpyhs), but the metrices must be Knuth's.

The common interpretation, however, has no discernable relation to the
actual license on the actual files.

Re: Knuth statement on renaming cm files and Licence violation.

[Dylan Thurston]

On Wed, 4 Sep, Brian Sniffen wrote:
> Sadly, I don't own a copy of Computers & Typesetting.  Can you quote
> the full copyright page, and give a general indication of the contents
> of Volume E?

Somewhat surprisingly, no-one has done this completely yet.  Computers
& Typesetting, Volume E, Computer Modern Typefaces, is a nicely
typeset version of the complete source code to the Computer Modern
fonts, typeset in the literate programming style with diagrams of the
characters and explanatory notes.  I would consider it a canonical
distribution of the Computer Modern typefaces.

Here is the complete copyright page.  The most relevant section (the
fourth paragraph) was already posted by Martin Schr??der.

---
The quotations on pages 7 and 351 have been excerpted from the Electra
file in the Dwiggins Collection of the Boston Public Library.

METAFONT is a trademark of Addison-Wesley Publishing Company.

TeX is a trademarke of the American Mathematical Society.

The programs for Computer Modern are in the public domain, and readers
may freely generate and hand-tune their own fonts using the algorithms
of this book.  However, use of the names is restricted: Any fonts
whose names cmr10 or cmbx12 or ... are identical to the standard font
names of this book should be fully compatible with the fonts defined
here; i.e., fonts with the same names are supposed to have precisely
the same character coding schemes and precisely the same font metric
files.

Library of Congress Cataloging-in-Publication Data

Knuth, Donald Ervin, 1938-
   Computer Modern typefaces.

   (Computers & Typesetting ; E)
   Includes indexes.
   1. Type and type-founding--Data processing.
2. Printing--Specimens.  3. METAFONT (Computer system).
4. Computerized typesetting.  I. Title.  II. Series:
Knuth, Donald Ervin, 1938-.  Computers &
typesetting ; E
Z253.K568  1986 686.2'2544  86-1235
ISBN 0-201-13446-2





Incorporates the final corrections made in 1992.

Copyright (c) 1986 by Addison-Wesley Publishing Company, Inc.

All rights reserved.  No part of this publication may be reproduced,
stored in a retrieval system, or transmitted, in any form or by any
means, electronic, mechanical, photocopying, recording, or otherwise,
without the prior written permission of the publishers.  Printed in
the United States of America.

ISBN 0-201-13446-2
4 5 6 7 8 9 10 11 12 13-HA-9998979695949392
--
(The "..." there is really in the original text.)

(By the way, volumes B and D are about the TeX and METAFONT programs,
respectively.  They presumably have similar copyright pages, but I
haven't checked.)

The statement about the programs for Computer Modern is suprisingly
vague for something that was presumably vetted by a lawyer.  Let's see
it again:

> The programs for Computer Modern are in the public domain, and readers
> may freely generate and hand-tune their own fonts using the algorithms
> of this book.

This place Computer Modern in the public domain, and furthermore
explicitly grants some of the rights Debian needs (although it leaves
out the right to redistribute).

> However, use of the names is restricted:

This is a slightly odd statement, since (AFAIK) names cannot be
restricted in the ways that follow.  The crucial issue seems to be
whether this statement (and what follows) are terms of the grant of
permission (above), or merely a request with no force in law.

> Any fonts whose names cmr10 or cmbx12 or ... are identical to the
> standard font names of this book should be fully compatible with the
> fonts defined here; i.e., fonts with the same names are supposed to
> have precisely the same character coding schemes and precisely the
> same font metric files.

To back up the notion that this is merely a request, I note the words
"should" and "supposed".  (I also note that there is no mention of
specific _filenames_, merely the name of the _font_; and that there is
some freedom to modify fonts under the same font name.)


I think that, for the fonts to be distributable by Debian under this
copyright notice, the statement about "public domain" has to be taken
seriously, since otherwise there is no permission to distribute.  Does
Debian need legal advice on whether this statement actually places the
files in the public domain?  Or does it make more sense to approach
Knuth directly?  If we do approach Knuth, the letter should be
carefully worded.

Best,
Dylan Thurston


pgpdiA0RxI0W9.pgp
Description: PGP signature

Re: Knuth statement on renaming cm files and Licence violation.

[Thomas Bushnell, BSG]

Dylan Thurston <[EMAIL PROTECTED]> writes:

> > However, use of the names is restricted:
> 
> This is a slightly odd statement, since (AFAIK) names cannot be
> restricted in the ways that follow.  The crucial issue seems to be
> whether this statement (and what follows) are terms of the grant of
> permission (above), or merely a request with no force in law.

The names could only be restricted if they are trademarked, which they
are not.  "Computer Modern" might be trademarked (I don't know), but
"cmr10" certainly is not.

And, if it were trademarked, trademark restrictions never apply to
functional elements.

Re: Knuth statement on renaming cm files and Licence violation.

[Dylan Thurston]

On Fri, Sep 06, 2002 at 03:35:17PM -0700, Thomas Bushnell, BSG wrote:
> > > However, use of the names is restricted:
> > 
> > This is a slightly odd statement, since (AFAIK) names cannot be
> > restricted in the ways that follow.  The crucial issue seems to be
> > whether this statement (and what follows) are terms of the grant of
> > permission (above), or merely a request with no force in law.
> 
> The names could only be restricted if they are trademarked, which they
> are not.  "Computer Modern" might be trademarked (I don't know),

It is, as indicated in the text I quoted and you snipped.  It's a
trademark of Addison-Wesley.

> but "cmr10" certainly is not.

Right.

> And, if it were trademarked, trademark restrictions never apply to
> functional elements.

... like the file name 'cmr10.tfm'.  (Just making everything explicit.)

--Dylan


pgpqH7RLjqHMA.pgp
Description: PGP signature

Re: Knuth statement on renaming cm files and Licence violation.

[Martin Schröder]

On 2002-09-06 18:59:45 -0400, Dylan Thurston wrote:
> On Fri, Sep 06, 2002 at 03:35:17PM -0700, Thomas Bushnell, BSG wrote:
> > The names could only be restricted if they are trademarked, which they
> > are not.  "Computer Modern" might be trademarked (I don't know),
> 
> It is, as indicated in the text I quoted and you snipped.  It's a
> trademark of Addison-Wesley.

Uh? METAFONT is a TM of Addison-Wesley. 
What makes you think "Computer Modern" is a TM of Addison-Wesley?

Best regards
Martin
-- 
http://www.tm.oneiros.de

Re: Knuth statement on renaming cm files and Licence violation.

[Kai Henningsen]

[EMAIL PROTECTED] (Brian Sniffen)  wrote on 04.09.02 in <[EMAIL PROTECTED]>:

> Bear in mind, Russ, nobody is questioning whether TeX (or LaTeX) are
> *good* software, or *useful* software, or even *open source* software.
> The question is whether they are free software.

Statements like this really piss me off.

Open source software *is* free software, by definition, without exception.

In fact, given that "open source" is the same as "DFSG free",

 *by definition*,

then if anything is not DFSG free then it's not open source, either.

MfG Kai

Re: Knuth statement on renaming cm files and Licence violation.

[Kai Henningsen]

[EMAIL PROTECTED] (Thomas Bushnell, BSG)  wrote on 04.09.02 in <[EMAIL 
PROTECTED]>:

> Russ Allbery <[EMAIL PROTECTED]> writes:
>
> > The arguments that have been presented that say that requiring file
> > renaming is an infringment on the freedoms guaranteed by the DFSG are
> > certainly reasonable ones and I can find much in them to agree with, but
> > the DFSG really *aren't* clear on this point, and a ruling on the subject
> > does not just obviously fall out of what's already there.
>
> DFSG says that you have to permit modification.  (By patches or
> directly.)  That is violated by a rule like "if you modify this, you
> must chant the kama sutra" or "if you modify this, you cannot name the
> output file foo.bar".

It is not actually clear to me that this is violated by any of the example  
rules.

>  The reason the latter is crucial is because it
> is an *operational* matter for the software, changing such things as
> APIs is exactly why we want the right to modify files.

And we usually get upset if a library changes its API without changing its  
soname or versioning its symbols. *Because* this is an operational matter.

> > (As an aside, once Debian reaches some sort of general conclusion on this,
> > it would be really nice to add that to either the DFSG or some supporting
> > material, since this has come up repeatedly for years and this exact
> > argument happens every time.)
>
> Eww, no.  The current method is actually better.  It takes time and
> patience, but the result is an ever-growing cadre of people who have
> thought it out and talked about it, and so much more deeply understand
> it than if they just read it in some document.

This is insane.


MfG Kai

Re: Knuth statement on renaming cm files and Licence violation.

[Steve Langasek]

On Sat, Sep 07, 2002 at 08:36:00PM +0200, Kai Henningsen wrote:
> [EMAIL PROTECTED] (Brian Sniffen)  wrote on 04.09.02 in <[EMAIL PROTECTED]>:

> > Bear in mind, Russ, nobody is questioning whether TeX (or LaTeX) are
> > *good* software, or *useful* software, or even *open source* software.
> > The question is whether they are free software.

> Statements like this really piss me off.

> Open source software *is* free software, by definition, without exception.

> In fact, given that "open source" is the same as "DFSG free",

>  *by definition*,

> then if anything is not DFSG free then it's not open source, either.

By whose definition?

Open Source is not defined in terms of the DFSG, nor are the DFSG defined
in terms of Open Source.  The two terms have two *independent*
definitions, and determination of whether a piece of software is "DFSG
free" or "Open Source" is made by two independent groups, in accordance
with their respective guidelines.  Though I can't cite any off-hand, ISTR
that there have been real cases where an OSI-approved license was
regarded as DFSG-noncompliant.

Despite Bruce's involvement in the crafting of both terms, they are not
synonyms.

Steve Langasek
postmodern programmer


pgpYhyX7CRxCm.pgp
Description: PGP signature

Re: Knuth statement on renaming cm files and Licence violation.

[Thomas Bushnell, BSG]

[EMAIL PROTECTED] (Kai Henningsen) writes:

> > DFSG says that you have to permit modification.  (By patches or
> > directly.)  That is violated by a rule like "if you modify this, you
> > must chant the kama sutra" or "if you modify this, you cannot name the
> > output file foo.bar".
> 
> It is not actually clear to me that this is violated by any of the example  
> rules.

Well, what about "if you modify this, you must pay a licensing fee"?
The point is that modifications must be practically unrestricted.  If
this was not clear to you before, I hope it is now.

> >  The reason the latter is crucial is because it
> > is an *operational* matter for the software, changing such things as
> > APIs is exactly why we want the right to modify files.
> 
> And we usually get upset if a library changes its API without changing its  
> soname or versioning its symbols. *Because* this is an operational matter.

Right.  Such a change is an unfriendly one, and one we don't do.  But
it's also crucial to freedom that we have the right to do it, because
someday it may be the only way to fix a bug.

> This is insane.

Well, how friendly!

Re: Knuth statement on renaming cm files and Licence violation.

[Dylan Thurston]

Martin Schr??der wrote:
> On 2002-09-06 18:59:45 -0400, Dylan Thurston wrote:
> > On Fri, Sep 06, 2002 at 03:35:17PM -0700, Thomas Bushnell, BSG wrote:
> > > The names could only be restricted if they are trademarked, which they
> > > are not.  "Computer Modern" might be trademarked (I don't know),
> > It is, as indicated in the text I quoted and you snipped.  It's a
> > trademark of Addison-Wesley.
> 
> Uh? METAFONT is a TM of Addison-Wesley. 
> What makes you think "Computer Modern" is a TM of Addison-Wesley?

Quite right; my mistake.  I think if "Computer Modern" were a
trademark it would doubtless have been mentioned on the copyright page
I quoted.

--Dylan


pgpLghKvDtuHE.pgp
Description: PGP signature

Re: Knuth statement on renaming cm files and Licence violation.

[Brian Sniffen]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1



Computer Modern appears to be a trademark of the American Mathematical
Society.  I don't know what impact a lack of mention of that in vol
E. would have.

- -Brian

-BEGIN PGP SIGNATURE-
Version: GnuPG v1.0.7 (GNU/Linux)

iD8DBQE9etQr03mlJHngJfERAujmAJwO1+T5StkLUcNIZ6rPynrP8RT8VgCgg+KQ
pSbodJJyUujt/bs67NUkS8M=
=QkWr
-END PGP SIGNATURE-

Re: Knuth statement on renaming cm files and Licence violation.

[Martin Schröder]

On 2002-09-08 00:37:59 -0400, Brian Sniffen wrote:
> Computer Modern appears to be a trademark of the American Mathematical
> Society.  I don't know what impact a lack of mention of that in vol

What makes you think that?

Best regards
Martin
-- 
http://www.tm.oneiros.de

Re: Knuth statement on renaming cm files and Licence violation.

[Brian Sniffen]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1



Some google searching and earlier discussion here indicated that CM
and TeX were trademarks of the AMS, while MF was a trademark of A-W
Publishing.

Knuth's own page on C&T lists only the trademarks of TeX and MF, so
perhaps the CM trademark has gone unenforced or been dropped.

- -Brian 

- -- 
Brian Sniffen [EMAIL PROTECTED]
Security Engineer day: (617) 444-2642cel: (617) 721-0927
Akamai Technologies   eve: (617) 354-1526 pi: (314) 159-2654 
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.0.7 (GNU/Linux)

iD8DBQE9e3gvJ6fLHGswGokRApEeAKCgttydXwBaq5bzzL+/lRJx6ypMPwCffpA1
KyJn1QchsEQy0ekLvYLDQqw=
=TWYJ
-END PGP SIGNATURE-

Re: Knuth statement on renaming cm files and Licence violation.

[Martin Schröder]

On 2002-09-08 12:17:44 -0400, Brian Sniffen wrote:
> Knuth's own page on C&T lists only the trademarks of TeX and MF, so
> perhaps the CM trademark has gone unenforced or been dropped.

Since none of the books (and none of the cm files) and no
literatur on TeX mention CM as a trademark, I strongly doubt that
there ever was one.

Best regards
Martin
-- 
http://www.tm.oneiros.de

Re: Knuth statement on renaming cm files and Licence violation.

[Glenn Maynard]

On Sun, Sep 08, 2002 at 10:56:47PM +0200, Martin Schröder wrote:
> > Knuth's own page on C&T lists only the trademarks of TeX and MF, so
> > perhaps the CM trademark has gone unenforced or been dropped.
> 
> Since none of the books (and none of the cm files) and no
> literatur on TeX mention CM as a trademark, I strongly doubt that
> there ever was one.

For the purposes of this discussion, does it matter?

-- 
Glenn Maynard

Re: Knuth statement on renaming cm files and Licence violation.

[Thomas Bushnell, BSG]

Brian Sniffen <[EMAIL PROTECTED]> writes:

> Computer Modern appears to be a trademark of the American Mathematical
> Society.  I don't know what impact a lack of mention of that in vol
> E. would have.

None.  A trademark holder has his trademark whether or not he
continually mentions the fact.