[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: TeX Licenses & teTeX (Was: Re: forwarded message from Jeff Licquia)

On Mon, Aug 05, 2002 at 06:59:56PM +0200, Frank Mittelbach wrote:
> but Don hasn't put his work out as a whole with a license

Then to what, exactly, do his statements in comp.text.tex on Wed, 23 Feb
1994 03:34:01 GMT apply?

To nothing at all?  Was he just talking to hear himself talk, or was he
talking about TeX, METAFONT, and the Computer Modern Fonts?

> but has put copyright notices and comments into individual files, for
> example, tex.web starting out as
> % This program is copyright (C) 1982 by D. E. Knuth; all rights are reserved.
> % Copying of this file is authorized only if (1) you are D. E. Knuth, or if
> % (2) you make absolutely no changes to your copy. (The WEB system provides
> % for alterations via an auxiliary file; the master file should stay intact.)
> or  cmr10.mf

Is a copyright holder allowed to re-license his work or clarify his
licensing statements?  Is Knuth not allowed to come back to tex.web
after 12 years and clarify his intent as he did in comp.text.tex?

> or from plain.tex (no copyright notice there and a somewhat contradictory
> statement if looked at it in isolation)
> % This is the plain TeX format that's described in The TeXbook.
> % N.B.: A version number is defined at the very end of this file;
> %       please change that number whenever the file is modified!
> % And don't modify the file under any circumstances.

Indeed, a somewhat contradictory pair of statements.  If your contention
is that Debian must ignore Knuth's statements to the public in
comp.text.tex, then prudence would dictate that Debian regard a
self-contradictory license as null; therefore, the license on the work
reverts to its default status under copyright law, which is "all rights
reserved".  The work would thus be non-DFSG-free, and not free software
under the TeX community's definition, either.

> etc. in addition he has explained his understanding of what he gave to
> the world and how he gave it, many times and that does not only
> involves the names "TeX" or "METAFONT" or "COMPUTER MODERN" but also
> that, if you run TeX on a simple document containing
>  \font\foo=cmr10
> then what is loaded is cmr10.tfm and precisely "THE cmr10 from
> Computer Modern" and not some modified copy. There had in fact be a
> huge uproar (by Don) once when somebody changed the CM fonts and
> distributed them under the original font names.

If that person had distributed the modified cmr10.tfm file on a floppy
disk in conjunction with a file that said "THIS IS NOT DON KNUTH'S
COMPUTER MODERN FONT", or in retail packaging that said "Waylon
Arbogast's Nifty Fonts (based on Computer Modern by Don Knuth)", do you
think Professor Knuth would have objected as strongly?

Is Professor Knuth claiming a poor man's trademark on the strings "cmr10"
or "cmr10.tfm"?

In other words, will Professor Knuth regard it as infrigment of his
copyright of someone uses the filenames "cmr10" or "crm10.tfm" for
something even if they aren't actually derived from his Computer Modern
fonts, but rather original works?

Will Professor Knuth regard it as infrigment of his copyright of someone
uses the filenames "cmr10" or "crm10.tfm" for something that isn't a
font at all?

>  > Such at-will revocation of a license would violate DFSG 1, 2, 3, 7, and
>  > arguably 5.
> at-will revocation of a license might violate anything, point is that in the
> past there hasn't been a real license  --- and what i was trying to
> communicate are the intentions behind it.

No, you're saying the actual licenses are right there in the files in
black and white, and that Debian cannot rely on Knuth's statements in
comp.text.tex to interpret them.

> I can't assert anything and I will not, i can only interpret how I interpret
> what Don said to me and to others in private and in public. To my best believe
> Don's intentions are file name based or more exactly interface based, eg on a
> system that runs TeX \input{plain} is loading a file plain.tex which is his
> plain.tex and not somebody elses, and the above paragraphs (to me) also says
> this with respect to fonts, ie
>   "And nobody is allowed to use the
>    names of the Computer Modern fonts 
> [note the plural, he means cmr10 cmr11 ...]
>   in Volume~E for any fonts that do not
>   produce identical {\tt.tfm} files.
> what he wants to prevent is that TeX picks up some cmr10.tfm if it isn't the
> original one, again the point isn't really the file name it is the work "The
> font CMR10" (which ends up being a file name restriction on most OS's)

Not a legal restriction, at least not one with any precedent of which I
am aware.

Note that there is a copyrighted work out there called "Microsoft Word".
At least the last time I used it, which was many years ago, the "file
name" that corresponded to this work was "MSWORD.EXE".  Microsoft has
legal remedies available to them if someone packages a work of their own
as "Microsoft Word", but to the best of my knowledge they can't attack
someone just for calling a file "MSWORD.EXE".

Again, we see the difference between the name of a work and the name of
a file on a computer.

> I'm however not going to proxy for Don even though I too wouldn't wish to see
> him interrupted in his other work.  But I don't think that anybody but him can
> in fact clarify this finally and it seems that such clarification is necessary.

That may be true.  If you insist that Knuth does not mean what I think
he said in
then I'll take your word for it, and TeX, METAFONT, and the Computer
Modern fonts will probably have to be regarded as DFSG-nonfree until we
can get a clarification from Professor Knuth.

> As far as I can see, the interpreation that I gave (as well as Boris and
> others) is broader than the interpretation that only the words TeX, METAFONT
> and Computer Modern are at Don's heart. So if Debian are accepting this
> broader interpretation (which in fact is the interface/filename restriction)
> Debian use that as a starting point to come to a conclusion what that means
> concerning DSFG.

There is not just what at "Don's heart".  There *are* practical
limitations on what one can accomplish with a copyright license.
Professor Knuth appears to be far more interested in issues that are
handled by trademark law than copyright law, however as far as I know he
has not filed for trademark protection in the terms TeX, METAFONT or
Computer Modern.  And even trademark law does not give the holder
completely unfettered power to remove words from unauthorized human

If I were a hostile interested party, I feel I could make a plausible
argument that Professor Knuth placed TeX in the public domain just like
his notice said he did, and therefore there is no copyright license on
it to stop me from calling a derivative work TeX.  However, this would
be dumb even if I were so motivated, because trademark protection can be
filed after quite a period of time (witness how William Della Croce
attempted to trademark "Linux" and was granted the trademark, whereupon
he began attempting to extort license fees from various parties,
prompting a lawsuit after which the trademark was transferred to Linus
Torvalds).  TeX, METAFONT, and Computer Modern are unique enough names
in their fields of commerce that unless there is some limitation on the
number of years one can go without filing for a trademark on a word
after it entered usage, I have little doubt that Knuth would be granted
his trademark application.

At any rate, I *suspect* that Professor Knuth is a reasonable man, and
does not intend absurd consequences.  I *suspect* that Professor Knuth
is interested in preserving the integrity of the identity of the works
into which he invested so much labor, and also in preserving freedom --
academic and otherwise -- among computer users and professionals.  I
*suspect* that Professor Knuth is willing to tolerate reasonable uses of
the terms "TeX", "METAFONT", and "Computer Modern", given that he has
stated specific and reasonable terms for their use.  I *suspect* that
Professor Knuth is not going to attempt to retract the copyright license
(ambiguously stated as it is, with a very confusing reference to the
"public domain") from someone who happened to use the string "cmr10" in
a filename.  (Let's say Carnegie Mellon University developed a regular
expression engine, and it saw 10 revisions -- should Knuth raise hell if
they called it "CMR10" for "Carnegie Mellon Regex version 10"?)

I *hope*, though I am not sure, that Professor Knuth would not mind if
someone modified the Computer Modern fonts and left the filenames the
same but provided conspicuous notices that "THIS IS NOT DON KNUTH'S
SAME" and distributed the fonts not as "Computer Modern", but as "Waylon
Arbogast's Nifty Fonts".

> If  Debian concludes that it means that all of Don's work is not DSFG-free then
> perhaps you better talk to him directly, assuming that that outcome is not in
> Debian's interest (after all that would then be only because people like me
> say that we interpret him in this way).

I suppose that if the TeX community shares your opinion, that will be

G. Branden Robinson                |
Debian GNU/Linux                   |     Cogitationis poenam nemo meretur.
branden@debian.org                 |
http://people.debian.org/~branden/ |

Attachment: pgpwDnNmi3Xq3.pgp
Description: PGP signature

Reply to: