[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 04:22:27PM -0400, Boris Veytsman wrote:
> Let us clarify this a bit. Suppose I change the file cmr10.tfm without
> changing its name. As long as TeX does not see it, I do not think
> Knuth objects. However, if TeX DOES see it, it does not behave like
> the trademarked TeX with the trademarked Computer Modern fonts,
> correct?  Therefore if I package and distribute the changed fonts with
> my TeX distribution without either (a) changing their name OR (b)
> taking reasonable effort to assert the user that my distribution is
> *not* TeX, I violate Knuth's TM, correct?

Well, as you noted, the TM (trademark) isn't Knuth's.  The trademarks
belong to the AMS and Addison-Wesley.  (Though I would hope they have
taken the time to consult with Knuth so as to not enforce the trademarks
in a way that he would find abhorrent.)

A license (whether it is a license on TeX as-a-work [copyright] or
TeX-as-a-name [trademark]), that requires that a modified cmr10.tfm
*EITHER* (a) change its filename OR (b) make a reasonable effort to
inform the user that the font itself, or the TeX system with which it is
distributed, it not official, would not violate the DFSG.

> Now suppose that I distribute these fonts *without* TeX. Do I violate
> the law? I do not know, but I think that if the inteneded use of the
> files is with TeX, I violate the spirit, if not the letter, of Knuth's
> license. 

I'd disagree.  I'd say that any license -- copyright or trademark --
that purports to tell people what they can or cannot name files on their
computers, even if they are original works, is a problem.

> Now the questions are:
> 1. Could we interpret Knuth's intentions as "you can do anything with
>    cmr10.tfm, as long as TeX(TM) behaves exactly in the same way" [and
>    I do not care about your unTeX]?

I am no longer comfortable speculating as to Knuth's intentions.

> 2. If we could, would this interpretation be deemed DFSG-free?

It depends on whether you mean Knuth's "cmr10.tfm" as a work, or as a

Your question is phrased in a way that makes it difficult for me to
answer clearly.  Let me make some assertions and hope I cover the bases.

* Users must have the freedom to modify cmr10.tfm, first and foremost.
* Debian can live with it if users are required to rename the derived
  Work that results from their modifications to cmr10.tfm.
* Users must have the freedom to rename cmr10.tfm as a file if they
  wish, irrespective of whether or not they have changed it.
* Debian tolerates the compelled inclusion of notices in human-readable
  form that a Work has been modified from its original state.  These
  notices can take many possible forms.  A license can *encourage* the
  inclusion of notices in as clear or as obfuscated a manner as it
  wishes, however the license must be able to be satisfied with a notice
  that is available on inspection of the modified work.  The GNU GPL
  contains an example of what the Debian Project considers reasonable.
* Users must have the freedom to create completely original works, and
  use *filenames* for these works that, coincidentally or not, happen to
  be same as the names of files that an already-existing licensed work
* Note that the title of a Work -- that is, its name -- is not itself
  subject to copyright protection in the United States and many other
  countries[1].  To effectively restrict authors' rights to name an
  original work the same as an existing work, trademarks must applied
  for in the jurisdictions of interest.  Debian tolerates -- via DFSG 4
  -- trademark practices, but only insofar as they refer to the naming
  or versioning of a work, and not to any other application.  Any other
  restrictive application of trademark protection can result in a DFSG
  violation and Debian will take whatever steps it deems necessary to
  protect itself and its users from legal liability or harassment from
  licensors, just as we have done with software patents.  (This means
  that sometimes a DFSG-free copyright license isn't enough to get your
  software into main, if you or some third party is using patent or
  trademark law to frustrate the exercise of rights that the DFSG
  attempts to guarantee.)

> 3. If (2) is true, can we model LPPL to reflect the same behavior?
> I hope we can find the middle ground along these lines. 

Me too!

[1] For example, both Led Zeppelin and the Rolling Stones wrote,
recorded, and released songs called "Good Times Bad Times".  They are
not the same song, and not even similar except insofar as they are part
of the same genre of music and were recorded using similar instruments.
To the best of my knowledge, neither group ever ever sued the over this
song title, or had a cause of action to do so.

G. Branden Robinson                |     One man's "magic" is another man's
Debian GNU/Linux                   |     engineering.  "Supernatural" is a
branden@debian.org                 |     null word.
http://people.debian.org/~branden/ |     -- Robert Heinlein

Attachment: pgpAJFtTVeEXD.pgp
Description: PGP signature

Reply to: