Discussion List Archives

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

Re: Trade and Service Marks

  • To: Multiple recipients of list <comcifs-l@iucr.org>
  • Subject: Re: Trade and Service Marks
  • From: Brian McMahon <bm@iucr.org>
  • Date: Mon, 17 Jan 2000 10:33:12 GMT
The case of "TeX" may be useful in providing analogies or precedents for what
we want to do with "CIF". Donald Knuth, the author of the program TeX, states
explicitly that users may modify the source code of the program if they so
desire; but the resulting modified program may not be called "TeX" unless it
passes a particular diagnostic test (called "TRIP") that he supplies. Hence,
the C port of TeX (which was originally written in a Pascal preprocessor
language) may be named TeX, because it produces identical output to the
original software in the "TRIP" test; but a modification that produces PDF
files as output is not permitted to be so called. However, such a modified
program exists, and is known as pdftex (indeed, possibly as pdfTeX).
Knuth has no problems with this.

The American Mathematical Society has registered as trademarks the names TeX
and AMS-TeX, the former on Knuth's behalf; the latter (referring to a
particular implementation of TeX upon which is layered a macro package
written by AMS) for its own use.

In practice I have seen software packages with names such as 
which may or may not pass the TRIP test; where they do, the package authors
usually declare proudly in their documentation that they are genuine TeX
implementations (usually just re-packaged for a different operating system)
and include the TRIP test file in the package; or they clearly state that
they are NOT compliant TeX implementations, and in what way they fail to
comply. In practice, however, such programs usually are compatible with the
canonical TeX software to a very high degree.

On the other hand, typesetting systems such as
   ArborText Publisher
do NOT claim to be "TeX" (sometimes make no reference to that program), even
though at least some of their formatting engine must be based directly on
the original TeX source.

So I would certainly favour registering "CIF" as the dundamental term; and
ideally providing some benchmark by which a file claiming to be CIF could be
adequately certified. Formal registration of other devices such as mmCIF,
pdCIF, coreCIF may not be necessary if it is expected that the community
would handle the suffixes with the same measure of respect as in the TeX
case. No doubt that respect arises in part from the trademark protection of
the name "TeX", but also from the very clear licence under which Knuth
released the software.