reduce-algebra/contrib/LuFy

Comments re rights to use and re-distribute changed
versions of TeX-based material.


Let's start with a remark that I will call the programs that I derive
here "LuFy". Specifically I will NOT call then TeX or anything that sounds
like that or that could cause potential confusion between them and their
origin. For those who have not already spotted it, LuFy is to TeX much what
HAL was to IBM (consider a Caesar cifer: just replace each letter by the next
one along in the alphabet!). And the stylized mixing of upper case puts the
upper case letter in the middle. The "proper" version of it that "\LuFy" will
generate will of course shift and kern the letters in a suitably
extreme manner so as to continue the respect and reference for TeX but still
to make it clear that this is not what my code is purporting to be.

ftp://ftp.cs.stanford.edu/pub/tex/README.KNUTH contains the text

"The files in subdirectories "dist" and "local" of this directory are
master files for TeX and METAFONT, maintained personally by Donald E. Knuth.
Nobody else is authorized to make any changes whatever to them!
If you modify the files for any purpose, you must give your
files a different name, so that installations of TeX throughout the world
will be 100% compatible when they use the official source files.

Please help preserve the integrity of TeX by reporting any violations of
these rules to the TeX User Group.

The dist/ directory contains the latest official sources for TeX,
METAFONT, Computer Modern, and accompanying TeXware/MFware.

The local/ directory contains the files Knuth uses at home to
customize the systems to his personal computer. It also contains special
macro files and semi-official fonts and software that you and others
may find useful, although this local collection is not maintained
as carefully as the officially distributed files are.

The CTAN archives put the dist/* files into directories systems/knuth/*
(except that font files go into fonts/cm/mf/*);
the local/* files are placed in in systems/knuth/local/*."

and in http://www.ctan.org/license/knuth you will find
"This software is copyright and you are explicitly granted a license which
gives you, the user of the software, legal permission to copy, distribute
and/or modify the software, so long as if you modify the software then it
carry a different name from the original software."

These two statements indicate that material derived from tex.web
may be re-used with very great freedom, and in particular that incorporating
parts of it within an otherwise BSD-licensed corpus of code should not
raise problems.

If, associated with this, some macro packages, fonts or other TeX-related
material is used then that will very often fall under the LaTeX Project
Public License (eg the GUST Fonts used the GUST Font License bu that is
based on the LPPL). Clause 10 of the LPPL reads
"10. a. A Derived Work may be distributed under a different license
provided that license itself honors the conditions listed in
Clause 6 above, in regard to the Work, though it does not have
to honor the rest of the conditions in this license.

b. If a Derived Work is distributed under a different license, that
Derived Work must provide sufficient documentation as part of
itself to allow each recipient of that Derived Work to honor the
restrictions in Clause 6 above, concerning changes from the Work."
and my concern here regards distribution under terms as close to the BSD
ones as feasible so that the complete composite Work incorporating some
of this material is available with as close to simple and consistent
license terms as possible.
Also note Clause 7 which reads
"7. If you are not the Current Maintainer of the Work, you may
distribute a Compiled Work generated from a Derived Work, as long as
the Derived Work is distributed to all recipients of the Compiled
Work, and as long as the conditions of Clause 6, above, are met with
regard to the Derived Work."

I expect that maintaining the Derived Work (ie the source code) in the
same publically available place as any Compiled Work (ie binary versions)
will satisfy this requirement.

So here is a recitation of clause 6 of the LPPL here to see how it can
be "honored":

6. If you are not the Current Maintainer of the Work, you may
distribute a Derived Work provided the following conditions are met
for every component of the Work unless that component clearly states
in the copyright notice that it is exempt from that condition. Only
the Current Maintainer is allowed to add such statements of exemption
to a component of the Work.

a. If a component of this Derived Work can be a direct replacement
for a component of the Work when that component is used with the
Base Interpreter, then, wherever this component of the Work
identifies itself to the user when used interactively with that
Base Interpreter, the replacement component of this Derived Work
clearly and unambiguously identifies itself as a modified version
of this component to the user when used interactively with that
Base Interpreter.

Since my (current) intent is to derive an embedded layout engine it
seeme improbable that I will end up with things that could be direct
replacement for part of a regular TeX installation. However just to be
careful I should arrange that it I make use of class files or other macro
packages that the changed versions always contain annotation that would
print a change-message if they ever got loaded into a native TeX. I should
do that even if I have not changed the individual file in any way at all,
bacase the LPPL considers distributing a subset of a full system to be
a modification. I do not see this as a problematic limitation.

b. Every component of the Derived Work contains prominent notices
detailing the nature of the changes to that component, or a
prominent reference to another file that is distributed as part
of the Derived Work and that contains a complete and accurate log
of the changes.

A "prominent reference to another file" seems best here! For small changes
(eg if I were to find a need to patch a macro file) this is not a big
issue. For code that has been transformed from one language to another and
then further re-purposed, modified and optimised the "complete and accurate
log of the changes" starts to look onerous. However scanning down I see that
I am relieved of this responsibility of the Derived Work is not a replacement!
So all seems comfortable, but as a matter of politeness it will still be
proper to include information about changes that have been made.

c. No information in the Derived Work implies that any persons,
including (but not limited to) the authors of the original version
of the Work, provide any support, including (but not limited to)
the reporting and handling of errors, to recipients of the
Derived Work unless those persons have stated explicitly that
they do provide such support for the Derived Work.

The BSD license has as one of its clauses an explicit disclaimer relating
to liability, and it will be easy to avoid suggesting that anybody
provides support when they have not explicitly offered to. I think that the
main sort of statement that this prohibits in the Derived Work would be
one that said "Since this Derived Work is just a modified version of the
Original one you can expect the Original Authors to respond to any bug
reports you wish to raise". Ha ha ha. Easy to avoid saying that!

d. You distribute at least one of the following with the Derived Work:

1. A complete, unmodified copy of the Work;
if your distribution of a modified component is made by
offering access to copy the modified component from a
designated place, then offering equivalent access to copy
the Work from the same or some similar place meets this
condition, even though third parties are not compelled to
copy the Work along with the modified component;

2. Information that is sufficient to obtain a complete,
unmodified copy of the Work.

A reference to the Comprehensive TeX Archive would satisfy d.2 easily,
and the wording of d.1 about "similar place" backs up my reading of the
reqirement that the (source version of a) Derived Work be distributed
to those who receive a Compiled Work can be satisfied in a similar manner.
But again the next paragraph of the LPPL tends to render this clause
inapplicable.





Later on the LPPL qualifies itself with a statement:

"Derived Works That Are Not Replacements
---------------------------------------

Several clauses of the LPPL specify means to provide reliability and
stability for the user community. They therefore concern themselves
with the case that a Derived Work is intended to be used as a
(compatible or incompatible) replacement of the original Work. If
this is not the case (e.g., if a few lines of code are reused for a
completely different task), then clauses 6b and 6d shall not apply."

So as a reminder of how simple the remaining conditions are:
6.a: If used with an original Base Interpreter changed components must
announce themselves when used interactively.
6.c: Must not give impression that anybody who has not explicitly
agreed to will provide support.

There is one more section in the LPPL that matters. This is the strong
recommendation that all files subject to LPPL be listed in a file
called manifest.txt so that it is clear just what is covered. Transcribing
any such file from the original source of any Work that is to be relied on
will satisfy this.