RE: [Gnu-arch-users] Front page to wiki now modifiable again

[Parker, Ron]

> From: Stephen J. Turnbull [mailto:address@hidden

> >>>>> "Mikhael" == Mikhael Goikhman <address@hidden> writes:
> 
>     Mikhael> I fail to understand why people use FDL as an example
>     Mikhael> against GPL v2+.  This is bogus.
> 
> The point is that software and its documentation should converge.  I
> think we all pay lip-service, at least, to the Book of Literate
> Programming by the Prophet Knuth.

[[I was going to keep my mouth shut until I had personally reviewed the DFSG
material, the OSI license information and all of the FSF licenses, but I did
have some independent thoughts related to this.  I cannot say whether I
believe the DFSG and the GNU FDL are, or should be made, compatible or not.
Please do not read a position on this into what is written below.  I am too
ignorant to take sides on the issue.  I only hope it can be resolved in
time, in a way satisfactory to both sides.]]

Knuth himself is an interesting example.  To my knowledge, he never claimed
to be a proponent of free-software.  Rather he has released and developed
documented-software and software-documentation.[1]  An example of the first
being tex.web and an example of the second being texbook.tex.  In this, his
intention and practice seem clear.  Firstly, that software should be well
documented, both internally and externally.  Secondly, that skilled manual
and book authors should have the protections afforded by copyright law to
support and sustain their work.

Whether it was his intention or not, the second point is backed up by U.S.
Constitution Article I, Section 8.  In part it reads, "The Congress shall
have power...To promote the progress of science and useful arts, by securing
for limited times to authors and inventors the exclusive right to their
respective writings and discoveries...."[2]  This section is the basis for
both U.S. copyright and patent law.  Now, there are no doubts in my mind
that the intention of this has been overstepped by both the Sonny Bono
Copyright Term Extension Act[3] and the DMCA[4].  

Yet, the point is that there are two equal sides to "promot[ing] the
progress of science and useful arts".  First, the authors and inventors need
to be able to support and sustain their work.  "[A] workman is worthy of his
hire."[5]   Second, the people have a right to profit from the knowledge and
artful contributions of such.  This leads to the third side of the triangle,
which is equally important.  Recognizing both of these things, the people
grant to the authors and inventors "exclusive rights" for a limited time.
At least such was the position of the founding fathers.  (I realize this is
an international mailing list and an international development effort, but
my understanding of copyright and its related licenses is restricted to U.S.
law.  I do not know the related deep history of such information in other
countries and cultures.  Although it might make an interesting study.)  If
any of the sides of this triangle become imbalanced, damage is done.  Either
the author and inventor are unable to support and sustain themselves, or the
people are unfairly robbed of the longer term benefits of the work done.  

Knuth seems to have found an interesting way to balance this triangle in his
own life and it lays in the separation of internal and external
documentation.  

The code and its internal documentation are themselves free for all to use
and study.  But, for the duration of D. E. Knuth's copyright only he may
modify the original source, cf. tex.web.  However, recognizing the need for
others to be able to modify and customize this software, he provided an
external mechanism for others to change it without changing the original.
This maintains both a reference standard and provides for extension and
improvement by others.  Also mechanical transformation of the program to
other languages, a la web2c, has become a standard and D.E.K. approved
practice.

In some ways, this reminds me of GNU arch.  There is an individually
maintained reference standard that all are free to use and modify
externally.  And like web2c one of those external version might become the
de facto standard implementation.

Knuth then appears to balance the rights of the people with his own right to
sustainability by creating external documentation that goes beyond
explaining how the code works and provides a user manual so that individuals
may become proficient in its use.  This is the source of (part of) his
revenue stream.[6]  Above and beyond what is required by copyright law, he
releases some of this software-documentation in its original source form so
that it may also be studied and learned from.  Yet at the start of the
texbook.tex source code the following is found:

        % This manual is copyright (C) 1984 by the American Mathematical
Society.
        % All rights are reserved!
        % The file is distributed only for people to see its examples of TeX
input,
        % not for use in the preparation of books like The TeXbook.
        % Permission for any other use of this file must be obtained in
writing
        % from the copyright holder and also from the publisher
(Addison-Wesley).
        \loop\iftrue
          \errmessage{This manual is copyrighted and should not be
TeXed}\repeat

In this, his or the AMS's intentions are clearly seen.  If I may paraphrase,
"Learn from this material but do not create competing books from its
source."  Also note the use of the word "should" in the error message, "This
manual is copyrighted and _should_ not be TeXed." (Emphasis mine.)  Legally
this expresses a preference, not a prohibition.  If re-TeXing was prohibited
it should read, "This manual is copyrighted and _shall_ not be TeXed."  It
is a subtle, but important difference.  When I was learning TeX, I modified
a local copy of texbook.tex so that I could view it online instead of having
to carry the book around.  I have also, in the past, modified it to skip the
dangerous sign sections.  Then later, I modified it to just skip the
double-dangerous sections.  This made it easier to read online and to read
it in its various stages, essentially folding the book to a more compact
form and then progressively unfolding it as I learned more and more.

What some have failed to recognize in this discussion is what has been
called "Fair Use Doctrine.  Its essence is captured in US Code Title 17,
Chapter 1, Section 107, "Limitations on exclusive rights: Fair use".[7]
According to U.S. law, as I understand it, Mikhael should not be prevented
from using portions "of the Wiki in projects licenced under canonical 'GNU
GPL; either version 2 of the License, or (at your option) any later
version'"[8], provided they adhere to Fair Use.

He also says, "I don't feel like continuing to contribute to something I
can't use in my own projects without licensing troubles."  However, it is
entirely possible for him to completely and fully reuse those portions which
he has authored.  He is ultimately the copyright holder unless he has
assigned those rights to someone else.  As copyright holder he is free to
release his contributions under multiple licenses.  This is common practice.
Many software authors release their work to the general public under GPL and
also make it available for inclusion in proprietary work under a separate
license.

I do not intend to take sides on the pros and cons of Knuth's or Stallman's
approaches to software and documentation.  Perhaps in the views of some, on
a completely mathematical level, they are internally inconsistent, drawing
artificial lines between documented-software and software-documentation.  I
have not studied enough and do not yet know enough to make that
determination for myself.  

I suppose my own opinions on the matter have been being forming over the
last several years.  Yet at this point in time, the compromises made by both
men look reasonable to me when laid alongside the U.S. Constitution and
basic copyright law.  Over time I expect my opinion on this will continue to
foment and change.  

At least Knuth seems to have produced a rather balanced triangle and in life
all things must be balanced.  I will need to sit back, learn more and
observe more to determine whether or not I believe RMS and the FSF have
arrived at a balanced solution.

Over the years I have helped author and revise materials that are
distributed, translated and used world wide.  The copyright holder of these
materials[9] has allowed them to be liberally copied, modified and
distributed, but has diligently protected them against abuse, theft,
misapplication and misrepresentation over the years.  This I believe is the
purpose of copyright and the proper use of license.  None of these materials
were developed for personal profit, neither were they required to support
those who worked upon them.  They were for the benefit of those into whose
hands they were given.

Likewise, I believe that software should be for the public good, but its
license must be defended against abuse.  Ultimately it is up to the author
to determine how he will support himself and his work.

It is interesting, this got me to thinking about software from the "writings
and discoveries" perspective.  Only in proprietary software and
software-embedded hardware does there exist a basic inability to see the
thing for what it is, to take it apart and learn how it works and for a
skilled individual to repair it independent of the original producer when it
breaks down.  This seems to me a basic violation of the principles of
Article I, Section 8 of the U.S. Constitution.  

I suppose RMS went through the same basic mental exercise at least two
decades ago.  While, I've benefited from, promoted and even contributed to
free software over the years.  I finally "get it."  I also understand the
position of the open-source movement.  But looking at what I have written, I
have to say that I disagree with allowing open-source to be included into
closed-source projects.  

Although, and this is may be a radical position.  I can see possible reasons
for including optional closed source components into an otherwise open
source project.  However, the example I am contemplating gets back to the
argument of "is documentation (or data) the same as software?"  From a legal
stand point, I suppose I have to say, "Not if the author has not released it
as such."  I may desire to treat it that way.  I may even wish it were so.
But if the holder of the copyright disagrees, I cannot treat it as such
until his copyright expires.  Only then, may I treat it as such and publish
an edited or expanded edition of his original work.  But, until such a time
I can only distribute it as he has allowed me to do so.

-
Footnotes:

[1]  At least in the views of some, these are different.

[2]
http://www.law.cornell.edu/constitution/constitution.articlei.html#section8.

[3] http://en.wikipedia.org/wiki/Sonny_Bono_Copyright_Term_Extension_Act.

[4] http://thomas.loc.gov/cgi-bin/query/z?c105:H.R.2281.ENR:. (Note the
trailing colon is part of the URL.)

[5] Luke 10:7, I Timothy 5:8.

[6] I fully recognize that the first few volumes of "The Art of Computer
Programming" have effectively supported him of their own accord.  While
TAoCP is not documented-software, it is no doubt documentation about
software.  The difference is why prepositions and adjectives both exist in
English.  Otherwise, a tangled ambiguity would ensue.

[7] http://www4.law.cornell.edu/uscode/17/107.html.

[8] http://mail.gnu.org/archive/html/gnu-arch-users/2004-03/msg00876.html.

[9] A 501c(3) non-profit corporation.