[WikiEN-l] RE: copyright and NOR policies re: math and science

[Brown, Darin]

> > I want to write an article which relies heavily on scientific
> > publications. (I will of course cite the relevant articles at the end
> > of `my' wikipedia article). What I have in mind is a summary of their
> > work, maybe one of ten sentences copied literally. Do I need the
> > permissions of the authors of these article (one author is already
> > dead) or could I avoid that problem in simply not copying even a
> > single sentence verbatim?
> >
> >
> > Or without their explicit permission I simply could not write about
> > the subject?
> >
> 
> You may quote these authors without their permission, to some extent.
> >From [[WP:Wikipedia:Fair use]];
> 
> "Brief, attributed quotations of copyrighted text may be used to
> illustrate a point, establish context, or attribute a point of view or
> idea. In general, extensive quotation of copyrighted news materials
> (such as newspapers and wire services) is not fair use and is
> prohibited by Wikipedia policy. Extensive quotation from copyrighted
> media such as movie scripts is also prohibited, as previous "fair use"
> case law has established that such usage may infringe on the future
> earnings of the copyright holder (i.e. on their ability to publish a
> book of said quotations)."
> 
> In general, try not to use copyrighted work under fair use unless
> really necessary. You can present the same facts as a copyrighted text
> using your own words without violating copyrights. It's the creative
> work that is protected, not the facts.  Also, remember that Wikipedia
> requires attribution whenever fair use is used.

But in many fields, esp. mathematics and some scientific fields, the
distinction between "facts" and "creative work" or "ideas" isn't so
clear-cut.  A previous poster seemed to imply, if I understand correctly,
that one cannot write, e.g. "E = mc^2" without attributing fair use to
Einstein.  This seems ridiculous to me.  In mathematics, there often is no
way to "paraphrase" a result, so that you are presenting "the facts" but not
the "creative work".  For example, suppose someone in analytic number theory
proves yet another technically tedious result on the zeta function.  Just
imagine an expression full of integrals and special functions that goes on
for an entire line (or more).  By the above interpretation, how is one to
present this result without violating copyright?  After all, there is no way
to "paraphrase" the equation.  And what about a proof?  One can always
paraphrase a proof, but at the end of the day, you're really violating
copyright in spirit under the above interpretation as much as if you copied
verbatim.

The math community generally approaches things from a different angle -- the
issue is not viewed as a copyright issue (no one is making tons of money off
their research papers) so much as a research, peer review, and intellectual
respect issue.  If someone uses another's result in their paper, you must
reference them -- not so much for copyright as for rigour and peer review.
If you are using someone's proof in a textbook, it's generally good to
attribute it, but this is a matter of subjectivity -- most proofs that are a
notch or two below research-level have saturated the collective
consciousness (at least among some handful of people), and no attribution is
given at all.  Even on relatively recent topics which are no more than 50
years old, (within the span of copyright) a typical textbook has few direct
references, except when a result or argument is sufficiently novel or
original.  No one is running around getting express "permission" from anyone
whose work they want to quote.  Providing references to every single
statement made would just be ridiculous.

This brings me to another point that I don't think was addressed when I
brought it up -- many people seem to think that NOR means that every single
mathematical argument in wikipedia requires a reference to the published
literature.  Again, this is ridiculous.  Some arguments are simply not
special enough to warrant publication.  We know how to integrate, take
limits, find upper bounds, play with inequalities, etc.  We don't have to
publish everything like this to know it's right.  If an argument can be
easily verified to be true _by a professional in the field_, it should be
acceptable at wikipedia, without reference to the literature.

darin