Re: [tlug] Open source license (wikipedia)

["Stephen J. Turnbull" <turnbull.stephen.fw@example.com>]

Benjamin Kowarsch writes:

 > It seems you are taking the term "reconstructing the original" too
 > literally.

What could it possibly mean but producing a copy that might be fuzzy
(as you describe with different names, passages elided or resequenced,
and so on) but recognizably the "same thing" as the original?  My
claim is that the process of copying that is the fundamental activity
restricted by copyright is *more*, not *less*, general than that.

 > Consequently, if you had a collection of analytics of a text that
 > would permit reconstructing the original text in such a way that
 > the reconstruction would constitute plagiarism, then this falls
 > under "reconstructing the original".

Plagiarism is a different concept from copyright infringement, more
closely related to patent than to copyright.  Specifically, plagiarism
is theft of *content* regardless of how expressed, while copyright
infringement is theft of *expression*, regardless of the content (thus
the claim about "Joycean 1-grams").  Of course in practice the overlap
is quite large.  E.g., in the typical undergraduate "kope-pe" form,
plagiarism is also a copyright violation.  Nevertheless, the two sins
are conceptually distinct.

But regarding this particular description of "reconstruction", I did
not say otherwise and I'm not sure why you think I did.  The
optimizing compiler example was intended to illuminate a different
point.  I.e., that it is not sufficient to show that reconstruction is
impossible to refute a claim of copyright infringement.

Steve