3.731 copyright, cont. (103)

Willard McCarty (MCCARTY@vm.epas.utoronto.ca)
Fri, 10 Nov 89 22:10:32 EST

Humanist Discussion Group, Vol. 3, No. 731. Friday, 10 Nov 1989.

(1) Date: Fri, 10 Nov 89 13:45:57 CST (40 lines)
From: "Michael S. Hart" <HART@UIUCVME>
Subject: Re: 3.725 copyright meditations (188)

(2) Date: Fri, 10 Nov 89 16:03 EST (40 lines)
Subject: Copyright

(1) --------------------------------------------------------------------
Date: Fri, 10 Nov 89 13:45:57 CST
From: "Michael S. Hart" <HART@UIUCVME>
Subject: Re: 3.725 copyright meditations (188)

The several notes regarding copyright have left out what is perhaps the
most important aspect of copyright, when copyright protection expires,
as well as the point of what actually constitutes a copy. The law is
quite specific in regard to both matters. A work can only be
copyrighted under certain rules, one of which is that the work must
contain the results of an intellectual or artistic effort. Therefore,
no matter how many years how many people spend, or pay others to spend,
creating a copy, per se, does not endow the creators with copyrightable
protection, unless the copy has artistic merit. This can not apply in
the case of typing in a copy. However a certain typesetting in a
certain font, etc. COULD BE copyrighted on the basis of artistic merit
but probably not in the case of using widespread computerized fonts to
copy text already in the public domain. With modern scanners, now no
less inexpensive than an XT-type computer, most books can be scanned in
less than a week - to effectively eliminate the labor value of the
person who typed a book for ten years (I would like to know what book it
was, and was it just copying or was the author actually composing -
which makes it copyrightable on that merit - and how many megabytes did
the finished work take? This is interested to me in the extreme, as I
spent years typing in books for our electronic library, before scanners
came along and reduced the 20 months of labor I spent on one 3.3M book
to about one month, including proofreading). At any rate, one of the
largest issues under discussion is the desire to get copyright
protection for works which are in the public domain and have been-
translated?!? into machine readable texts. Even if such a work were
allowed copyright protection for artistic merit, all one would have to
do to create? a newly copyrightable copy would be to change the font.
There was a similar issue once created in respect to WordCrucher(TM) in
which it was argued that a text prepared with WordCruncher was protected
under WordCruncher's program copyright. Needless to say, this did not
take much effort (though the usual legal delay) to solve, as this
precedent would allow any word processor that same right to demand a
royalty for any file created with it.

Enough for now. I would like to hear why so many are against having
machine readable texts of public domain materials easily and cheaply

(2) --------------------------------------------------------------47----
Date: Fri, 10 Nov 89 16:03 EST
Subject: Copyright

The current discussion of copyright has raised the question
of whether an electronic text may itself be copyrighted. In
the United States, the answer is probably no. According to
the guidelines published by the Library of Congress (specifically
Circular 65: Copyright Registration for Automated Databases),
<quote> Copyright protection is not available for . . . the
selection and ordering of data in a database [including full-
text database] where the collection and arrangement of the
material is a mechanical task only, and represents no original
authorship; e.g., merely transferring data from hard copy to
computer storage.</quote>

I assume that merely transferring data applies to text that
has been scanned with an OCR or keyboarded at a wordprocessor.
However, the introduction of markup language may qualify as
original authorship -- because it may constitute an act of
interpretation as well as manual labor -- and therefore may
entitle the developers of such an encoded text to claim
copyright protection of their own. Such an inference, it
should be noted, is based only upon the preceding quotation
and not upon any other details in Circular 65.

Incidentally, the Library of Congress has compiled dozens of
circulars on the subject of copyright (including one on the
implications of the United States' participation in the
Berne Convention as of March 1989). For a list of material
published by the Copyright Office, write for Circular 2:
Publications on Copyright. The address is Copyright Office,
LM455, Library of Congress, Washington, DC 20559. Phone:
(202)479-0700 between 8:30 am and 5:00pm Eastern Standard Time.

Mike Neuman
Georgetown Center for Text and Technology
Georgetown University
Washington, DC 20057