Date   : Thu, 12 Apr 1984 00:58:43-MST
From   : Ron Fowler <RFOWLER@SIMTEL20.ARPA>
Subject: INFO-COPYRIGHT AM DIGEST: APRIL 12, 1984

[ copyright (c) 1984 Ronald G. Fowler ]

There has been some controversy lately regarding the distribution of
public domain software with respect to the copyright law.  I've done
some preliminary research, and thought I'd share my findings with
the group.

Specifically in question has been Irv Hoff's copyright of his MDM
modem series, especially as it relates to Ward Christensen's orig-
inal work and Mark Zeiger's extensive enhancements to the program.
Prevention of "profit-taking" and sale of public-domain software
has also been discussed.

Fundamental to securing copyright protecton is the publication of a 
work; generally you may copyright unpublished work without restrict-
ion (i.e., you don't have to maintain a copyright notice in the work).
Legally, 'publishing' is the "distribution of copies ... to the pub-
lic, by sale or other transfer of ownership, or by rental, lease or
lending".  I think we can safely say that work distributed by SIG/M
and the CPMUG can be deemed legally "published".

Both the old (1909) copyright law and the new (1978) require a that
a copyright notice be placed in the work, in order to secure the copy-
ight.  While the newer law addresses procedures for omission of this
notice (in order that a mistaken omission not cause loss of copyright
protection) the older law did not. Work published prior to 1 January,
1978 falls under the jurisdiction of the older law.  In fact, Copyright
Office Circular "R1" specifically mentions this circumstance:

       "If a work was published under the copyright owner's 
       authority before January 1, 1978, without a proper
       copyright notice, all copyright protection for that
       work was permanently lost in the United States.  The
       new copyright law does not provide retroactive pro-
       tection for those works."

Now MODEM2 was published by CPMUG, without copyright notice, in
1977.  So much for any claim poor Ward might have.

MODEM2 is therefore unquestionably in the public domain.

(Note that the terms "copyright" and "public domain" are mutually
exclusive, under the law.  "Public domain" is work without copyright,
either by expiration of an existing copyright, or by forfeiture of
copyright by the author.  Hence, a disclaimer like "Copyright (c)
1984 by Calvin C. Codehacker: contributed to the public domain, may
not be sold commercially" is a contradiction in terms, and may very
well jeopardize the author's right to copyright.  For that reason,
I *never* refer to the "public domain" in any programs I introduce
to the user community for which I desire to retain copyright pro-
tection).

(Side note: there is absolutely *nothing* to prevent a commercial in-
terest from selling public domain work, legally or ethically.  Have
you ever seen an anthology of Edgar Allen Poe, or a recently reprinted
"Moby Dick", both of which have fallen into the public domain?  Has
sale of these caused any kind of public outcry of "ripoff!"?  Think
about it...)

Back to MODEM2 and MODEM7: Zeiger and Hoff's enhancements have been
published (SIG/M, I believe) after 1 January, 1978, and thus fall under
the jusrisdiction of the 1978 law, which provides for "derivative work".
A "derivative work" is "a work based on one or more preexisting works".
This seems to be subject to copyrights held by the author of the
previous work, but is not spelled out specifically in the material
I've seen so far.  I'd hazard to guess that Hoff and Zeiger are on
pretty stable ground, though, *unless* either has failed to actually
secure the copyright by registration.  That's where things begin to
get a little shaky.

Generally, you don't have to register to claim copyright; you can reg-
ister anytime within five years of publication and still establish
prima facie evidence in court of the copyright's validatity. There
is a "gotcha" though: if the work is not registered within 3 months
after publication, then no statutory damages or attorney's fees are
available to the plaintiff in an infringement suit.  In this case,
only actual damages and lost profits are available, and in "free"
distribution software, there are very little damages indeed (perhaps
if the author is a professional who receives value from the circulation
of his name in "free" software, the loss of that circulation could
constitute a damage....).

Now I think we have a clear idea of how to proceed with new "freeware":

1) Claim copyright in the published work (copyright circular "R61"
   suggests placing the notice in the program's sign-on message,
   if it has one. By the way, this pub is specific to computer software).

2) Do not fail to actually register the copyright with the Copyright
   office, within 3 month of publication.  

3) Never mention the public domain in relation to your work.

With these requirements met, the software is protected to the extent
that you can confidently proceed legally against infringers, including
attorney's fees and statutory damages. Merely by virtue of your author-
ship (and, of course, your copyright).


Interesting aside: if you publish work with the Copyright notice
included, you *must* deposit a copy with the Copyright office for the
use of the Library of Congress; failure to do so can result in fines
and other penalties.  (Note that there are certain exceptions and
modifications to this rule for various types of copyright works).

Also: Forms are available from the U.S. Government Copyright Office:
dial (202) 287-9100.  Ask for form TX and Circulars R1 and R61.  The
cheapskates will only send you five TX's for one phone call.