Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.1 6/24/83; site mit-athena.UUCP
Path: utzoo!watmath!clyde!burl!ulysses!allegra!mit-eddie!mit-athena!martillo
From: martillo@mit-athena.UUCP (Joaquim Martillo)
Newsgroups: net.emacs
Subject: Gnu Emacs & Copyright Laws
Message-ID: <287@mit-athena.UUCP>
Date: Mon, 15-Jul-85 12:08:39 EDT
Article-I.D.: mit-athe.287
Posted: Mon Jul 15 12:08:39 1985
Date-Received: Wed, 17-Jul-85 06:27:29 EDT
Reply-To: martillo@mit-athena.UUCP (Joaquim Martillo)
Distribution: net
Organization: MIT Project Athena
Lines: 80
Unlike RMS I am not driven by ideological fervor to take a stand
against "software hoarding." I was willing to "prostitute" myself
and work for ATT Bell Laboratories. But I learned Unix in the days
when the government forbade ATT to sell Unix as a product. The
material which came out of the Computer Research Group then was much
more interesting than the substandard Unix called 5V2. The computer
research software was interesting because the research group was able
to give and receive ideas from the university environment. Now ATT
is so paranoic about protecting "trade secrets" that they NIH
(NDH)'ed almost every interesting idea in the system.
Xerox's paranoia basically kept almost everyone from learning about
some of the interesting new computer ideas developed in their
research environment and in the end this paranoic concealment of
ideas is harming Xerox.
RMS could easily be correct that the current system of industrial
software development hinders both research and harms business. The
justification that someone must pay for the development environment
is incorrect because hardware is worthless without good software.
Therefore the hardware manufactures have an interest in seeing good
new software freely available.
The current situation needs people to take strong stands against
ridiculously rigid interpretations of copyright laws. If I compose a
piece of written material, the convention is I can use three
consecutive words from someone else's writings unattributed. In
scholarly literature, properly attributed quotation is considered
acceptable up to about 20% of the composition. A ridiculously rigid
interpretation of the copyright laws could easily require an author
to get permission for every single quote, but this is silly and
everyone knows it. Software copyright law is still relatively new
and no one knows the commonsense measure. This lack of commonsense
measure is being used to scare people.
I would suggest 1 pages of unattributed code (of generic functions
like strcpy) in a software project of at least moderate size is
acceptable. In a similar project, complaining about 10% or less
attributed code is silly. Now the disputed modules in Gnu Emacs
constitue about 5% of the code and I suspect less than 10% of the
original version of Gosling's emacs. Further of the modules which
contain Gosling's copyright, Trm.h (a 3 page module) has about 50%
(by line excluding comments) of its code in common with the current
version of Unipress Emacs. The difference between the current
version of Unipress Emacs and Gnu Emacs dsp.c (a 5 page module) is
90%. The differences between current version of Unipress Emacs and
Gnu Emacs display.c (a 19 page module) is about 80%. For all the
modules which Fen LeBalme gave RMS permission to use, the differences
are similar. Unipress is not even using the disputed software
anymore! Now, these modules contain code people like Chris Torek and
others contributed when Gosling's emacs was in the public domain. I
must wonder whether these people would have contributed had they
known their freely-given code was going to become part of someone's
product.
The use of this freely-given code in the product seems rather
questionable if this code does not remain freely available for others
to use even despite the copyright notice. Trying to forbid RMS from
using discarded code so that he must spend time to reinvent the wheel
supports his contention that "software hoarders" are slowing down
progress in computer science.
If using the discarded code is wrong, I must wonder why display.c
(p. 10) contains the following notice.
/****************************************************************************\
* *
* The following hack and all following lines involving the variable *
* "DoneEarly" cause the bottom line of the screen to be redisplayed before *
* any others if it has changed and it would be redrawn in-place. This is *
* purely for Emacs, people using this package for other things might want to *
* lobotomize this section. *
* *
\****************************************************************************/
Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.1 6/24/83; site pegasus.UUCP
Path: utzoo!watmath!clyde!burl!ulysses!mhuxr!mhuxt!houxm!mtuxo!pegasus!mzal
From: mzal@pegasus.UUCP (Mike Zaleski)
Newsgroups: net.emacs
Subject: Re: Gnu Emacs & Copyright Laws
Message-ID: <2432@pegasus.UUCP>
Date: Thu, 18-Jul-85 03:18:29 EDT
Article-I.D.: pegasus.2432
Posted: Thu Jul 18 03:18:29 1985
Date-Received: Fri, 19-Jul-85 02:02:33 EDT
Organization: AT&T Information Systems, Lincroft NJ
Lines: 63
[ Insert your favorit clever quote here. ]
Excerpt from: martillo@mit-athena.UUCP (Joaquim Martillo)
<allegra!mit-eddie!mit-athena!martillo>
Software copyright law is still relatively new and no one knows
the commonsense measure. This lack of commonsense measure is
being used to scare people.
I would suggest 1 pages of unattributed code (of generic functions
like strcpy) in a software project of at least moderate size is
acceptable. In a similar project, complaining about 10% or less
attributed code is silly.
A tangential point here which might be of interest. When depositing
code with the Copyright Office, one is required only to submit:
"... one copy of identifying portions of the program ... 'identifying
portions' shall mean either the first and last 25 pages of the
program ... together with the page or equivalent unit containing the
copyright notice, if any." [1]
Also, use of generic functions does not necessarily constitute copyright
violation or preclude copyrighting. After submiting a work for
copyright, if the person submiting the forms has claimed exclusive
authorship, a form letter which basically asks:
[To what extent does the deposited program contain preexisting
materials? Your application implies ...] "... that the program
is entirely new and contains no SUBSTANTIAL (my emphasis) amount of
material that has been previously registered or published or that
has become standard usage for certain purposes. Although some
programmers crate from scratch all parts of their programs, we
understand that many draw from libraries or previously developed
programs or routines in creating a new program." [1]
A general comment: Many of the submissions "explaining" the law
regarding copyrights, patents, and trade secret which I see on
the net seem to have many misconceptions. (I do not mean to
imply anything negative about the above excerpted article, as it
expresses wishes, rather than claims things are facts.) It seems
to me that some of these people should do some reading before they
start writing. Herewith, some suggestions:
[1] "A Software Law Primer", Frederic W. Neitzke, 1984, Van
Nostrand Reinhold, page 52.
[2] "Law and the Computer", Michael C. Gemignani, 1981, CBI
Publishing.
A quick quiz: If I put:
/* great.c - Great C program
* (C) 1984 by Michael Zaleski */
at the beginning of my great C program, have I preserved my
legal copyright rights?
The surprising answer is NO! See page 50 of [1].
-- "The Model Citizen" Mike^Z
Zaleski@Rutgers [ allegra!, ihnp4! ] pegasus!mzal
Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.2 9/18/84; site masscomp.UUCP
Path: utzoo!watmath!clyde!bonnie!masscomp!z
From: z@masscomp.UUCP (Steve Zimmerman)
Newsgroups: net.emacs
Subject: Re: Gnu Emacs & Copyright Laws
Message-ID: <736@masscomp.UUCP>
Date: Fri, 19-Jul-85 09:16:27 EDT
Article-I.D.: masscomp.736
Posted: Fri Jul 19 09:16:27 1985
Date-Received: Sat, 20-Jul-85 05:48:15 EDT
References: <2432@pegasus.UUCP>
Organization: Masscomp - Westford, MA
Lines: 39
> Also, use of generic functions does not necessarily constitute copyright
> violation or preclude copyrighting.
Although the second part of this statement is true, I think you're
making some false assumptions in the first part. As supporting
evidence, you quote:
> [To what extent does the deposited program contain preexisting
> materials? Your application implies ...] "... that the program
> is entirely new and contains no SUBSTANTIAL (my emphasis) amount of
> material that has been previously registered or published or that
> has become standard usage for certain purposes. Although some
> programmers create from scratch all parts of their programs, we
> understand that many draw from libraries or previously developed
> programs or routines in creating a new program." [1]
This certainly corresponds to my understanding of the copyright law.
However, if you read it (and the copyright law itself) carefully, it is
clear that this is merely stating the eligibility requirements for
copyrighting a program. That is, a program must be substantially
original in order to be eligible for copyright, but use of library
routines and similar programs (which may or may not be copyrighted) does
not necessarily detract from that originality. I think a close reading
of these texts will show that they do not say that it is all right to
use copyrighted subroutines or similar code in a new program without
permission, and that the proper use of such code implies that permission
has been received.
I liked your example of the invalid copyright notice. The problem, of
course, is that the international copyright symbol is a C with a circle
around it, and the parentheses don't quite qualify. Adding the word
"Copyright" is sufficient protection under U.S. law. However, this word
does not satisfy international copyright law, which accepts only the
circle C. This leads to the interesting situation that there is no way
no put an internationally valid copyright notice into an ASCII source
file.
Steve Zimmerman
Masscomp