[TUHS] Why Linux not another PC/UNIX [was Mach for i386 ...]

Random832 random832 at fastmail.com
Fri Feb 24 06:31:04 AEST 2017

On Thu, Feb 23, 2017, at 14:15, Clem Cole wrote:
> On Wed, Feb 22, 2017 at 6:51 PM, Paul Ruizendaal <pnr at planet.nl> wrote:
> > I'm not a lawyer, but wasn't part of the background that prior to 1988
> > in US law one could not claim both copyright and trade secret protection,
> > and that for something to be copyrighted it had to expressly claim to
> > be copyrighted material, and be registered as such?
> >
> ​Take this with what its worth (it came for free and I'm not a lawyer
> ....) > Your comment got me thinking, why would try to change and can you.  So I
> asked on our patent counsel this am to explain the difference.  For
> context in the USA we have Patent, Trade Secret, Copyright Registration and
> Copyright Protection.   Her reply to me was:
> Copyright *protection* is automatic – as soon as the code is written it
> is
> considered protected.  Copyright *registration* is just a formality
> necessary to instigate litigation.  There is no time limit for
> registration.

That's true today, but to my understanding wasn't true in 1988. (Well,
registration wasn't a requirement to be copyrighted - that requirement
went away retroactively in 1978, and only applied to unpublished works
then.) The change seems to have been March 1, 1989 from what I can find.

I think AT&T *tried* to construct a basis to claim that UNIX source code
was "unpublished", even when distributed to source licensees (or e.g.
shell scripts which were by necessity distributed to all licensees),
possibly in service of this trade secret theory. Remember, the infamous
comment on the otherwise empty SVR2 /bin/true had two lines of copyright
notice and three lines of assertion that it was unpublished.

And if that attempt involved removing all copyright notices from V6, V7,
and 32V (and presumably not registering any copyright on any
"unpublished" works pre-1978) then that might have killed any
copyright-based case. By the time the actual lawsuit happened, they were
fully committed to the trade secret theory.

> Dennis Ritchie wrote:
> >
> > "Paul" <pssawyer at comcast.net.INVALID> wrote in message >>
> >   ....
> >> ISTR there was a copyright notice in a Sys V /bin/true; explaining
> >> why we thought this was funny might be a violation of that copyright!
> >
> > The local lawyers here also flip-flopped over the copyright issue.
> > I can't recall whether it was because of law changes or reinterpretation
> > or whether they just changed their minds.  The issue partly had to do
> > with the question of whether a copyright claim amounted to publication,
> > while their primary protection theory had to do with trade secret
> > protection, and its possible conflict with publication.
> >
> > At any rate, here is the whole contents of /bin/true in SVr2:
> >
> > # Copyright (c) 1984 AT&T
> > #   All Rights Reserved
> >
> > # The copyright notice above does not evidence any
> > # actual or intended publication of such source code.
> >
> > #ident "@(#)true:true.sh 1.4"

More information about the TUHS mailing list