*BSD News Article 10826


Return to BSD News archive

Received: by minnie.vk1xwt.ampr.org with NNTP
	id AA768 ; Sun, 07 Feb 93 11:01:42 EST
Newsgroups: alt.suit.att-bsdi,comp.unix.bsd
Path: sserve!manuel.anu.edu.au!munnari.oz.au!sgiblab!zaphod.mps.ohio-state.edu!cs.utexas.edu!wupost!csisun!jburnes
From: jburnes@csisun.uucp (Jim Burnes)
Subject: Re: George William Herbert's Challenge - Part 4 (copyright & derived works)
Message-ID: <C1wyMs.1x4@csisun.uucp>
Organization: CompuSci, Inc
References: <1993Feb3.175211.13214@igor.tamri.com>
Date: Thu, 4 Feb 1993 07:40:03 GMT
Lines: 107

jbass@igor.tamri.com (John Bass) writes:
> In regard to derived works, standards from other parts of the publishing
> industries need to be examined. Take music for example.
> 
> If we have a well known piece, like "Jingle Bells" and someone comes along
> with a tune that sounds just like it ... IE has the same rhythm and melody
> ... then we without question would call it a rendition of "Jingle Bells",
> no matter how much the author claims it to be a new piece ... even if
> EVERY note, EVERY chord, and every other technical description is different
> from the original.
>

Speaking as a musician and computer scientist I don't think your
analogy holds water.  Music is not the same thing as an operating system.

Music is a type of emotional expression.  The only close analogy to
this would be the direct expression of code with the same exact structure.
The "changing the chords" analogy doesnt fly either, considering that
if someone were to change the chords and the notes in jingle bells
and keep the same rythym you most likely would have something that
wouldnt be jingle bells.

Computer programs are more like mechanical artifices.  If you kept the
code but just changed the variable names, then I think AT&T might have
a case.  On the other hand since people at CSRG have seen the algorithms,
ideas etc etc they cant be expected to just forget these things.  The
chance that someone intimately familiar with BSD kernel sources
would create something similar to AT&T's code would be high.  Not only
that but internal functionality might be intentionally similar to
avoid breaking the higher layers of BSD.

Now of course you could completely replace the BSD kernel with a 
radically different architecture (say like MACH), but then that was
not the goal.  They still wanted something very much like the
UNIX monolithic kernel.  Once their minds had been exposed to the
old AT&T kernel/methodology it might be very hard to think in
some other modality.  Much like an apprentice to a composer might
very well structure their symphonies similar to their master.

The IDEAS on how to boot can not be protected.  A very specific
implementation can be.  Now at what level you say "Thats too close
to the original ...your using the exact same call sequence, the same
variables, the same temporary variables, the same loop structure,
the same buffering mechanism" ...after about 10 of these you start
encroaching heavily on the specific expression.  I imagine that
would be up to a judge.

In any case I think youre passing judgement on nit picking details
like this is mean and spiteful.  In a system as large as BSD you
could probably find some code fragment that looks like AT&T.  I
think the main question is "Did Jolitz et al willfully distribute
unaltered AT&T code that contained proprietary algorithms that
will damage AT&T & USL."  I think the answer to that question is
no.  The real damage to AT&T is that guru-level UNIX engineers
rewrote what little was left of the orginal unix sources.  Do 
you actually believe that $5,000,000,000 worth of software was
re-written by  Jolitz et al?  I mean it was eventually going to
happen anyway.  Someone with intimate knowlege of unix would
essentially cut USL out of the deal by making something very
similar.  AT&T/USL looses because the made something that
was replaceable by a significant effort.

Jeez...Linux was thunk up and implemented in less than a year.
On the outside it looks a lot more like AT&T unix than anything
else.  BSD has made innumerable contributions to UNIX.  It has
made UNIX a viable commodity for the long term.  

In any case I think its a case of sour grapes.  You seem to be
tied closely to the SysV community and the PC/UNIX community
also.  Could it be that a widespread, free unix system might eat
into your funds?


 
> No matter how much the author whines, we presume that he heard it atleast
> once and his song was derived from the original melody. To play the song
> and with a striaght face call it original is plagiarizm.

The place where original music springs is different from where
software springs.  Composing original music is a unique emotional
expression not likely to be duplicated.  Software is almost
always an engineered artifice that builds stepwise on previous
innovations.  This music analogy doesnt work.  Besides the question
is not one of copying ideas, but of source code.

> 
> This is the context on how to apply the scrutiny of "methodology and
> algorithms, including the sequence of processes adopted by the programmer"
> to works that are suspected of derivation.
> 
>

Variations on a theme.  If students of Vivaldi wrote music that sounded
very much like Vivaldi then they would be composing in the manner of
Vivaldi.  I think epression is the keyword here.  If you express an
idea exactly like some other composer.  Indeed if you just lift a
particular expression then you are stealing.  If you have similar
phrasing, composition and other structure yeilding the same net
effect (bop jazz, baroque, pop-rock) then youre using ideas and not
expressions.

In any case it would be nice if Bill, Lynee or someone else with
intimate knowlege of the specifics of 386BSD/BSD386 would comment.

J. Burnes
jburnes@compusci.com