*BSD News Article 3124


Return to BSD News archive

Path: sserve!manuel!munnari.oz.au!spool.mu.edu!hri.com!ukma!darwin.sura.net!mips!sdd.hp.com!think.com!news!rlk
From: rlk@underprize.think.com (Robert Krawitz)
Newsgroups: alt.suit.att-bsdi,comp.unix.bsd
Subject: Re: UNIGRAM's article on the USL-BSDI suit
Message-ID: <RLK.92Aug5104856@underprize.think.com>
Date: 5 Aug 92 14:48:56 GMT
Article-I.D.: underpri.RLK.92Aug5104856
References: <1992Aug3.143259.23897@crd.ge.com> <7045@skye.ed.ac.uk>
	<KANDALL.92Aug4161214@globalize.nsg.sgi.com>
	<15lc26INNlpq@agate.berkeley.edu>
	<KANDALL.92Aug5143839@globalize.nsg.sgi.com>
Organization: Thinking Machines Corporation, Cambridge Mass., USA
Lines: 71
NNTP-Posting-Host: underprize.think.com
In-reply-to: kandall@nsg.sgi.com's message of 5 Aug 92 19:38:39 GMT

(note that I'm speaking only for myself)

In article <KANDALL.92Aug5143839@globalize.nsg.sgi.com> kandall@nsg.sgi.com (Michael Kandall) writes:

   >>>>> On 4 Aug 92 07:36:38 GMT, gwh@soda.berkeley.edu (George William Herbert) said:
   }> What?
   }> I understand some parts of how UNIX internals work.
   }> I have not, to my knowledge, looked at encumbered code.
   }> Rather, I've read articles and books and talked to people
   }> about how it works.  Every little detail of the UNIX operating
   }> system has been disected time and time again in journals,
   }> books, and online and offline discussions.

   Fortunately or unfortunately, in order to create a successful software
   system, one must reveal as much as possible about one's product.  The
   more people know about your implementation, the better they can use
   it, and the more successful the product will become.  If nobody knew
   anything about how UNIX works, it would not have much of a chance of
   being a commercial success.  

That's life.  When you introduce a new product, and it's successful,
people compete with you.  So ATT/USL should be happy that it made the
right choice by making Unix reasonably open in the first place.

"The bulls can win, the bears can win, but the hogs can never win".

Mind you, Apple has done fairly well (unfortunately, in my personal
opinion) by being rather closed.  While I detest this practice, at least
they've been consistent since day 0.

   }> If that knowledge, gained without reading AT&T's proprietary
   }> code, is then used to recreate part of UNIX, is it violating their
   }> copyright?  That's NOT what current copyright law generally holds.
   }> That's not what current intellectual propertly law holds.

   As I mentioned, independent of the legality, this practice has a
   deleterious effect on open systems, discourages industry investment in
   creating licensable technology, and shows no respect for the value of
   the specification and design time invested by the original creators of
   the system.  That the legal system may no protect such things is 
   unfortunate.

The free market means you have the opportunity to sink as well as swim.
You are not guaranteed a return on your investment.  If someone outplays
you at your own game, that's part of life.  If I buy a house, and the
market drops, am I "entitled" to have its value increase when I want to
sell it because of the work I did saving for the down payment, putting
in a deck, remodeling the kitchen, etc.?

Mind you, CSRG/BSD invested considerable time and effort into improving
Unix, too.  Why should USL be able to take advantage, gratis, of CSRG
and BSD's work?

As for reverse engineering, redesigning, etc. hurting open systems, that
seems rather contrary to real world experience.  The major system
architectures that have the most market share in the low end and
workstation segments of the market -- the PC and Sparc -- have all been
cloned, improved, etc. by companies that show no respect for what IBM or
Sun did other than than the highest form of flattery of all --
imitation.  IBM and Sun have made out quite handsomely from this.  And
the GNU software is rapidly becoming the software of choice on many Unix
platforms (and some others) -- GNU Emacs is "the world" for many people,
GCC is one of the most solid and portable compilers in existence --
despite the FSF taking NO "intellectual property" claims on these
products.
--
ames >>>>>>>>>  |	Robert Krawitz <rlk@think.com>	245 First St.
bloom-beacon >  |think!rlk				Cambridge, MA  02142
harvard >>>>>>  .	Thinking Machines Corp.		(617)234-2116

Member of the League for Programming Freedom -- write league@prep.ai.mit.edu