Return to BSD News archive
Xref: sserve comp.org.eff.talk:9311 misc.int-property:567 comp.unix.bsd:6007 Path: sserve!manuel!munnari.oz.au!spool.mu.edu!nigel.msen.com!math.fu-berlin.de!Sirius.dfn.de!chx400!sicsun!disuns2!dgay From: dgay@di.epfl.ch (David Gay) Newsgroups: comp.org.eff.talk,misc.int-property,alt.suit.att-bsdi,comp.unix.bsd Subject: Re: Patents: What they are. What they aren't. Other factors. Message-ID: <DGAY.92Oct3124402@disun46.epfl.ch> Date: 3 Oct 92 11:44:02 GMT References: <10880.Sep3008.43.0892@virtualnews.nyu.edu> <1992Oct1.090209.9474@netcom.com> <1992Oct1.134749.5671@cae.prds.cdx.mot.com> <1992Oct1.230931.7833@netcom.com> Sender: news@disuns2.epfl.ch Followup-To: comp.org.eff.talk Organization: Ecole Polytechnique Federale de Lausanne Lines: 67 Nntp-Posting-Host: disun46.epfl.ch In-reply-to: mcgregor@netcom.com's message of 1 Oct 92 23:09:31 GMT In article <1992Oct1.230931.7833@netcom.com> mcgregor@netcom.com (Scott Mcgregor) writes: In article <1992Oct1.134749.5671@cae.prds.cdx.mot.com> dan@cae.prds.cdx.mot.com (Dan Breslau) writes: >Regardless of that problem, it's ridiculous to claim that the end >result of LZW is a rearrangement of the polarity of electrons. The >end result is the transformation of information, regardless of the >medium. This is the essence of what algorithms are. I've never disagreed that the end result of LZW is a rearrangement of information, not polarities of electrons. What I have said is that one could construct a physical process patent that described the transformation of polarities of electrons without appealing to mental states, or even without assigning the polarities significance of being information. It just focusses on the physical world changes. In such a patent, the fact that you are modifying the polarities is essential, because that is EXACTLY what you are claiming is the benefit. If you are comfortable with this not being a software patent then so am I. But if the inputs and outputs and other characteristics are defined correctly, a computer system that effectuated the changes as a result of running an LZW program (not the theoretical algorithm, but the actual electronic switching) could infringe. Certain mechanical cam systems would be equally infringing. So this wouldn't be "the LZW algorithm patent" (because algorithms aren't patentable) but it would be the "patent that restricts use of LZW to achieve increased disk storage packing densities. I agree with Mr. Margolin that such a patent shouldn't be extended to prohibit LZW application to other domains (such as modems) unless they too are explicity claimed (presumably by showing how one could convert polarities on media to control signals on a wire and vice versa). Even though this could be a "physical process patent" without any appeal to software algorithms, I suspect the effect on software use would be the same as if it were a "software patent" because the reason someone usually wants to use an algorithm is to achieve some real world result, and not just to do some theoretical mental world only transformations. In other words, you are using the existence of the computer as a physical element capable of following algorithms as a way of extending patents to cover algorithms (as now these algorithms can have a "physical" existence). This smacks of sophistry to me. If you feel that algorithms should be patentable, say so (and I can understand people having that point of view). I would argue that computers are basically information processing devices (*), and that the existence of the physical storage systems (and in this case, the polarities on the media) is inessential. LZW is obviously an information-transforming process, the fact that this information has to be physically stored at all points in time (be it in memory, on disk, etc) for it to be useful is beside the point. -- *: When the computers start controlling external devices, things change. -- Scott L. McGregor mcgregor@netcom.com President tel: 408-985-1824 Prescient Software, Inc. fax: 408-985-1936 3494 Yuba Avenue San Jose, CA 95117 David Gay dgay@di.epfl.ch Nothing much at all. Ecole Polytechnique Federale de Lausanne Switzerland