Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP Posting-Version: version B 2.10.2 9/18/84; site ut-sally.UUCP Path: utzoo!watmath!clyde!burl!ulysses!gatech!ut-sally!crandell From: crandell@ut-sally.UUCP (Jim Crandell) Newsgroups: net.music.synth,net.legal Subject: Re: Marble Madness & FM Music Synthesis Message-ID: <3027@ut-sally.UUCP> Date: Fri, 27-Sep-85 23:42:41 EDT Article-I.D.: ut-sally.3027 Posted: Fri Sep 27 23:42:41 1985 Date-Received: Sun, 29-Sep-85 07:24:11 EDT References: <2614@ihnss.UUCP> <267@weitek.UUCP> <2882@ut-sally.UUCP> <269@weitek.UUCP> <2953@ut-sally.UUCP> <710@sfmag.UUCP> Reply-To: crandell@sally.UUCP (Jim Crandell) Organization: U. Texas CS Dept., Austin, Texas Lines: 89 Xref: watmath net.music.synth:550 net.legal:2399 >> >Patenting FM synthesis is a lot like patenting the color blue to my mind. >> >I'm well aware that others were there first. They didn't get that patent >> >however. There exists a feature of patent law which makes an idea unpatentable >> >if "it is obvious to anyone skilled in the art." I think FM qualifys here. >> ... >> This seems to be a good example of a type of incident that one hears >> about now and then and which usually leaves me with a profound sinking >> feeling. I am assuming, of course, that Yamaha has a US patent, else >> much of this will seem rather silly.... [I love a self-fulfilling prophecy. But much as I admire my own flames, I really think everyone's seen enough copies of this, thanks partly to people who followup without bothering to edit the quote. :-) ] > As has already been amply mentioned, the patent was issued to >John Chowning and Stanford U.... The point I want to make, >however, is that it is not at all clear to me that the idea of using >FM for music synthesis was not patentable, at least at the time >the patent was issued. You're not alone. >There are four requirements for an invention to be patentable. It >must be in a class of inventions defined as patentable, it must be >useful, it must be novel, and it must be non-obvious. The first two >are straightforward in this case. Whoa, not so fast. Useful? No argument. ``In a class of inventions defined as patentable''? Not so easy. One of my correspondents made the bold concession that FM was a ``mathematical formula'' (I might have said ``theoretical concept''), and is therefore ineligible. Appealing as that line of reasoning may be, I reject it not only because it invites the question whether the application of the formula/concept is the new, patentable entity (FM itself demonstrably isn't), but because it's unnecessary. I have been familiar with software implementations of digital FM synthesis, some of which I've written, for over a dozen years. (No, that sentence doesn't contain a typo.) Actually, the time frame is irrelevant. The fact that FM synthesis is realizable via writing programs for ``general purpose'' computers implies that at least for some purposes, digital FM synthesis is really just a computational problem for which there exists a family of algorithms and is therefore not patentable. Neither are any of those algorithms, as I assume we all know. A practical consequence of the same phenomenon is that you might find that your market-cornering, one-of-a-kind FM synth chip suddenly has to contend with some competitor's custom ROM pattern in, say, a TMS32020. (Before you flame, please note that I haven't actually said that there IS a program for a 32020 that will emulate the Yamaha FM synth chip. Okay? I just want to be sure everyone's straight on that. [:-)]) Now it's quite true that certain legal eagles have been kicking around the notion of patent protection for programs for several years, but they haven't gotten very far with it. One of the reasons is a widespread agreement that patents are intended to cover fairly tangible things, like black boxes. For most legal purposes, software is considered intangible. But let's face it, the ``idea of using FM for music synthesis'' is about as intangible as you can get. > Novelty most likely also passes >because it is only concerned with whether others were publicly using >the invention or already had a patent on it. Another very ``iffy'' one, actually. Definition of ``publicly using'' is crucial. > The real question is non-obviousness, and I think that if you look >at what was going on at the time Chowning "invented" FM synthesis, you >will see that no one had done anything like it. Perhaps. I'll take you up on the challenge at any rate; I'm currently working to acquire a copy of the patent. In terms of the theroretical concepts involved -- well, we'll see. I have been told, however, that the actual scope of Chowning's patent is essentially limited to certain hardware implementations. Yes, that's hearsay. (And yes, this is USENET. Caveat lector.) > Even so, I agree that FM being >patented does put a crimp in the plans of those who would like to make >their own chip or otherwise capitalize on the technique. At least. Actually, it's potentially worse than that, because if company X has the exclusive legal right to design, make and sell all forms of any device that does Y, then if you need a device that does Y but has some little wrinkle which X happens to feel would render it unmarketable, then you're fresh out of luck. Make no mistake about it -- lots of people are injured when an overly broad patent is successfully granted; it's a matter that legitimately concerns all of us. -- Jim Crandell, C. S. Dept., The University of Texas at Austin {ihnp4,seismo,ctvax}!ut-sally!crandell Brought to you by Super Global Mega Corp .com