libdts patent issue?
This came up in the course of a curiosity-driven review of the patent status of various audio codecs, and also in a recent MPlayer thread. It would probably be wise not to wait for a cease-and-desist letter before quietly discontinuing the distribution of libdts. See http://developers.videolan.org/libdca.html and http://www.via.ecp.fr/via/ml/vlc-devel/2005-04/msg00230.html . Might want to give derivatives a heads-up too. I don't say I like it, but it's probably the wrong battlefield on which to fight the "software patent" wars. The patent (US #5,956,674, EP 864 146) is "presumptively valid" to within the relevant legal standard, we already collectively have knowledge of it and can't really pretend otherwise given things that are already in the mailing list archives, and AFAICT (IANAL, TINLA) it can't possibly describe anything other than the tweak of a hack on a kludge that is the DTS format. Contrariwise, given the thicket of claims (49 in the US version, which is a monstrous 261K in HTML), it strikes me as quite impossible to determine whether it is 100% unoriginal without litigation. If it is important to enough DD's to do something other than drop it and give its derivatives and CD/DVD distributors a heads-up (IANADD and it's not actually my problem), then SPI and/or one of its EU sister organizations had better obtain opinion of competent counsel. IANAL, TINLA. FWIW it appears that Philips has also stopped supporting DTS in its DVD-R/RW drives: http://www.licensing.philips.com/licensees/patent/dvdrw/documents760.html . Cheers, - Michael
Re: libdts patent issue?
On Sat, Jul 16, 2005 at 12:44:11AM -0700, Michael K. Edwards wrote: > This came up in the course of a curiosity-driven review of the patent > status of various audio codecs, and also in a recent MPlayer thread. > It would probably be wise not to wait for a cease-and-desist letter > before quietly discontinuing the distribution of libdts. See > http://developers.videolan.org/libdca.html and > http://www.via.ecp.fr/via/ml/vlc-devel/2005-04/msg00230.html . Might > want to give derivatives a heads-up too. > > I don't say I like it, but it's probably the wrong battlefield on > which to fight the "software patent" wars. The patent (US #5,956,674, > EP 864 146) is "presumptively valid" to within the relevant legal > standard, Software patents are not legal in Europe. Period. The European patent convention from 1972 explicitly excludes software from patentability. Attempts to pass legislation that would have allowed software to become patentable have failed. The worst thing we could do now is give in to the patent scare tactic and stop developing and distributing software that might infringe patents that might have some validity. If DTS Inc. would have prevailed in court is very very questionable. Regretfully the university hosting libdts and the VideoLAN developers did not have the motivation, energy and money to fight this out. > we already collectively have knowledge of it and can't > really pretend otherwise given things that are already in the mailing > list archives, and AFAICT (IANAL, TINLA) it can't possibly describe > anything other than the tweak of a hack on a kludge that is the DTS > format. Contrariwise, given the thicket of claims (49 in the US > version, which is a monstrous 261K in HTML), it strikes me as quite > impossible to determine whether it is 100% unoriginal without > litigation. Why stop at libdts anyway? During the course of the discussion we came up with a lot of MPEG-related patents, so why not remove all multimedia packages same as Red Hat did? This is a very slippery slope, if a few software patents get accepted as valid soon all of main will have to be removed. Diego -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: libdts patent issue?
On 7/16/05, Diego Biurrun <[EMAIL PROTECTED]> wrote: > Software patents are not legal in Europe. Period. The European patent > convention from 1972 explicitly excludes software from patentability. > Attempts to pass legislation that would have allowed software to become > patentable have failed. The worst thing we could do now is give in to > the patent scare tactic and stop developing and distributing software > that might infringe patents that might have some validity. Is there no "presumptively valid" standard in any country in Europe -- or does it not apply to patents that actually issue under that country's laws, unless and until they are demonstrated to be invalid in court? Otherwise, if Debian collectively wants to keep distributing libdts in Europe, I think it would be prudent to obtain the advice of competent counsel, which I emphatically am not (and neither AFAIK are you). And for that matter, do you think Debian is magically insulated from US law? Put aside the knee-jerk hostility to the phrase "software patent" for a moment, and ask yourself: what would you do if your assets were on the line? > If DTS Inc. would have prevailed in court is very very questionable. > Regretfully the university hosting libdts and the VideoLAN developers > did not have the motivation, energy and money to fight this out. I've read that thread, including your contributions to it, and it does not surprise me in the least that you are the first to advocate fingers in ears. Which may be the right answer, or may not, depending on factors like whether CD vendors get the C&D letters before Debian does -- making it a little bit socially awkward for Debian to piss away any possible goodwill that might obtain, from the entity which AFAICS has the upper hand under real-world law, some forbearance with respect to copies of Sarge already pressed. > Why stop at libdts anyway? During the course of the discussion we came > up with a lot of MPEG-related patents, so why not remove all multimedia > packages same as Red Hat did? Not a very successful reductio ad absurdum. The scope of patents differs, the declared enforcement intentions differ, the grounds for a laches (unreasonable delay of prosecution) defense are a lot stronger WRT projects that the patent holder has known about for many years, etc., etc. > This is a very slippery slope, if a few software patents get accepted as > valid soon all of main will have to be removed. Yeah, right. The sky is falling. Read a software patent that has withstood close scrutiny sometime; you are likely to find that it contains a non-trivial invention, arrived at after much experimentation, that is not just a matter of natural law -- just like any other patent fit to be granted. The examiners let a lot of crap go by but that's not what drives the patent system, as you can tell by looking at the size of the "expired for non-payment of maintenance fees" list each year. Cheers, - Michael (IANADD, IANAL, TINLA)
Re: libdts patent issue?
Michael K. Edwards wrote: > Is there no "presumptively valid" standard in any country in Europe -- > or does it not apply to patents that actually issue under that > country's laws, unless and until they are demonstrated to be invalid > in court? There's no explicit law that says patent are presumed valid, but it's what the courts will assume unless the defendant raises a good defense. But again this differs from country to country. The situation is a little more complex, because the European Patent Office issues the patents following its own interpretation of the European Patent Convention. National courts then determine infringement and validity. There's no requirement that the national courts follows the EPO's interpretation. It is the usual practice though. And I don't want to restart the whole "software patents are valid/ invalid in Europe" discussion. The attempt to pass a European directive to regulate the issue failed. All that means is that we're back at square one: waiting for national caselaw. I don't know of any caselaw in any European country in the past ten years that says "This European patent is invalid because it's a computer program as such." Arnoud -- Arnoud Engelfriet, Dutch & European patent attorney - Speaking only for myself Patents, copyright and IPR explained for techies: http://www.iusmentis.com/ -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: libdts patent issue?
On Sat, Jul 16, 2005 at 03:41:03AM -0700, Michael K. Edwards wrote: > On 7/16/05, Diego Biurrun <[EMAIL PROTECTED]> wrote: > > Software patents are not legal in Europe. Period. The European patent > > convention from 1972 explicitly excludes software from patentability. > > Attempts to pass legislation that would have allowed software to become > > patentable have failed. The worst thing we could do now is give in to > > the patent scare tactic and stop developing and distributing software > > that might infringe patents that might have some validity. > > Is there no "presumptively valid" standard in any country in Europe -- > or does it not apply to patents that actually issue under that > country's laws, unless and until they are demonstrated to be invalid > in court? Otherwise, if Debian collectively wants to keep > distributing libdts in Europe, I think it would be prudent to obtain > the advice of competent counsel, which I emphatically am not (and > neither AFAIK are you). Neither of us is. > And for that matter, do you think Debian is > magically insulated from US law? DTS Inc. used a European patent to go after VideoLAN. > Put aside the knee-jerk hostility to > the phrase "software patent" for a moment, and ask yourself: what > would you do if your assets were on the line? Defend them (in court if necessary). > > If DTS Inc. would have prevailed in court is very very questionable. > > Regretfully the university hosting libdts and the VideoLAN developers > > did not have the motivation, energy and money to fight this out. > > I've read that thread, including your contributions to it, and it does > not surprise me in the least that you are the first to advocate > fingers in ears. I'm not advocating "fingers in ears". I'm fully aware of the problem I just believe that running away from it is exactly the wrong reaction and does a disservice to the free software community. Let me add that when I speak about fighting software patents I do more than just talk about the subject. I spent four days on the FFII booth at CeBIT, I've talked to my representatives in person and on the phone, I've been in Brussels, etc. > Which may be the right answer, or may not, depending > on factors like whether CD vendors get the C&D letters before Debian > does -- making it a little bit socially awkward for Debian to piss > away any possible goodwill that might obtain, from the entity which > AFAICS has the upper hand under real-world law, some forbearance with > respect to copies of Sarge already pressed. As far as you can see, but you are not are not a lawyer and the only lawyer in the MP3 thread did not agree with your position at all... And which of our two pieces of advice is better for free software in the long run? > > Why stop at libdts anyway? During the course of the discussion we came > > up with a lot of MPEG-related patents, so why not remove all multimedia > > packages same as Red Hat did? > > Not a very successful reductio ad absurdum. The scope of patents > differs, the declared enforcement intentions differ, the grounds for a > laches (unreasonable delay of prosecution) defense are a lot stronger > WRT projects that the patent holder has known about for many years, > etc., etc. Debian has been distributing libdts for quite some time now without problems... > > This is a very slippery slope, if a few software patents get accepted as > > valid soon all of main will have to be removed. > > Yeah, right. The sky is falling. Read a software patent that has > withstood close scrutiny sometime; you are likely to find that it > contains a non-trivial invention, arrived at after much > experimentation, that is not just a matter of natural law -- just like > any other patent fit to be granted. The examiners let a lot of crap > go by but that's not what drives the patent system, as you can tell by > looking at the size of the "expired for non-payment of maintenance > fees" list each year. Feel free to point me at a software patent you consider worth its salt. All the ones I read weren't. It's an inherent problem. Software just does not work like, e.g. cars. It's much more abstract and much more complex. While a car has about 1-15000 parts, programs can easily have ten times that and more, if you consider complete operating systems or - say - Debian we're talking about several magnitudes of complexity. So there are many more parts that might be patented and due to the abstract nature of software it is much harder to tell whether program X really falls under the scope of patent Y. This has been explained before by other people, the paper I like best is "SOFTWARE PATENTS: AN INDUSTRY AT RISK" by The League for Programming Freedom: http://lpf.ai.mit.edu/Patents/industry-at-risk.html I suggest you to read it, it's brilliant if a little longuish, but this way it really explores the subject in depth. I'm sure length is not a problem for you. But please let us not veer off into the pr
Re: libdts patent issue?
On 7/16/05, Diego Biurrun <[EMAIL PROTECTED]> wrote: > DTS Inc. used a European patent to go after VideoLAN. One with a US equivalent, paint by numbers. Presumably they used its EP number in the C&D to the VideoLAN folks because they're in France. C'est la même chose. > Defend them (in court if necessary). At what odds? Why pick this battle? > I'm not advocating "fingers in ears". I'm fully aware of the problem I > just believe that running away from it is exactly the wrong reaction and > does a disservice to the free software community. Survive to fight another day. If you persist in seeing "software patents" (a false category if I ever heard one) as evil, don't be the means of letting this particular tentacle of evil screw over Debian's CD/DVD distributors if they see fit to respond decently to a quiet, voluntary retreat that reflects the reality upstream. > Let me add that when I speak about fighting software patents I do more > than just talk about the subject. I spent four days on the FFII booth > at CeBIT, I've talked to my representatives in person and on the phone, > I've been in Brussels, etc. But have you _thought_ about the problem? Have you _read_ one? Do you have a sane distinction to propose between "software patents" and everything else, or is this just NIMBYism? I think patenting pure math is just as stupid as you do -- but I can tell pure math from the sort of semi-empirical semi-invention disclosed in, say, US #5,924,060 (one of the Fraunhofer suite, which I half-analyzed in the XMMS/MP3 thread). And I can tell you that many signal electronics patents that don't involve RF and higher frequencies could equally well (if not equally cheaply) be implemented with off-the-shelf A/D - DSP - D/A. Does that make them "software patents"? > As far as you can see, but you are not are not a lawyer and the only > lawyer in the MP3 thread did not agree with your position at all... If you are referring to Mr. Ravicher, he seems to be a decent guy (and vastly more qualified than I) but speaking out of where the sun don't shine insofar as he encourages willful disregard of a presumptively valid patent without benefit of the advice of competent counsel. If SPI (or Debian, to the extent that it legally exists) decides that he or someone else constitutes competent counsel, obtains an opinion on the basis of which they decide to continue the distribution of libdts, and makes some sort of plan for what to do if and when CD/DVD vendors (among others) get C&D letters and come to Debian for advice, that is of course a completely different story. > And which of our two pieces of advice is better for free software in the > long run? IMHO, the one that makes a distinction between free speech and free beer (or free brewing equipment), and doesn't demand that everyone die of thirst because the beer isn't free when there's perfectly good water sitting right next to it. Which one do you think I am describing? > Debian has been distributing libdts for quite some time now without > problems... Let's see, a single upload in February 2004, which appears to have gotten through NEW in March and to have been installed by less than one in 40 popcon users, and never to have been put on a Debian CD/DVD before sarge released. Yeah, that's a basis for a laches defense -- not. > Feel free to point me at a software patent you consider worth its salt. > All the ones I read weren't. It's an inherent problem. Software just > does not work like, e.g. cars. It's much more abstract and much more > complex. While a car has about 1-15000 parts, programs can easily > have ten times that and more, if you consider complete operating systems > or - say - Debian we're talking about several magnitudes of complexity. > So there are many more parts that might be patented and due to the > abstract nature of software it is much harder to tell whether program X > really falls under the scope of patent Y. Did you read the Fraunhofer suite? I haven't looked closely at all of them, but I've seen patents a whole lot more bogus than any of them that have nothing to do with software (and aren't in the "crackpot" category either). Any car made for sale in a developed country in the last five years, anyway, has software in it that is a lot more complex than an MPEG-2 encoder, let alone an MP3 decoder (and yes, I've hacked on both); and the spec for a single overhead cam (metallurgy and machining) probably exceeds the length of that MPEG-2 encoder's source code. Software, my friend, is not rocket science. > This has been explained before by other people, the paper I like best is > "SOFTWARE PATENTS: AN INDUSTRY AT RISK" by The League for Programming > Freedom: > > http://lpf.ai.mit.edu/Patents/industry-at-risk.html Yeah, that would be Moglen and RMS's crowd again. Funny how these monomaniacal, non-reality-based, but strangely compelling (until you pick at them) position papers come from a common source A few concepts
Re: libdts patent issue?
On Sat, Jul 16, 2005 at 06:16:14AM -0700, Michael K. Edwards wrote: > On 7/16/05, Diego Biurrun <[EMAIL PROTECTED]> wrote: > > Defend them (in court if necessary). > > At what odds? Why pick this battle? According to Mr Ravicher the odds are not that bad. Why give in before the battle even started? What if there is no problem? Software will not remain free if you don't defend it and you will not keep your freedom if you are not willing to fight for it. But if you want to give away your freedoms beforehand just in case someone might want to take them from you... Please remember that this is my answer to your question of what _I_ would do, I didn't say what Debian should do. > > I'm not advocating "fingers in ears". I'm fully aware of the problem I > > just believe that running away from it is exactly the wrong reaction and > > does a disservice to the free software community. > > Survive to fight another day. If you persist in seeing "software > patents" (a false category if I ever heard one) as evil, don't be the > means of letting this particular tentacle of evil screw over Debian's > CD/DVD distributors if they see fit to respond decently to a quiet, > voluntary retreat that reflects the reality upstream. And which day would that be? And what would be left to fight for anyway? Giving in to patent FUD is a way of strengthening it. If you accuse me of "fingers in ears" I'll have to accuse you of defeatism. Reality upstream is that the hosting university could not be bothered to fight the patent, even though the consensus was that the patent is invalid. > > Debian has been distributing libdts for quite some time now without > > problems... > > Let's see, a single upload in February 2004, which appears to have > gotten through NEW in March and to have been installed by less than > one in 40 popcon users, and never to have been put on a Debian CD/DVD > before sarge released. Yeah, that's a basis for a laches defense -- > not. Mandriva also has it, BTW: http://www.mandriva.com/products/101/powerpack/packages?p=media-main6-dtsdec-0.0.2-1mdk.i586.rpm.html Just for everybody's information. > > This has been explained before by other people, the paper I like best is > > "SOFTWARE PATENTS: AN INDUSTRY AT RISK" by The League for Programming > > Freedom: > > > > http://lpf.ai.mit.edu/Patents/industry-at-risk.html > > > > I suggest you to read it, it's brilliant if a little longuish, but this > > way it really explores the subject in depth. I'm sure length is not a > > problem for you. > > If you call that "depth" then frankly you can't tell scholarship from > demagoguery. It seems that at least I have read it, while you haven't... > But the reality here on Planet Earth is that the law is a blunt tool > and that it's just not smart to hold one of the best collections of > non-fictional free speech yet assembled hostage to a half-finished > implementation of a half-assed specification that a marginal but > cash-rich entity with current law on its side wants to suppress. Given that more and more DVDs come with DTS audio, this software is useful whether you like the specification or not. It's also not half-finished, it works just fine. Multimedia is a field that is plagued by particularly many software patents. As I said, it's a slippery slope, if you drop multimedia today, what is going to be left tomorrow? Michael, let's try to keep this ontopic please. To cut a long story short, you suggest dropping the package, unless legal advice can show it to be harmless. I suggest keeping it unless legal advice shows it to be harmful. Diego P.S.: Please don't CC me, I'm subscribed. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: libdts patent issue?
On 7/16/05, Diego Biurrun <[EMAIL PROTECTED]> wrote: > According to Mr Ravicher the odds are not that bad. Why give in before > the battle even started? What if there is no problem? Software will not > remain free if you don't defend it and you will not keep your freedom > if you are not willing to fight for it. But if you want to give away > your freedoms beforehand just in case someone might want to take them > from you... Let me quote you part of the "Caveat" from the research project ( http://www.patstats.org/editors_page.html ) from which Mr. Ravicher said he got his numbers: While we believe it is important for actual and potential litigants and their attorneys to know what has been happening recently in the courts on various patent law points, the value of statistics can easily be exaggerated. We caution that any set of statistics should be used with care, in that it is not always apparent what the data may signify. Shifting figures for court results on a particular issue may reflect not so much a predisposition of a court as they do variations in the strengths of positions that litigants bring to that court. For example, if patentees are more often willing to litigate the on-sale issue on weaker facts than at some previous time, we should expect to see a decrease in the percentage of times the patentee prevails on that issue, regardless of any judicial attitudes on the subject. If defendants become more selective in their use of the inequitable conduct defense, we should expect to see a rise in their success rate on that point, again independent of any shift in judicial attitude. Moreover, since very few issued patents are actually litigated to judgment, the reported statistics do not necessarily reflect on the overall quality of patents being issued by the Patent and Trademark Office, but at most only the quality of those chosen by litigants to be asserted in lawsuits. That's not my idea of evidence that it's a dice roll, nor is it a set of odds on which to base a large gamble with other people's money -- and by definition such a cavalier attitude does not meet the "duty of due care" even if it is supported by opinion of counsel. Have you _read_ Knorr-Bremse? Do you understand what was at stake and by how much -- or how little -- the Federal Circuit relaxed the "duty of due care" calculation that courts of fact are expected to apply? Do you understand that the Federal Circuit en banc is absolutely the last word on the subject in US law unless overruled by the Supreme Court or by act of Congress? (IANAL, TINLA) You will not "keep your freedom" if you run under a tractor, rob your next-door neighbor in broad daylight, or show up in court in response to a non-frivolous suit using your pet lizard for an attorney, either. In this case, if continued distribution of libdts is a "freedom" that Debian is "willing to fight for", then somebody had better start shelling out for competent counsel and/or making the rounds to find someone to take the case pro bono. Mr. Ravicher and the SFLC appear to be making the latter offer, and I would be very entertained to watch them in action, but AFAIK no one who may feel a need to do so has taken them up on the offer -- not Sam Hocevar, not the ftpmasters, not the DPL, and not the board of SPI. > Please remember that this is my answer to your question of what _I_ would > do, I didn't say what Debian should do. So suppose a buddy of yours goes around handing people, including you and your next door neighbor, bags of Everlasting Gobstoppers -- candies that, in a way, contain their own recipe and never run out. Now Willy Wonka comes to your neighbor's door with his lawyer in tow and says, "actually, those Gobstoppers were stolen from me; here's a certificate issued by the administrative authorities of half a dozen nations, duly constituted by their legislatures in accordance with several centuries of legal precedent, that says so; please flush them down the toilet and we'll call it quits." Your neighbor isn't particularly impressed but his lawyer tells him it's stupid to challenge the validity of that certificate, so the Gobstoppers go to Fish Heaven. You have advertised in a small way that you hand paper bags full of candy, Gobstoppers included, to anyone who wants them. People who trust you sell nice gift boxes with those candy mixes in them for the price of the gift box. There's no hiding your or your distributors' dealings in candy, there's no pretending you didn't see Wonka calling on your neighbor, there's no reason he would smile on your actions and not your neighbor's. The Candy Laws say that, under those circumstances, failing to either quit handing out Gobstoppers or make a reasonably honest _and_competent_ effort to evaluate the legitimacy of Wonka's claim constitutes "willful infringement" and may triple the penalties Wonka can inflict on you in court if you and he can't settle it amicably. Now you're totally opposed to the idea that an
Re: libdts patent issue?
Arnoud Engelfriet wrote: I don't know of any caselaw in any European country in the past ten years that says "This European patent is invalid because it's a computer program as such." That's not the caselaw you're looking for. The caselaw you're looking for is "This European patent is invalid because it is a mathematical method." "Mathematical methods" are explicitly unpatentable under the European Patent Convention. (Unlike in the US, where they were only unpatentable due to caselaw, which was overturned by the Federal Circuit, which didn't realize or admit that it was breaking stare decisis.) So-called "software patents" are in fact patents on mathematical algorithms. If necessary, I can get a near-infinite number of computer scientists and mathematicians of the top caliber to testify that algorithms for computers *are* mathematical methods. (Conversely, most mathematical methods can be transformed losslessly into algorithms for computers.) The equivalence was established at least as early as Shannon. This is a matter of fact, not law. If a judge thinks otherwise, it's probably because he or she hasn't been given the expert testimony on the issue. If there is case law to the contrary, it is due to misunderstanding of the facts -- because this is an issue of fact, not law -- and any competent judge would ignore it for that reason. It's actually an excellent idea to establish some case law by killing a patent on the basis of that clause of the European Patent Convention. The key requirement, however, is to pull together a *lot* of *very* strong expert testimony before the fact, because history has shown that judges are quite easily swayed by false claims when they don't understand the field. In reality every pro-software-patent ruling has been violating centuries-old established law against the patenting of mathematics. Judges have had the wool pulled over their eyes; they didn't realize that they were allowing the patenting of mathematical methods. If a judge actually understood that, the weight of stare decisis would be seen to be very much against "software patents". This is why I feel no compunction about saying that the European Patent Convention disallows software patents: because it does. Now, it may be an unenforced provision of the European Patent Convention, one which judges have been (knowingly or unknowingly) ignoring due to a lack of understanding of the facts. But it is clear and unmistakable. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: libdts patent issue?
Nathanael Nerode wrote: > Arnoud Engelfriet wrote: > >I don't know of any caselaw in any European country in the past ten > >years that says "This European patent is invalid because it's a > >computer program as such." > > That's not the caselaw you're looking for. > The caselaw you're looking for is "This European patent is invalid because > it is a mathematical method." As such. Art. 52(3) EPC. It's actually easier to make sure you don't claim math as such than software as such. Just state a practical application in the claim. Here's a claim that would _not_ be maths as such under European law: "A method of encrypting a bitstream A using a key B that is the same length as A, comprising computing A XOR B". Of course this method is long known, but that's question number two. I occasionally see courts dealing with the computer program exception, but I've never seen caselaw about maths as such. Probably because me and my colleagues know that you must claim a practical application of the mathematics, and fortunately that's usually easy. > So-called "software patents" are in fact patents on mathematical algorithms. Probably. Still, the EPC mentions "computer programs as such" and "mathematical methods as such" as two separate categories. See items 2(a) and 2(c) of article 52 EPC. So I guess the drafters of the EPC thought it worth mentioning both. Whether this means they consider maths and software equivalent I don't know; but both are excluded to the same extent, so I guess it doesn't matter. I agree with you that the distinction may seem artificial. But it does seem logical to me to say "you can't patent A XOR B but you can patent a computer program that does that." Then the formula remains public domain; you just can't make, use or sell a program that implements the formula. Were the formula patented, then you couldn't even publish a textbook. It's the same in my eyes as saying "you can't patent a discovery but you can patent a machine that applies this discovery in practice." > It's actually an excellent idea to establish some case law by killing a > patent on the basis of that clause of the European Patent Convention. Try it. I'm able to kill business methods[1], but I would not advise my client to oppose a European patent on the ground that it's a computer program as such. Not even if I had Knuth as expert witness. [1] See the lowermost example on my webpage http://www.iusmentis.com/patents/businessmethods/epoexamples/ > In reality every pro-software-patent ruling has been violating > centuries-old established law against the patenting of mathematics. Judges > have had the wool pulled over their eyes; they didn't realize that they > were allowing the patenting of mathematical methods. If a judge actually > understood that, the weight of stare decisis would be seen to be very much > against "software patents". This is Europe. We don't do stare decisis. (England does, but they usually don't call themselves Europe :) ) European law is civil law, and it's the statute that determines what is patentable. The European Patent Convention does not say that "computer programs" are excluded, but makes a point of saying that only "computer programs as such" are excluded. Since that's what the law says, that must mean something. Arnoud -- Arnoud Engelfriet, Dutch & European patent attorney - Speaking only for myself Patents, copyright and IPR explained for techies: http://www.iusmentis.com/ -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: libdts patent issue?
> Software patents are not legal in Europe. Period. The European > patent convention from 1972 explicitly excludes software from > patentability. Attempts to pass legislation that would have > allowed software to become patentable have failed. The worst > thing we could do now is give in to the patent scare tactic and > stop developing and distributing software that might infringe > patents that might have some validity. Ditto, for Brasil. Software patents are explicitly excluded in our Industrial Property (= Patents + Trademarks) Act (Law 9279/96), section 10, V: " 10. It shall not be considered invention or utility model: I - discoveries, scientific theories and mathematical methods; II - purely abstract conceptions; III - schemes, plans, principles or methods of commerce, accounting, finance, education, advertising, lottery or fiscalization; IV - literary, architetonic, artistic, scientific works or any aesthetical creation; V - computer programs by themselves; VI - information presentation; VII - game rules; VIII - surgical techniques or methods, as well as therapeutic or diagnosis methods for application both in human or animal bodies; IX - all or part of natural living beings and biological materials found in nature, even if in isolation, including the genome and germoplasma of any natural living being and biological natural processes. " Obviously, only inventions (or utility models) can be patented. -- HTH, Massa -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: libdts patent issue?
On 7/18/05, Humberto Massa Guimarães <[EMAIL PROTECTED]> wrote: > Ditto, for Brasil. Software patents are explicitly excluded in our > Industrial Property (= Patents + Trademarks) Act (Law 9279/96), > section 10, V: " [snip] > Obviously, only inventions (or utility models) can be patented. Now that Arnoud has explained how the "no patents on mathematics, software, or business methods" premise works in Europe -- indistinguishable in my eyes from how it works under current law in the US, except that EPO's examiners may or may not be on average more competent, less captured, or less overwhelmed than the USPTO's -- I would not be at all surprised if it works similarly under the statute you quote. Humberto, can you give me some kind of pointer to the primary literature (i. e., what judges cite in their opinions as interpretive guidance) in your legal system? Cheers, - Michael
Re: libdts patent issue?
Summary: it looks to me like current US and European law on the patentability of math, software, and business methods are already very, very closely aligned. Gripe, if you like, about the USPTO's ignorance of the prior art in software-intensive fields, and about the unholy alliance between patent agents with big thesauruses and corporate assignees who are more interested in obtaining ammunition for FUD than an accurate statement of the extent of the novelty in their "invention". But don't blame Congress or the Federal Circuit, and don't believe everything you read in the funny papers. IANAL, TINLA. On 7/17/05, Arnoud Engelfriet <[EMAIL PROTECTED]> wrote: > As such. Art. 52(3) EPC. It's actually easier to make sure you > don't claim math as such than software as such. Just state > a practical application in the claim. > > Here's a claim that would _not_ be maths as such under European law: > "A method of encrypting a bitstream A using a key B that is the > same length as A, comprising computing A XOR B". > > Of course this method is long known, but that's question number two. That is precisely the distinction applied under current law in the US as I understand it (IANAL). The latest word on the topic appears to be AT&T Corp. V. Excel Comm. Inc. [50 USPQ2d 1447, 1452 (Fed. Cir. 1999)] at http://caselaw.lp.findlaw.com/data2/circs/Fed/981338v2.html . AT&T had patented a method of identifying whether both ends of a long distance call were in the same network that used simple Boolean algebra. The district court invalidated the patent on the grounds that it was not statutory subject matter, for lack of any "physical limitations" (as required under the older Freeman-Walter-Abele test). Judge Plager's appellate opinion points out that, subsequent to "In re Alappat" (a 1994 decision of the Federal Circuit en banc), it applies a "useful, concrete, tangible result" test instead: "our inquiry here focuses on whether the mathematical algorithm is applied in a practical manner to produce a useful result." The circuit court reversed and remanded, with a reminder that "the ultimate validity of these claims depends upon their satisfying the other requirements for patentability such as those set forth in 35 U.S.C. 102, 103, and 112" -- and sure enough, on remand the patent was invalidated for the right reasons, anticipation and obviousness. > I occasionally see courts dealing with the computer program > exception, but I've never seen caselaw about maths as such. > Probably because me and my colleagues know that you must > claim a practical application of the mathematics, and fortunately > that's usually easy. And it limits the scope of the monopoly granted to the cited application(s). US Patent law appears to apply a criterion not dissimilar from the copyright law "doctrine of merger" -- the Supreme Court, in Diamond v. Diehr [450 U.S. at 175, 209 USPQ 1 (1981)], confirmed that "abstract ideas" are excluded from patentability and hence any "invention" which constitutes an obvious application of an abstract idea is denied patent protection. I think it's fair to say that an abstract idea in the form of an algorithm, "applied" to the making of a "machine" that executes it (those absurd "register means", "processor means", and "storage means" of the 80's and 90's), isn't really "applied" yet. It becomes patentable subject matter when it is applied to a concrete problem domain in which that algorithm is not an obvious solution and produces a useful result; and it doesn't really matter whether it can be built with commodity hardware or has to be realized as an ASIC to be commercially competitive. Hence it is essentially also true in the US that software is not patentable "as such", only its application to a concrete problem domain. The fact that this distinction is widely misunderstood or misrepresented (IMHO, IANAL) by commentators, especially those with a nest to feather or an axe to grind, is not the Federal Circuit's fault. The USPTO isn't helping much, as they haven't fully understood this opinion either as far as I am concerned. See discussion of the continuing relevance of the Koo patent case http://www.uspto.gov/web/menu/pbmethod/trangmaterials.ppt (Google will convert it to HTML for you). The "storage in computer-readable memory" shibboleth from the mid-90's had to do with uncertainty about whether "concrete", where signals were concerned, meant "must be a machine instead of a process" (the claim formulas are different in USPTO guidance), and AFAICT this has also been superseded by AT&T v. Excel. > Try it. I'm able to kill business methods[1], but I would not > advise my client to oppose a European patent on the ground that > it's a computer program as such. Not even if I had Knuth as > expert witness. > > [1] See the lowermost example on my webpage > http://www.iusmentis.com/patents/businessmethods/epoexamples/ Note that the relevant decision in the US, State Street v. Signature ( http://laws.f
Re: libdts patent issue?
Arnaud Engelfriet wrote: >Here's a claim that would _not_ be maths as such under European law: >"A method of encrypting a bitstream A using a key B that is the >same length as A, comprising computing A XOR B". That *is* math. If a judge has ruled that it isn't, he doesn't know what the hell he's talking about. The bitstream A, the key B, and the output are mathematical objects, and the "method" is a mathematical algorithm. Unless it involves measuring the local wind velocity or some other non-mathematical element. Everything I said remains true. If a judge really has ruled that that isn't a mathematical method, then there is indeed caselaw which is wrong as a matter of fact. Which would suck, but there you are. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: libdts patent issue?
Arnoud Engelfriet wrote: >I agree with you that the distinction may seem artificial. But it >does seem logical to me to say "you can't patent A XOR B but you can >patent a computer program that does that." If you can patent the class of computer programs which do A XOR B, you have patented the abstract operator which does that. >Then the formula remains >public domain; you just can't make, use or sell a program that >implements the formula. Were the formula patented, then you couldn't >even publish a textbook. Unfortunately, that's a distinction without a difference. If you're prohibited from making a computer program implementing the algorithm, you're prohibited from writing a formal description of the algorithm, which is a standard textbook technique. (A computer program *is* a formal description of an algorithm.) If you're prohibited from selling such a program, you're prohibited from selling such a textbook. The use prohibition is at least different: if only the use prohibition were present, you could indeed publish a textbook, but nobody would be allowed to use its techniques without a license. According to this "distinction", we could distribute Debian as a "computing textbook" rather than as a "system", and we would then be exempt from these patent considerations. (The current US rule is that that every such patent is for a "program plus a generic computer", so this should actually work. Only the silly people who use Debian on their computers who are violating the patent, although Debian might be in trouble for encouraging them. Actually, the users might be fine too, because they're not using the process on an industrial scale. Hey -- there's a solid line of argument: the program itself doesn't violate anything, only the program plus a computer, and it's only combined with the computer at the end-user's house. This is silliness, of course, but that's what you get for making a distinction without a difference.) I am sure ill-informed judges are making this "distinction", but it's something like saying "The ideas are free; it's only thinking them which is illegal". It's complete and utter bullshit. Most likely judges are going by some sort of incompetent gut instinct, and allowing "dead-tree" publication of algorithms while prohibiting online publication. A deterministic algorithm which takes one bit sequence as input and produces another as output is a piece of pure mathematics. And that is *exactly* the sort of thing which is being patented under the name of "software patents". The IEEE magazine (Spectrum) had an article about this recently, which made much the same point: you cannot make a valid distinction of this sort. It went into some detail on why various "distinctions" used by the US courts are hopeless and illusory. The current US excuse is that any mathematical algorithm "plus a generic computer" is patentable. Is that what's being used in Europe too? There's a reason the FFII preferred standard is that the inventive part of a patent must be on some method of manipulating the physical world. That's the last place you can make a distinction where there's a difference. Once you allow patents on abstract algorithms (even if you require the pro forma utterance "plus a generic computer"), you allow patents on large swathes of mathematics, whether you admit it or not. (Not *all* mathematics is algorithms, of course. If European law said that "mathematical methods, except algorithms" were unpatentable, that could be different. But algorithms are quite definitely mathematics. However, there are lossless transformations to convert much of mathematics into algorithms and back.) >It's the same in my eyes as saying "you can't patent a discovery >but you can patent a machine that applies this discovery in >practice." It's not, though. Look again. Now, all algorithms are indeed processes, and most processes can be patented. However, they're purely *mathematical* processes. They must be excluded from patentability if mathematical methods are excluded from patentability. "Mathematica" allows mathematical theorems to be expressed formally (as programs). So according to the theory used currently in the US -- and perhaps in Europe as well -- I can patent a mathematical theorem (plus a generic computer) right now. As it is, all pure mathematics is patentable by law in the US (via the "just add generic computer" method), thanks to the incompetent rulings of the Federal Circuit and others. The basic problem is this: a computer is a machine for doing mathematics. A computer program is a formal description of a piece of mathematics. It is a "method" for computing something, composed of mathematical elements (all the basic bit operations are purely mathematical) -- which means that it is a piece of mathematics. Judges, and apparently you :-), have been fooled by the language of "a method for doing X", which looks like the langua
Re: libdts patent issue?
Nathanael Nerode wrote: > Arnoud Engelfriet wrote: > >Then the formula remains > >public domain; you just can't make, use or sell a program that > >implements the formula. Were the formula patented, then you couldn't > >even publish a textbook. > Unfortunately, that's a distinction without a difference. If you're > prohibited from making a computer program implementing the algorithm, you're > prohibited from writing a formal description of the algorithm, which is a > standard textbook technique. (A computer program *is* a formal description > of an algorithm.) If you're prohibited from selling such a program, you're > prohibited from selling such a textbook. No, you're not. The textbook is not the same thing as a computer program in machine-readable form. This is the same as with those crypto algorithms that can't (couldn't) be exported as .c files but could be exported as printed pages. > The use prohibition is at least different: if only the use prohibition were > present, you could indeed publish a textbook, but nobody would be allowed to > use its techniques without a license. That's basically how patent law works in every area. You can publish the knowledge but not apply the knowledge to make, use or sell a working device or actual product. And a book that humans can read is not a product in patent law. > According to this "distinction", we could distribute Debian > as a "computing textbook" rather than as a "system", and we would then be > exempt from these patent considerations. I think it actually has to _be_ a textbook before you can call it a textbook. > (The current US rule is that that every such patent is for a "program plus a > generic computer", so this should actually work. Right, although there's lots of legal concepts like contributory or vicarious infringement that could come into play. But I don't see how distributing a .c file can be held a direct infringement of a patent on a "computer equipped with code for performing X". > A deterministic algorithm which takes one bit sequence as input and produces > another as output is a piece of pure mathematics. And that is *exactly* the > sort of thing which is being patented under the name of "software patents". To me the distinction is clear: you have to add something to the algorithm before you arrive at patentable matter. You apparently consider the addition (a computing device with a memory) to be irrelevant, and hence you don't see a distinction. > The IEEE magazine (Spectrum) had an article about this recently, which made > much the same point: you cannot make a valid distinction of this sort. It > went into some detail on why various "distinctions" used by the US courts are > hopeless and illusory. The current US excuse is that any mathematical > algorithm "plus a generic computer" is patentable. Is that what's being used > in Europe too? The European standard is that the claim must cover a piece of technology: a device or method that exhibits a "technical effect". And no, I don't have a definition for "technical". In Europe we look at this effect to determine if it's patentable. The question whether it's an algorithm or software is largely irrelevant. You could build dedicated hardware that performs the algorithm instead of software. If the end result is something "technical", then it's patentable. Perhaps this is more sensible to you than the US approach? > There's a reason the FFII preferred standard is that the inventive part of a > patent must be on some method of manipulating the physical world. That's the That's what European patent law also pretends to be. FFII is pushing a very restrictive definition of what "manipulating the physical world" means, but otherwise they're completely in line with how patent law works. > I guess "the mathematics is public domain, but any use of it is patented"? More like, the expression in .obj is patented, but the expression in .PDF is not. Feel free to publish papers; don't distribute devices that execute the algorithm disclosed in those papers. > > but I would not > >advise my client to oppose a European patent on the ground that > >it's a computer program as such. > Again, the grounds should be that it's a *mathematical method*. The "method" > or "process" in such a patent is in fact a *mathematical* method. You get > mathematicians specializing in fields like theory of computation to testify > to that. The inventive step is solely in the new mathematical method (not in > the generic computer). The problem is exactly the same: European patent law does not exclude patents on mathematical methods, but only on mathematical methods _as such_. Apparently this is not the same thing for the people who wrote that law. They may have been wrong, but if this is the law, then that's what we have to work with. But again, you have to show a relation to the physical world. In Europe, it has to be technology, or it's not patentable. Arnoud -- Arnou
Re: libdts patent issue?
On 7/19/05, Nathanael Nerode <[EMAIL PROTECTED]> wrote: > Arnaud Engelfriet wrote: > >Here's a claim that would _not_ be maths as such under European law: > >"A method of encrypting a bitstream A using a key B that is the > >same length as A, comprising computing A XOR B". > > That *is* math. If a judge has ruled that it isn't, he doesn't know what the > hell he's talking about. The bitstream A, the key B, and the output are > mathematical objects, and the "method" is a mathematical algorithm. Unless > it involves measuring the local wind velocity or some other non-mathematical > element. I think you're missing Arnoud's point. It's not math, it's an application of math to the problem domain of message encryption. That makes it statutory subject matter for patenting, which math as such is not. And it doesn't need to spell out some particular way of converting back and forth between physical observation and bitstream; the relevant courts (in the US and, AIUI, in Europe) have gotten over their Platonic muddle when it comes to "abstract ideas" vs. "processes" and "machines", and focus instead on the requirement that the invention solve a practical problem. > Everything I said remains true. If a judge really has ruled that that isn't a > mathematical method, then there is indeed caselaw which is wrong as a matter > of fact. Which would suck, but there you are. Case law isn't dispositive in Arnoud's system, and where it is it is only dispositive on points of law, not points of fact. Appeals judges in the US have the discretion to review factual determinations only under a "clear error" standard, and even that is constitutionally denied them with respect to jury rulings -- but in any case that "clear error" relates only to the facts in the court record of that trial and no prior case or external knowledge is supposed to be applied. A case could get the facts completely wrong and yet be an excellent precedent on a point of law. Cheers, - Michael (IANAL, TINLA)
Re: libdts patent issue?
On 7/19/05, Michael K. Edwards <[EMAIL PROTECTED]> wrote: > I think you're missing Arnoud's point. It's not math, it's an > application of math to the problem domain of message encryption. That > makes it statutory subject matter for patenting, which math as such is > not. "it" is rather unclear here. There's several values for "it", one of which includes the math, and the other which includes the physical process of encrypting messages. Furthermore, "xor" is a rather trivial transposition cipher (at the bit level), so regardless of whether or not laws can say whether or not the method is patentable they do, existing law seems to say that this would not be patentable. [This would not be the case if bits themselves were non- obvious, but for that to be the case you'd have to be dealing with people from a different field of activity.] If a court has ruled that the patent is valid Nathanael's point that the court having problems recognizing the facts of the case is indeed a salient point. -- Raul
Re: libdts patent issue?
On 7/19/05, Nathanael Nerode <[EMAIL PROTECTED]> wrote: > Arnoud Engelfriet wrote: > >I agree with you that the distinction may seem artificial. But it > >does seem logical to me to say "you can't patent A XOR B but you can > >patent a computer program that does that." > If you can patent the class of computer programs which do A XOR B, > you have patented the abstract operator which does that. Arnoud said that computer programs as such are not patentable subject matter under the EPC, and as I read it that's also true in the US under Alappat's "useful, concrete, tangible result" test. You can patent a computer program that does A XOR B in its character as a solution to the problem of encrypting messages encoded as digital data, but that same program used to achieve some unrelated tangible result isn't covered by the patent. Arnoud, am I getting your argument straight? > >Then the formula remains > >public domain; you just can't make, use or sell a program that > >implements the formula. Were the formula patented, then you couldn't > >even publish a textbook. > Unfortunately, that's a distinction without a difference. If you're > prohibited from making a computer program implementing the algorithm, you're > prohibited from writing a formal description of the algorithm, which is a > standard textbook technique. (A computer program *is* a formal description > of an algorithm.) If you're prohibited from selling such a program, you're > prohibited from selling such a textbook. Patent is not copyright; you don't obtain a monopoly on describing your method, you obtain a monopoly on its commercial application. No patent prohibits you from making a computer program implementing any algorithm you like; but if you sell it as a solution to the problem addressed in the patent, without authorization from the patent holder, you are infringing. The same goes for selling its output, if that's covered by the patent -- compare against the enforcement of chemical process patents. So I think Arnoud's point is that, if a formula or other abstract idea were patentable without any indication of the result being achieved, then a textbook would be just as much an infringement of this counterfactual patent as a computer program or a machine that embodies it. His "make, use, or sell" language is a little bit over-broad, but essentially accurate insofar as the maker may be liable for infringing use of the program by third parties even if he cannot be demonstrated to have made infringing use of it himself or to have profited from its sale. > The use prohibition is at least different: if only the use prohibition were > present, you could indeed publish a textbook, but nobody would be allowed to > use its techniques without a license. If people bought the textbook principally so they could copy down sections that amounted to an implementation of the patented invention, and proceeded to use them in an infringing way, then AIUI you could be liable for contributory infringement. Don't cry First Amendment, either -- it's not the writing and publishing that are getting you in trouble, it's the collusion in tortious and/or criminal activity, and freedom of speech/press doesn't cover that any more that it exonerates a mafia don who orders a hit with a little help from a printing press. > According to this "distinction", we could distribute Debian > as a "computing textbook" rather than as a "system", and we would then be > exempt from these patent considerations. Doesn't make a bit of difference what you call it. What matters is what it is and what people use it for. > (The current US rule is that that every such patent is for a "program plus a > generic computer", so this should actually work. Only the silly people who > use Debian on their computers who are violating the patent, although Debian > might be in trouble for encouraging them. Actually, the users might be fine > too, because they're not using the process on an industrial scale. Hey -- > there's a solid line of argument: the program itself doesn't violate > anything, only the program plus a computer, and it's only combined with the > computer at the end-user's house. This is silliness, of course, but that's > what you get for making a distinction without a difference.) That is not the current US rule as I understand it. Despite the patent-attorney-encouraged "computing means" shibboleths of the 80's and 90's, applications of software techniques to practical problems are just as patentable when stated using "process" lingo as when using "machine" lingo, certainly now (per AT&T v. Excel) but AFAICT all along. The only appellate precedent I have ever seen alluded to which suggests otherwise is an administrative appeal regarding "the Koo patent"; I'll track it down when I have time. As for the "personal use" exemption: thread at http://lists.debian.org/debian-legal/2005/07/msg00267.html . My own opinion (IANAL, TINLA) is that the statutory exemption wouldn't
Re: libdts patent issue?
On 7/19/05, Arnoud Engelfriet <[EMAIL PROTECTED]> wrote: > Nathanael Nerode wrote: > > Unfortunately, that's a distinction without a difference. If you're > > prohibited from making a computer program implementing the algorithm, you're > > prohibited from writing a formal description of the algorithm, which is a > > standard textbook technique. (A computer program *is* a formal description > > of an algorithm.) If you're prohibited from selling such a program, you're > > prohibited from selling such a textbook. > > No, you're not. The textbook is not the same thing as a computer > program in machine-readable form. This is the same as with those > crypto algorithms that can't (couldn't) be exported as .c files > but could be exported as printed pages. If you're talking about the Bernstein v. United States and Junger v. Daley cases, the export authorities made this bogus distinction but the courts (the appellate courts, anyway) did not. > > The use prohibition is at least different: if only the use prohibition were > > present, you could indeed publish a textbook, but nobody would be allowed to > > use its techniques without a license. > > That's basically how patent law works in every area. You can > publish the knowledge but not apply the knowledge to make, use > or sell a working device or actual product. And a book that > humans can read is not a product in patent law. Still a bogus distinction. If the book says, "scan me and use me as an implementation, screw the patent holder", I see no reason which it wouldn't be every bit as infringing as a machine sold to end users in build-it-yourself kit form. When such a kit is sold to OEMs by a DVD chipset vendor, they make it very clear what patents they believe OEMs need to license in order to market the result. > > According to this "distinction", we could distribute Debian > > as a "computing textbook" rather than as a "system", and we would then be > > exempt from these patent considerations. > > I think it actually has to _be_ a textbook before you can call > it a textbook. It at least has to have substantial non-infringing uses in its character as a textbook. How much leeway this buys you with respect to a textbook that doubles as an infringing implementation depends on how the judge views your surrounding conduct. > > (The current US rule is that that every such patent is for a "program plus a > > generic computer", so this should actually work. > > Right, although there's lots of legal concepts like contributory > or vicarious infringement that could come into play. But I don't > see how distributing a .c file can be held a direct infringement > of a patent on a "computer equipped with code for performing X". AFAICT this is not correct under your law or ours; statutes and rulings that protect "staples of commerce" and components with substantial non-infringing uses would not protect a "just add water (solder, computer)" kit. Whether this goes by the name of "contributory" or "vicarious" instead of "direct" infringement makes little or no difference. > > A deterministic algorithm which takes one bit sequence as input and produces > > another as output is a piece of pure mathematics. And that is *exactly* the > > sort of thing which is being patented under the name of "software patents". > > To me the distinction is clear: you have to add something to the > algorithm before you arrive at patentable matter. You apparently > consider the addition (a computing device with a memory) to be > irrelevant, and hence you don't see a distinction. The addition of generic computing means is irrelevant as I understand it. You of course are more qualified than I, but I find it hard to believe unless you can show me that judges care about that distinction. As I understand it, what has to be added is an application to a field of endeavor in which it constitutes a practical invention. > > The IEEE magazine (Spectrum) had an article about this recently, which made > > much the same point: you cannot make a valid distinction of this sort. It > > went into some detail on why various "distinctions" used by the US courts > > are > > hopeless and illusory. The current US excuse is that any mathematical > > algorithm "plus a generic computer" is patentable. Is that what's being > > used > > in Europe too? > > The European standard is that the claim must cover a piece of > technology: a device or method that exhibits a "technical effect". > And no, I don't have a definition for "technical". In the US, it's part of what the courts have read into 35 USC 101's "new and useful process, machine, manufacture, or composition of matter". Even a "process" has to be "useful" for a particular industrial or commercial purpose in order to be patentable. Again, see In re Alappat. > In Europe we look at this effect to determine if it's patentable. > The question whether it's an algorithm or software is largely > irrelevant. You could build dedicated hardware t
Re: libdts patent issue?
On 7/19/05, Raul Miller <[EMAIL PROTECTED]> wrote: > On 7/19/05, Michael K. Edwards <[EMAIL PROTECTED]> wrote: > > I think you're missing Arnoud's point. It's not math, it's an > > application of math to the problem domain of message encryption. That > > makes it statutory subject matter for patenting, which math as such is > > not. > > "it" is rather unclear here. Sorry. "It" = "the patentable subject matter of the invention". > There's several values for "it", one of which includes the math, and the > other which includes the physical process of encrypting messages. The "physical" part is IMHO a Platonic, dualist hang-up. What matters is the application to a practical problem, such as hiding the information content of message A from people who don't have key B. > Furthermore, "xor" is a rather trivial transposition cipher (at the bit > level), > so regardless of whether or not laws can say whether or not the method > is patentable they do, existing law seems to say that this would not be > patentable. [This would not be the case if bits themselves were non- > obvious, but for that to be the case you'd have to be dealing with > people from a different field of activity.] Arnoud's example is of course not patentable -- but on obviousness grounds, not statutory subject matter grounds. It could, however, easily form part of a patentable invention in which the nature of key B is specified -- and if you had a novel way of generating cryptographically strong pseudo-random bitstreams and wanted to patent its cryptographic applications, that's exactly how you would do it. > If a court has ruled that the patent is valid Nathanael's point that the > court having problems recognizing the facts of the case is indeed a > salient point. Judges do not wade into highly technical areas unaccompanied. You might find Alex Kozinski's speech at http://notabug.com/kozinski/claimschief informative. He was the first chief judge of what is now the US Court of Federal Claims, which is the court of first instance for all intellectual property claims against the United States including appeals from administrative decisions of the USPTO. The Federal Circuit, which handles all appeals from the USCFC as well as appeals from district courts involving the validity and scope of patents, has an even more formidable army at its command. Cheers, - Michael (IANAL, TINLA)
Re: libdts patent issue?
On Tue, 19 Jul 2005 15:10:10 +0200 Arnoud Engelfriet wrote: > More like, the expression in .obj is patented, but the expression > in .PDF is not. Feel free to publish papers; don't distribute > devices that execute the algorithm disclosed in those papers. And how is literate programming dealt with? -- :-( This Universe is buggy! Where's the Creator's BTS? ;-) .. Francesco Poli GnuPG Key ID = DD6DFCF4 Key fingerprint = C979 F34B 27CE 5CD8 DC12 31B5 78F4 279B DD6D FCF4 pgp9YEShbRIX6.pgp Description: PGP signature
Re: libdts patent issue?
I wrote: > I am not "pro-software-patent". I think that the USPTO (and, from the > look of it, the EPO) are doing a profoundly incompetent job of > filtering out the trivial and the erroneous from _all_ kinds of patent > applications, not just those which permit an implementation in terms > of a Von Neumann machine with some useful peripherals. But the > reality here on Planet Earth is that the law is a blunt tool ... ... and if you think that US legislators, and even the PTO itself, are not concerned about the "volume of patents" and "cost and difficulty of invalidating cruft" problems, read the CongressDaily article at http://www.govexec.com/dailyfed/0405/042605cdam1.htm . A sane quotation from Sen. Orrin Hatch? Who knew? Cheers, - Michael
Re: libdts patent issue?
I wrote, with regard to aspersions cast by Nathanael on the competence and consistency of judicial opinions in intellectual property arenas: > I am glad that I do not live in the dystopic fantasy world you > describe, with incompetent judges obsessed by sophomoric deductions > from Plato and easily led by the nose. Most judges are not software > engineers but few are utter fools, and to argue otherwise you're going > to need to adduce real evidence. To my own opinion I will add the testimony before Congressional committee of Hon. Alex Kozinski of the Ninth Circuit with respect to the consistency between published and unpublished dispositions in the Ninth, found at http://notabug.com/kozinski/unpublisheddispositions . Whatever else you may say about the Ninth Circuit, it is indisputable that many complex IP cases come before its benches, and it is highly reassuring to me that interested parties are apparently unable to identify previously unknown conflicts in its record when invited to do so. Cheers, - Michael (And while I'm citing Judge Kozinski and his law clerks, you might enjoy his comments on the "fair use" doctrine at http://notabug.com/kozinski/fairuse . I haven't decided how I feel about the merits of his proposal, but he's certainly thinking about the problem.)
Re: libdts patent issue?
Michael K. Edwards wrote: > On 7/19/05, Arnoud Engelfriet <[EMAIL PROTECTED]> wrote: > > That's basically how patent law works in every area. You can > > publish the knowledge but not apply the knowledge to make, use > > or sell a working device or actual product. And a book that > > humans can read is not a product in patent law. > > Still a bogus distinction. If the book says, "scan me and use me as > an implementation, screw the patent holder", I see no reason which it > wouldn't be every bit as infringing as a machine sold to end users in > build-it-yourself kit form. Don't confuse inducement or contributory infringement with direct infringement. You can't claim "a book containing instructions that when loaded into a computer execute the method of claim 1" so there cannot be direct infringement by selling such a book. There are many people who come up with tricks like build-it-yourself kits, omitting a small element and so on. That's why the law also recognizes contributory infringement and penalties for inducing others to infringe. But those concepts are based on a different principle. > > Right, although there's lots of legal concepts like contributory > > or vicarious infringement that could come into play. But I don't > > see how distributing a .c file can be held a direct infringement > > of a patent on a "computer equipped with code for performing X". > > AFAICT this is not correct under your law or ours; statutes and > rulings that protect "staples of commerce" and components with > substantial non-infringing uses would not protect a "just add water > (solder, computer)" kit. Like I said, that wouldn't be _direct_ infringement. > Whether this goes by the name of > "contributory" or "vicarious" instead of "direct" infringement makes > little or no difference. For contributory infringement you need additional evidence. Contributory infringement is knowingly selling or supplying an item for which the only use is in connection with a patented invention. If you had a good faith belief the item had non-infringing uses, then you're not contributing to infringement. Also the penalties tend to be different. So I do think there is a difference. > > The European standard is that the claim must cover a piece of > > technology: a device or method that exhibits a "technical effect". > > And no, I don't have a definition for "technical". > > In the US, it's part of what the courts have read into 35 USC 101's > "new and useful process, machine, manufacture, or composition of > matter". Even a "process" has to be "useful" for a particular > industrial or commercial purpose in order to be patentable. Again, > see In re Alappat. Right, although your "useful" is generally broader than our "technical". State Street Bank's patent was "useful" but not "technical". > > The problem is exactly the same: European patent law does not > > exclude patents on mathematical methods, but only on mathematical > > methods _as such_. Apparently this is not the same thing for the > > people who wrote that law. They may have been wrong, but if this > > is the law, then that's what we have to work with. > > Can you point me to an appellate decision that speaks to this > distinction, even if that isn't dispositive under your system? The law says so: articles 52(2) and (3) EPC. http://www.european-patent-office.org/legal/epc/e/ar52.html The EPO's Board of Appeals has issued several decisions that discuss this article: http://legal.european-patent-office.org/dg3/pdf/t971173ex1.pdf (see page 12 of the PDF, sections 4 & 5) http://legal.european-patent-office.org/dg3/pdf/t950931eu1.pdf (see page 12, similar reasoning for a business method) The Germans issued a somewhat similar decision in 2000: http://swpat.ffii.org/papers/bgh-sprach00/index.en.html Arnoud -- Arnoud Engelfriet, Dutch & European patent attorney - Speaking only for myself Patents, copyright and IPR explained for techies: http://www.iusmentis.com/ -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: libdts patent issue?
Francesco Poli wrote: > On Tue, 19 Jul 2005 15:10:10 +0200 Arnoud Engelfriet wrote: > > More like, the expression in .obj is patented, but the expression > > in .PDF is not. Feel free to publish papers; don't distribute > > devices that execute the algorithm disclosed in those papers. > > And how is literate programming dealt with? No one has brought a case to court yet, so I can't say. If you provide the program loaded into a computer, ready to execute, then the court may likely hold that you infringe. If you publish a printed piece of paper with the program's source, then you likely do not infringe. Arnoud -- Arnoud Engelfriet, Dutch & European patent attorney - Speaking only for myself Patents, copyright and IPR explained for techies: http://www.iusmentis.com/ -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: libdts patent issue?
Summary: I can still find no substantive difference between US and EPO law on software patentability. On 7/20/05, Arnoud Engelfriet <[EMAIL PROTECTED]> wrote: [snip good stuff] > For contributory infringement you need additional evidence. > Contributory infringement is knowingly selling or supplying an item > for which the only use is in connection with a patented invention. > If you had a good faith belief the item had non-infringing uses, > then you're not contributing to infringement. > > Also the penalties tend to be different. So I do think there is > a difference. Fair enough. AIUI the standard here is that it must actually have substantial non-infringing uses and the vendor must be making a good faith effort to sell it for its non-infringing uses rather than using them as cover for a larger, infringing economic opportunity. When I get around to finding precedents from the Federal Circuit that clarify this question, and evaluate the associated penalties, I'll let you know. :-) [snip more good stuff] > The law says so: articles 52(2) and (3) EPC. > http://www.european-patent-office.org/legal/epc/e/ar52.html Understood that that's the statutory basis for the subject matter test (parallel to 35 USC 101), insofar as the EPC constitutes statute rather than treaty. > The EPO's Board of Appeals has issued several decisions that > discuss this article: > http://legal.european-patent-office.org/dg3/pdf/t971173ex1.pdf > (see page 12 of the PDF, sections 4 & 5) I've made a quick pass through this opinion (In re IBM) and find it unsurprising. Methinks the EPO's law clerks need to take some lessons in concise and vivacious writing from some of ours. I also observe that the case law established by previous Boards of Appeal is very much the primary source of citations, and if your courts aren't bound by stare decisis except in the same sort of extraordinary circumstances that ours are, I can't tell the difference. See Sections 7 and 11 of this ruling, and particularly this sentence from 11.3: This means that the question to be decided upon in the present appeal has not been answered earlier by the boards of appeal. Strictly speaking, the cited reasonings may therefore be considered to constitute obiter dicta and not ratio decidendi. As for the content of the ruling, it discusses the "physical" effect silliness quite a bit (and rather inconclusively), then tells the lower court to go back and consider the disputed claims on their merits, evaluating them for non-trivial "technical effect", rather than assume that they describe software "as such". Here is Section 12.2, the heart of the ruling: The present appeal relates to whether the subject matter of claims 20 and 21 is excluded from patentability under Article 52(2) and (3) EPC. The examining division decided that it was. The Board wishes to emphasize that it has decided only that a computer program product is not excluded from patentability under all circumstances. To the Board, those circumstances include the exact wording of the claims at issue. As the wording of present claims 20 and 21 shows, there are various ways in which a claim to a computer program product may be formulated. >From the fact that these claims were refused by the examining division on the basis of the cited passage in the Guidelines, the Board concludes that the examining division did not consider the exact wording of those claims in detail, and acknowledges that, from the examining division's point of view, there was little need to do so. However, now that the Board has decided that not all computer program products are prima facie to be excluded from patentability, a thorough examination of the exact wording of the claims has to be carried out. In order to preserve the appellant's right [not? -- MKE] to have this determined at two instances, the case is remitted to the first instance for further examination of this point. Apart from the substantial difference in writing style, it seems to me to be rather more similar to the opinions in Diehr, Alappat, and AT&T v. Excel than it is different. I have already spoken to AT&T v. Excel; the unanimity of this opinion, together with the denial of certioriari in State Street, suggests that this is now quite settled law in the US. Note that Alappat (1994) was decided en banc, with the "majority" opinion written by the same judge who wrote State Street. Alappat has one of the more complex opinion structures that I have seen: Rich, J., with whom: as to Part I (Jurisdiction): Newman, Lourie, and Rader, JJ., join; Archer, C.J., Nies, and Plager, JJ., concur in conclusion; and Mayer, Michel, Clevenger, and Schall, JJ., dissent; and as to Part II (Merits): Newman, Lourie, Michel, Plager, and Rader, JJ., join; Archer, C.J., and Nies, J., dissent; and Mayer, Clevenger, and Schall, JJ., take no position. This alone tends to be enough to trigger a lot of loose talk about the conformability of the decision with pre-existing l
Re: libdts patent issue?
Arnoud Engelfriet <[EMAIL PROTECTED]> wrote: > Michael K. Edwards wrote: > > Arnoud Engelfriet <[EMAIL PROTECTED]> wrote: > > > The problem is exactly the same: European patent law does not > > > exclude patents on mathematical methods, but only on mathematical > > > methods _as such_. Apparently this is not the same thing for the > > > people who wrote that law. They may have been wrong, but if this > > > is the law, then that's what we have to work with. > > > > Can you point me to an appellate decision that speaks to this > > distinction, even if that isn't dispositive under your system? > > The law says so: articles 52(2) and (3) EPC. > http://www.european-patent-office.org/legal/epc/e/ar52.html If the EPO is an artefact of the EPC, it can't be "the people who wrote that law" so why is EPO reinterpreting the EPC? Is it actually known whether the drafters meant the claimed "you can patent maths as part of a machine" view rather than the "maths is not patentable" from the UK Patent Act? That's something that I've wondered more than a few times. I know that's not the end of it, as pro-patenters seem to be pushing for their interpretation just in case it's not clear. I'm still amazed that the commercial UKPO has such a vital role in informing legislation - I thought the idea of Executive Agencies and Trading Funds was to offer some independence of government. Having them involved is like inviting Unisys (of BurnAllGIFs fame) to help run an information service about Free and Open Source software... -- MJR/slef My Opinion Only: see http://people.debian.org/~mjr/ Please follow http://www.uk.debian.org/MailingLists/#codeofconduct -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: libdts patent issue?
On 20 Jul 2005 23:14:28 GMT, MJ Ray <[EMAIL PROTECTED]> wrote:
> If the EPO is an artefact of the EPC, it can't be "the people
> who wrote that law" so why is EPO reinterpreting the EPC?
> Is it actually known whether the drafters meant the claimed
> "you can patent maths as part of a machine" view rather
> than the "maths is not patentable" from the UK Patent Act?
The EPO's job (particularly that of its Board of Appeals) is to
interpret the EPC in specific cases, helping inventors work through
mostly uncontroversial criteria for a well-crafted patent application
and recommending approval or denial to national patent offices. While
it is not precisely a court, it operates on similar principles and
rules of law, and the rarity (as I understand it) of judicial review
of its rulings (as opposed to those of national patent offices) puts
it in a position of subject matter authority not unlike the US Federal
Circuit's.
The EPC doesn't say "anything with math in it is unpatentable", nor
does it say "anything with math in it, plus any physical object, is
patentable". It says:
Art. 52 (3) The provisions of paragraph 2 [including the prohibitions
on patenting "mathematical methods" and "programs for computers" --
MKE] shall exclude patentability of the subject-matter or activities
referred to in that provision only to the extent to which a European
patent application or European patent relates to such subject-matter
or activities as such.
That "as such" has to be given some practical meaning by the Board of
Appeals in order to provide a concrete and reliable guide to examiners
and applicants as to what does and doesn't pass this "subject matter"
test. It's not just a couple of throw-away words -- it's the entire
point of an entire clause of the Convention. AFAICT there is no
analogue of the House Report on the Copyright Act of 1976 for the
Board to consult as evidence of legislative intent, so they have to
roll their own.
>From the one Board of Appeals decision I've looked at closely
(discussed ad nauseam in my reply to Arnoud), it appears to me that
the Board takes almost exactly the same stance articulated in recent
years by the Federal Circuit in the face of similarly ambiguous
guidance from Congress and the Supremes. Dualism is on the retreat,
processes and machines are on an equal footing, and what makes
something not an abstract idea "as such" is that it be "susceptible of
industrial application" to reliably achieve a particular useful
result.
The In re IBM ruling is not what I would call a model of clarity and
vigor, but it seems to me to stand for the same proposition as AT&T v.
Excel -- the subject matter test is there to distinguish the
theoretical from the applied, not to exclude applied science whose
underlying natural law is that of computational complexity and theory
of approximation instead of chemistry or physics. The "novelty and
inventiveness" criterion is still there to bar inventions that are
trivial in light of the prior art.
In re IBM directs the examining division, not to grant the claims at
issue, but to evaluate them for "further technical effect" as a
question of fact rather than of law. ("Further", here, means that the
examiner should ignore "technical effects" incidental to the
computer's internal mechanisms for executing programs.) The
"technical effect" formula is facially different from the US's
"useful, concrete, tangible result", but they are equally judicially
created, and leave ambiguities regarding the patentability of
inventions at the fringe of the technical world such as automated
securities trading.
The EPC does explicitly exclude "methods for ... doing business", with
the same "as such" caveat; and in this it facially differs from US law
following State Street. It's quite off-topic here so I won't go into
any detail, but In re PBS Partnership (the business methods case
Arnoud cited) denied issuance of a patent on quite strict "absence of
technical effect" grounds, distinguishing it on the facts from other
cases granting patents involving business methods such as Sohei and
Petersson. The latter presumably remain controlling authority under
which the EPO must grant business method patents that meet the
"technical effect" test.
> That's something that I've wondered more than a few times.
> I know that's not the end of it, as pro-patenters seem
> to be pushing for their interpretation just in case it's
> not clear. I'm still amazed that the commercial UKPO has
> such a vital role in informing legislation - I thought the
> idea of Executive Agencies and Trading Funds was to offer
> some independence of government. Having them involved is
> like inviting Unisys (of BurnAllGIFs fame) to help run an
> information service about Free and Open Source software...
Do you really think it's fair to characterize as "pro-patenters"
people who are simply pointing out:
the actual state of the law as embodied in decisions by the
relevant administrative and judicial autho
Re: libdts patent issue?
Michael K. Edwards wrote: > On 7/20/05, Arnoud Engelfriet <[EMAIL PROTECTED]> wrote: > > The law says so: articles 52(2) and (3) EPC. > > http://www.european-patent-office.org/legal/epc/e/ar52.html > > Understood that that's the statutory basis for the subject matter test > (parallel to 35 USC 101), insofar as the EPC constitutes statute > rather than treaty. The EPC is a treaty. All EPC member states have changed their law to be the same as the treaty. Furthermore European patents may only be declared invalid under the conditions set by the EPC. So it's not law, but as close as you can get. > > The EPO's Board of Appeals has issued several decisions that > > discuss this article: > > http://legal.european-patent-office.org/dg3/pdf/t971173ex1.pdf > > (see page 12 of the PDF, sections 4 & 5) > > I've made a quick pass through this opinion (In re IBM) and find it > unsurprising. Methinks the EPO's law clerks need to take some lessons > in concise and vivacious writing from some of ours. I've heard people say "this is what you get when you let Germans write common law decisions." > I also observe > that the case law established by previous Boards of Appeal is very > much the primary source of citations, and if your courts aren't bound > by stare decisis except in the same sort of extraordinary > circumstances that ours are, I can't tell the difference. This is the EPO's internal board of appeals. National courts have no obligation whatsoever to follow the BoA. And even the boards themselves can deviate from earlier decisions, although they have to explain why. And the matter can then be referred to the Enlarged Board (sort of like a rehearing en banc). And yes, they heavily rely on their earlier caselaw. I agree with you that the approach is similar in may aspects to the US system. Partially that's because patent law is heavily harmonized around the world. But still there are differences, especially when it comes to patentable subject matter. The example I gave earlier is http://legal.european-patent-office.org/dg3/pdf/t950931eu1.pdf which is European patent application http://v3.espacenet.com/textdoc?IDX=EP0332770 that was rejected for being a business method as such. And if you read the decision, you'll see that there was no way they were going to allow this application. The corresponding US patent US4750121 was allowed without any such problem. Claim 1 does not even recite "a computer", only several "means" that perform certain steps to a master trust. So in my view, the fact that this kind of difference occurs shows that in the US a lot more is patentable than in Europe. Arnoud -- Arnoud Engelfriet, Dutch & European patent attorney - Speaking only for myself Patents, copyright and IPR explained for techies: http://www.iusmentis.com/ -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: libdts patent issue?
MJ Ray wrote: > Arnoud Engelfriet <[EMAIL PROTECTED]> wrote: > > The law says so: articles 52(2) and (3) EPC. > > http://www.european-patent-office.org/legal/epc/e/ar52.html > > If the EPO is an artefact of the EPC, it can't be "the people > who wrote that law" so why is EPO reinterpreting the EPC? Because the EPC tells them to. The examiners approve or reject patent applications, and if you're not satisfied (or if you try to oppose a patent and lose) you can appeal. In appeal the boards rely on the EPC and the EPC alone to decide whether the case is to be upheld or not. http://www.european-patent-office.org/legal/epc/e/ar23.html And if the boards disagree or have a question of law on the EPC, they refer it to the enlarged board (sort of like an "en banc" hearing in the USA). http://www.european-patent-office.org/legal/epc/e/ar22.html > Is it actually known whether the drafters meant the claimed > "you can patent maths as part of a machine" view rather > than the "maths is not patentable" from the UK Patent Act? Not really. The fact that they didn't exclude "mathematical methods" but only "mathematical methods as such" suggests they had something like that in mind. Otherwise why add "as such"? Keith Beresford [1] has dived into the history of the EPC. In his book I see that the earliest drafts did not even mention computer programs or mathematical methods. And when maths does appear, it's as "mathematical THEORIES as such" (emphasis mine). Later "computer programs" did appear in the list of exclusions, with lots of debate about how far this should go. The most frequently heard objection was that "items which were traditionally patentable should not be excluded merely because they contained computer programs." The British did oppose allowing patents on computer programs, as they felt it was too early to say whether patents or copyright was the best regime. And there was still talk of a separate regime for protection of software. [1] http://swpat.ffii.org/papers/beresford00/index.en.html Arnoud -- Arnoud Engelfriet, Dutch & European patent attorney - Speaking only for myself Patents, copyright and IPR explained for techies: http://www.iusmentis.com/ -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: libdts patent issue?
On 7/21/05, Arnoud Engelfriet <[EMAIL PROTECTED]> wrote: > The example I gave earlier is > http://legal.european-patent-office.org/dg3/pdf/t950931eu1.pdf > which is European patent application > http://v3.espacenet.com/textdoc?IDX=EP0332770 > that was rejected for being a business method as such. And > if you read the decision, you'll see that there was no way they > were going to allow this application. Agreed. As well they shouldn't have. > The corresponding US patent US4750121 was allowed without any such > problem. Claim 1 does not even recite "a computer", only several > "means" that perform certain steps to a master trust. The US examiner was clearly not on the ball, and/or not properly advised as to the significance of State Street. The PBS Partnership patent would, I suspect, be tossed out if litigated in the US, because there is not a reliable engineering relationship between the process being followed and the desired outcome. It's a collection of organizational tactics, which might represent an effective business approach, but it's not an invention. > So in my view, the fact that this kind of difference occurs > shows that in the US a lot more is patentable than in Europe. I think that cuts both ways. Look how long some of the MP3 patents, first issued in Germany, took to prosecute through the US system -- in a couple of cases, a decade or so after the German version had issued. Although I have not yet correlated against the originals, I would not be at all surprised to find that the allowed claims of the US versions are much narrower, judging from the continuations-in-part, abandoned applications, and all that. I have yet to see any evidence that persuades me that what is patentable in the US in some application area is a strict superset of what is patentable in Europe. Cheers, - Michael (IANAL, TINLA)
Re: libdts patent issue?
Michael K. Edwards wrote: > On 7/21/05, Arnoud Engelfriet <[EMAIL PROTECTED]> wrote: > > The corresponding US patent US4750121 was allowed without any such > > problem. Claim 1 does not even recite "a computer", only several > > "means" that perform certain steps to a master trust. > > The US examiner was clearly not on the ball, and/or not properly > advised as to the significance of State Street. US Patent class 705 is full of such patents where the examiner wasn't "on the ball". Only very recently have I seen US office actions where the examiner talks about "technological progress". > The PBS Partnership > patent would, I suspect, be tossed out if litigated in the US, > because there is not a reliable engineering relationship between the > process being followed and the desired outcome. It's a collection of > organizational tactics, which might represent an effective business > approach, but it's not an invention. My understanding was that this kind of patent was perfectly allowable under US law. There's a useful, concrete and tangible result, and you can't throw it out because it's a method of doing business. That's how "everyone" interpreted State Street Bank. > I have yet to see any evidence that persuades me that what is > patentable in the US in some application area is a strict superset of > what is patentable in Europe. It's not. They are two partially overlapping circles. The criteria to determine whether something is statutory are different in both regions. To a certain degree they give you the same outcome, but this is not necessarily true. Arnoud -- Arnoud Engelfriet, Dutch & European patent attorney - Speaking only for myself Patents, copyright and IPR explained for techies: http://www.iusmentis.com/ -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: libdts patent issue?
On 7/21/05, Arnoud Engelfriet <[EMAIL PROTECTED]> wrote: > US Patent class 705 is full of such patents where the examiner > wasn't "on the ball". Only very recently have I seen US office > actions where the examiner talks about "technological progress". Oh, I agree with you completely that this is one of the worst areas of incompetence and/or regulatory capture in the USPTO. But it's hard for me to lay the majority of the blame at the feet of the legislative or judicial branches, based on what I understand to be the public record. Maybe the State Street case was chosen for litigation because it's an extraordinarily convincing presentation of the facts illustrating the principle that an invention can solve a microeconomic problem rather than engineering as such. If so, it has its parallels in EPO case law, including two cases cited by your PBS Partnership example. > My understanding was that this kind of patent was perfectly > allowable under US law. There's a useful, concrete and tangible > result, and you can't throw it out because it's a method of > doing business. That's how "everyone" interpreted State Street Bank. There still needs to be an authentic engineering relationship between the claimed process and a tangible outcome, which I just don't see in PBS Partnership. The State Street opinion didn't change that requirement, and I find it difficult to imagine the Federal Circuit authorizing the acceptance of the disclosure and claims of #4,750,121 as an "invention". Generally speaking, the precedential value of an appellate opinion doesn't extend far beyond its holdings. The only holdings I can find in the State Street opinion are: We hold that declaratory judgment plaintiff State Street was not entitled to the grant of summary judgment of invalidity of the '056 patent under § 101 as a matter of law, because the patent claims are directed to statutory subject matter. OK, that's just a summary of the holding below; no risk of that statement by itself being dispositive in a later case. Today, we hold that the transformation of data, representing discrete dollar amounts, by a machine through a series of mathematical calculations into a final share price, constitutes a practical application of a mathematical algorithm, formula, or calculation, because it produces "a useful, concrete and tangible result" -- a final share price momentarily fixed for recording and reporting purposes and even accepted and relied upon by regulatory authorities and in subsequent trades. That's a statement that Alappat's "useful, concrete, and tangible result" test is applicable when the result is market activity rather than seed-free cotton or goop in a test tube. Now, if you ask me (IANAL, TINLA), this holding is unfounded in Supreme Court authority. The main Supreme Court precedents (Diehr and Benson) cited as support by the State Street court rely on the 1877 Cochrane v. Deener case, cited in part: "A process is a mode of treatment of certain materials to produce a given result. It is an act, or a series of acts, performed upon the subject-matter to be transformed and reduced to a different state or thing." That doesn't sound to me like carte blanche to patent a mode of pricing analysis. Flook isn't on point, and to my eye the claim of support from Chakrabarty is disingenuous at best; I cannot agree that "the Supreme Court has acknowledged that Congress intended § 101 to extend to 'anything under the sun that is made by man" based on a passing quotation from legislative history in the Chakrabarty opinion, nor does a pricing proposal strike me as a "_thing_ made by man". But I see no indication that State Street Bank sought certioriari; so we may have to wait for a later business methods case to see whether the Supremes rip the Federal Circuit a new one for abuse of discretion. By the way, I need to correct something I said earlier. The State Street court did not direct the USPTO to cease the practice of automatically denying patents on business methods. The USPTO had already done so in 1996, presumably in part as a consequence of the Alappat ruling. The State Street opinion comments approvingly on the 1996 editions of the Manual of Patent Examining Procedures and Examination Guidelines for Computer Related Inventions, closing with: "We agree that this is precisely the manner in which this type of claim should be treated. Whether the claims are directed to subject matter within § 101 should not turn on whether the claimed subject matter does 'business' instead of something else." Cheers, - Michael
Re: libdts patent issue?
[EMAIL PROTECTED] wrote: > To me the distinction is clear: you have to add something to the > algorithm before you arrive at patentable matter. You apparently > consider the addition (a computing device with a memory) to be > irrelevant, and hence you don't see a distinction. The addition should be irrelevant because there's no inventiveness involved in the addition. :-P This is why I noted that the US standard, theoretically, allows a new piece of *artwork* plus a generic computer to be patented. The inventiveness is entirely in the artwork, but by adding the computer, you could arrive at "patentable matter" (a "device which displays a person holding some dogs, for the entertainment of the user", for instance). This is obviously bad law. >The European standard is that the claim must cover a piece of >technology: a device or method that exhibits a "technical effect". >And no, I don't have a definition for "technical". The dispute is quite specific here: it is over whether (a) the innovation must be in the technical area. If it doesn't have to be, you get results exactly like the US results: patentable artwork. (b) the technical area must be a patentable area. If it doesn't have to be, you get results exactly like the US results: patentable artwork. > If the end result is something >"technical", then it's patentable. This is no good if there's no innovation in the "technical" area. I "invent" a machine to display a woman holding dogs in a certain manner. It's "inventive" because the woman is holding dogs in an unusual and inventive manner. It's "technical" because it's a machine to display the image. Well, actually it's just a computer plus a unique piece of art, and I've used twisted language to get a patent on something unpatentable. Now nobody can display my artwork -- or indeed any similar artwork -- on a computer without getting a patent license. This is, sadly, precisely analogous to the patents being issued on mathematics today. "A method of solving linear equations"? Come on. The inventiveness is in the mathematics, not in the relation to the physical world. > > There's a reason the FFII preferred standard is that the inventive part of a > > patent must be on some method of manipulating the physical world. That's the > > That's what European patent law also pretends to be. FFII is pushing > a very restrictive definition of what "manipulating the physical > world" means, but otherwise they're completely in line with how > patent law works. I'm not sure whether you caught the key point here. The key point here is that the *inventive* part must be attached to the manipulation of the physical world. Innovative "software patents" generally feature an *uninventive* part which manipulates the physical world -- a generic bit-twiddling machine -- combined with inventive mathematics. > The problem is exactly the same: European patent law does not > exclude patents on mathematical methods, but only on mathematical > methods _as such_. Apparently this is not the same thing for the > people who wrote that law. This is such an overuse of "as such" so as to render the entire list of exclusions from patentability meaningless, and as such (ahem) is invalid under traditional rules of statute construction. By the same argument, only artwork *as such* is unpatentable, and my art patent should work. I could go on and write a similar, equally valid patent relating to putting the artwork on walls. Do you see the problems with this line of argument? The "as such" phrase is presumably intended to allow patents on material which happens to use a mathematical method/artwork/etc., not on material for which the entirety of the inventive portion is the mathematical method/artwork/etc. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: libdts patent issue?
Michael Edwards wrote: > Dualism is on the retreat, > processes and machines are on an equal footing, and what makes > something not an abstract idea "as such" is that it be "susceptible of > industrial application" to reliably achieve a particular useful > result. In practice, that's another distinction without a difference. *sigh* If "industrial" was any sort of limitation at all, it might mean something. As it is -- with "solving a system of linear equations" being an "industrial application" -- this means that EVERY piece of mathematics is "susceptible of industrial application". Without exception. Anyway, this interpretation, although it is apparently the interpretation taken by the Board of Appeals, is again contrary to standard rules of statutory interpretation, and therefore presumably wrong. The requirement that something be "susceptible of industrial application" to be patentable is given explicitly in a different clause of the EPC. If "not a mathematical method as such" is read to mean merely "susceptible of industrial application", then it is redundant surplusage and means nothing. Accordingly, it should be assumed to mean something stronger. Ahem. >the subject matter test is there to distinguish the >theoretical from the applied Nice idea, except that it's being used that way. *Everything* is applied under the current standards and *nothing* is theoretical. The first problem is that all theoretical mathematics is *applicable*, although not *applied* until it is actually applied to a specific problem. "Software patents" are not limited to a particular inventive application (say, playing mp3 files), but apply to future applications using the same mathematical method (say, converting mp3 files to ogg vorbis files). They are patenting the applicable, not the applied. The second (smaller) problem is that applications to mathematics are apparently considered "applied" (viz. "a method to solve a system of linear equations"), which really puts everything in the applied category without exceptions. I don't dispute that European judges are reading the EPC to allow all kinds of crap. I say that it does not allow that kind of crap, and I have the better legal arguments. It is sad that so many judges are incompetent, but there you are. > not to exclude applied science whose > underlying natural law is that of computational complexity and theory > of approximation instead of chemistry or physics. This so-called "applied science" is a field of mathematics. Ahem. Computational complexity theory is *not* a description of the natural world -- you can't test its predictions or do physcial experiments to see how accurate it is -- it's a description of mathematics. > Do you really think it's fair to characterize as "pro-patenters" > people who are simply pointing out: > the actual state of the law as embodied in decisions by the > relevant administrative and judicial authorities; Well, that's fine. I never denied that. Some of these decisions are contrary to statute, of course. Arguing that they are supported by statute is a pro-patent argument. > the difficulty courts had in applying a "physical effect" test > consistently when it was believed to be the prescribed procedure; Pointing out the incompetence of courts is all well and good, but saying that it means they should give up is not. > and the absence of a public policy rationale for denying the same > sort of encouragement to applied researchers (and their financial > backers) in fields where the work is done at a computer keyboard as in > those where it is done at a lab bench? Nobody has "pointed this out", and you've misframed the debate with several false assumptions in this paragraph. * Standard economic theory considers monopolies bad as a public policy matter except when they are (a) natural monopolies, which patents never are, or (b) they can be proven to have benefits outweighing their costs. Amazingly, nobody has ever collected any evidence that any patents have benefits outweighing their costs (except in the pharmaceutical industry, where there are some inconclusive results which might tilt that way). The evidence that "software patents" inhibit innovation, drive up costs, and generally waste economic resources is quite complete. Therefore there is a strong public policy rationale for denying them. The burden of proof is on *you* to show that the incentive of patents is *necessary*, not on us to show that it is unnecessary. * Your argument imagines people "working at keyboards" who are "applied researchers" much like those working at "lab benches". But what are they actually doing? Software "engineering" is not engineering, as anyone who compares the two carefully will realize. Research in the fundamentals of engineering is research in the natural sciences, but practical engineering requires a lot of additional skills too. "Research" in software *is* pure
Re: libdts patent issue?
Nathanael Nerode wrote: > [EMAIL PROTECTED] wrote: > > To me the distinction is clear: you have to add something to the > > algorithm before you arrive at patentable matter. You apparently > > consider the addition (a computing device with a memory) to be > > irrelevant, and hence you don't see a distinction. > > The addition should be irrelevant because there's no inventiveness > involved in the addition. :-P Inventiveness has nothing to do with whether something is statutory subject matter. Today, the wheel is patentable in principle, but you will never get that patent because the wheel is known. A wheel of exactly 38.25" diameter is also statutory subject matter, it's (probably) new but such a specific choice of diameter is obvious. > This is why I noted that the US > standard, theoretically, allows a new piece of *artwork* plus a > generic computer to be patented. US patent law would say that this "computer with artwork" is statutory subject matter. It'd probably be rejected for lack of inventiveness though, as computers that show artwork are well-known so it would be obvious to a skilled person to pick yet another piece of artwork and display it. > The dispute is quite specific here: it is over whether > (a) the innovation must be in the technical area. If it doesn't have to be, > you get results exactly like the US results: patentable artwork. > (b) the technical area must be a patentable area. If it doesn't have to be, > you get results exactly like the US results: patentable artwork. See above. The claim _as a whole_ must be in a patentable area. In Europe that means it must be in a technical area. And in Europe, the _innovation_ must be in a technical area as well. For instance, "A method of selling items, involving giving a 10% discount if you buy more than five items at once" is not in a patentable area in Europe. That's what we would call a pure business method. (I don't know if it is statutory in the USA). I can rewrite that claim to "A cash register that is configured to count the number of items rung up during one transaction and to give a 10% discount if the counted number exceeds five." This is _statutory subject matter_ in Europe, because cash registers are technical devices. Same for the USA. But the application will be rejected because it does not involve an inventive step. The innovation (giving the discount) is purely a business method, which by definition cannot be inventive. > This is no good if there's no innovation in the "technical" area. I "invent" > a machine to display a woman holding dogs in a certain manner. It's > "inventive" because the woman is holding dogs in an unusual and inventive > manner. It's "technical" because it's a machine to display the image. In Europe, we'd say "yes, this is technical -it's a machine- but there's no inventiveness involved in displaying yet another picture." So it's just a trivial variation on a known machine. If I take a washing machine and paint it purple, would you say it's no longer a machine? > > That's what European patent law also pretends to be. FFII is pushing > > a very restrictive definition of what "manipulating the physical > > world" means, but otherwise they're completely in line with how > > patent law works. > I'm not sure whether you caught the key point here. The key point here is > that the *inventive* part must be attached to the manipulation of the > physical world. There is a difference in approach here. FFII (and apparently you as well) judges whether something is _statutory_ by looking only at the contribution over the prior art. European patent laws by and large determine this question by looking at the _claim as a whole_. A washing machine is a washing machine is a washing machine. That kind of machine is statutory subject matter. If the contribution you claim to have invented is no _technical contribution_ then you have an obvious variation on a washing machine. FFII would say in this example "the invention is not statutory." The EPO would say "the invention is statutory but trivial". This is the difference between 'Kerntheorie' and 'whole claim approach'. > Innovative "software patents" generally feature an > *uninventive* part which manipulates the physical world -- a generic > bit-twiddling machine -- combined with inventive mathematics. The inventive step is part maths and part practical (technical) application of the mathematics. A formula by itself is not a patentable contribution. "A method of recognizing spoken words, comprising using new formula X to recognize the words" provides a technical contribution: the _use_of_ the formula to manipulate the physical world. > > The problem is exactly the same: European patent law does not > > exclude patents on mathematical methods, but only on mathematical > > methods _as such_. Apparently this is not the same thing for the > > people who wrote that law. > This is such an overuse of "as such" so as to render the entire l
Re: libdts patent issue?
On Sat, Jul 16, 2005 at 01:25:27PM -0700, Michael K. Edwards wrote: > On 7/16/05, Diego Biurrun <[EMAIL PROTECTED]> wrote: > > Please remember that this is my answer to your question of what _I_ would > > do, I didn't say what Debian should do. > > [...} > > But you're telling me you won't at least call your lawyer? I never said something to this effect, please don't misrepresent my words. > > Reality upstream is that the hosting university could not be bothered to > > fight the patent, even though the consensus was that the patent is > > invalid. > > What consensus is that? Consensus among qualified commentators (of > which I am not one)? That I really, really, doubt. Consensus among VideoLAN developers and the university's lawyer. > > It seems that at least I have read it, while you haven't... > > Au contraire, mon frère. I have read it, and thought about it, and > spot-checked some of its "facts", and I stand by my assessment that it > is rubbish. I advise you again to ask yourself: is this a > dispassionate, scholarly analysis or is it a polemic that, at best, > uses outside evidence to exhort rather than to inform? I _have_ thought about it and consider your continued suggestion that I have not insulting. People arrive at different conclusions without being complete idiots. C'est la vie, mon ami. I stand by my assessment that it is well thought-out, much better than what I have read from the other side of the fence. Feel free to point me at your spot-checks and texts on the subject that you consider "dispassionate, scholarly analysis". In an attempt to keep the discussion ontopic for this thread and this mailing list I suggest that we continue this part of the discussion in private, should you wish to answer. > > Given that more and more DVDs come with DTS audio, this software is > > useful whether you like the specification or not. It's also not > > half-finished, it works just fine. > > Even if it were true that "more and more DVDs come with DTS audio" -- > I don't think I've ever actually seen one that wouldn't play on an > AC-3 + stereo player, but I'm sure it varies radically by region and > genre They generally come with an additional AC3 track, that's why they still work for you. > -- what does that have to do with whether it is the sort of > thing _you_ want to go to the wall for? I might go to the wall for > Larry Flynt's right to publish material that may or may not differ > from my personal taste in literature, but would I go there for someone > who (IMHO) is less exercising free speech than circumventing his > society's prevailing bargain of temporary monopoly on a design in > exchange for permanent documentation of how it works -- and not > putting his own money and/or liberty on the line? Basically I'm willing to go to the wall for free software and I consider software patents the biggest threat to free software at this moment in time. The prevailing bargain might have to be revisited every once in a while to see if it is (still) a good one. Seeing how patent systems are coming under fire in recent years around the world I appear not be entirely alone with that opinion. > > Michael, let's try to keep this ontopic please. > > What could possibly be more on-topic for debian-legal than the > discussion of a strategy for dealing with a foreseeable legal problem > for Debian and its distributors and users? This is ontopic. Discussing the pros and cons of software patents in general is not. Besides, I doubt you will find much support for your position that (some) software should be patentable in this forum. Diego -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: libdts patent issue?
On Wed, Jul 20, 2005 at 05:54:40PM -0700, Michael K. Edwards wrote: > > Do you really think it's fair to characterize as "pro-patenters" > people who are simply pointing out: > the absence of a public policy rationale for denying the same > sort of encouragement to applied researchers (and their financial > backers) in fields where the work is done at a computer keyboard as in > those where it is done at a lab bench? You're turning things on the head. Patents are a restriction and as such _you_ need to present a compelling rationale to extend this restriction to other fields of endeavour, not the other way around. And to answer your question: Yes, it's fair. The argument you just presented appears in every pro-patenter's portfolio, btw. Diego -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Re: libdts patent issue?
Arnoud Engelfriet wrote: > If you provide the program loaded into a computer, ready to execute, > then the court may likely hold that you infringe. If you publish > a printed piece of paper with the program's source, then you likely > do not infringe. Like I said somewhere, non-tech-savvy judges making a very questionable "dead-tree vs. electronic" distinction. What if I provide it on diskette, ready to do whatever with? -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Re: libdts patent issue?
Nathanael Nerode wrote: > Arnoud Engelfriet wrote: > > If you provide the program loaded into a computer, ready to execute, > > then the court may likely hold that you infringe. If you publish > > a printed piece of paper with the program's source, then you likely > > do not infringe. > > Like I said somewhere, non-tech-savvy judges making a very questionable > "dead-tree vs. electronic" distinction. What if I provide it on diskette, > ready to do whatever with? There's no caselaw on this point. I doubt anyone would bring a case if you were distributing diskettes to give people electronic versions of the material of your speech or whatever. If you were selling the diskettes with the intent that people can now get the patented feature on their computer, then it gets tricky. I honestly don't know. Arnoud -- Arnoud Engelfriet, Dutch & European patent attorney - Speaking only for myself Patents, copyright and IPR explained for techies: http://www.iusmentis.com/ -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
