Microsoft v. AT&T Oral Arguments:

AT&T sued Microsoft for patent infringement and wants to collect damages for foreign sales under Section 271(f) of the patent act. That section calls for infringement of a US patent if “components” of a patented invention are supplied abroad to be combined as in a would-be infringing manner. Microsoft ships software code abroad that is then copied before being installed in an infringing manner. Microsoft argues that its actions do not satisfy 271(f) because (1) software code itself (separated from a physical disk) cannot be a component because it is not physical and (2) only copies of the code are installed. Notes from the transcript:

Patentable Subject Matter:

MR. OLSON [For Microsoft]: The ‘580 patent is a program, as I understand it, that’s married to a computer, has to be married to a computer in order to be patented.
JUSTICE SCALIA: You can’t patent, you know, on-off, on-off code in the abstract, can you?
MR. OLSON: That’s correct, Justice Scalia.
JUSTICE SCALIA: There needs to be a device.
MR. OLSON: An idea or a principle, two plus two equals four can’t be patented. It has to be put together with a machine and made into a usable device.

JUSTICE BREYER: I take it that we are operating under the assumption that software is patentable? We have never held that in this Court, have we?
MR. JOSEFFER: No, but . . . that’s not relevant here . . .

JUSTICE STEVENS: Your time is up, but I want to ask you one yes or no question. In your view is software patentable?
MR. JOSEFFER: Standing alone in and of itself, no.

. . .

MR. WAXMAN [For AT&T]: The code is not patentable.

What is a Component:

MR. JOSEFFER [For the Government]: I think the most important point here is that the components of patented inventions do not have to be patentable. Many patented inventions are comprised of a bunch of parts where the parts themselves would not be patentable because say they were standard off the shelf parts.

MR. OLSON: Physical things must be components under 271(f) because they must be supplied from somewhere. Ideas have no physical from. They’re in the air. The words used, “supplied from” tells us that it must be a physical thing combined with. Ideas don’t combine with physical things to make a patented invention. Physical things do.

Scalia’s Jokes:

JUSTICE SCALIA: I hope we can continue calling it the golden disk. It has a certain Scheherazade quality that really adds a lot of interest to this case.

Scared that AT&T will Control our Thoughts:

JUSTICE BREYER: But I then would be quite frightened of deciding for you and discovering that all over the world there are vast numbers of inventions that really can be thought of in the same way that you’re thinking of this one, and suddenly all kinds of transmissions of information themselves and alone become components. So I’m asking you, is there any outside the computer field analogous instance where the transmission of information has itself been viewed as the transmission of a component?

Notes:

79 thoughts on “Microsoft v. AT&T Oral Arguments:

  1. WOW. I read this thread last night and in all honesty I couldn’t stop laughing. I am a former programmer who has turned…lawyer?…no…psychologist. I am just so thankful that that our founders had the foresight to put multiple people on the court and require a majority because this is a classic example of where the choice is not between good or bad, silly or serious, law or physics, but rather a choice between positions that are both equally reasonable. Instead of calling people names I found myself thinking, “well, that is true. And, yes, that is true too; uh huh, I think that is right also” even though all the posters were disagreeing with each other.

    How about this: there is no right answer in the abstract. There is only the answer that the SC will come up with and everybody else will have to live with until either Congress changes the law or the SC changes its mind.

  2. Tom Kulaga wrote, “After considering the technology, a Golden Master could be considered to be ‘especially adapted’ for use in the invention. Maybe ‘especially adapted’ includes replicating the Golden Master. After all, that is the purpose of the Master.”

    If the claims recited a master disc replication system, then I’d agree with you that “especially adapted” would apply since the master disc would be “in the invention.” However, that’s not what AT&T claimed and it’s not the Court’s job to adapt the statute to AT&T’s choice of claims.

  3. “Technology changes things, creates new capabilities and new problems. My point is that analogies will always be imperfect when used to describe new technology.”

    There is some truth to it. Of course, statutes will always be imperfect when used to describe new technology.

    Here’s a question: were electromagnetic signals considered “new” technology in 1984 when 271(f) was drafted? What about software? Was anybody contemplating that computers would be important way back then?

  4. Stepback wrote:
    “Exercise: Using nothing but the English language describe with exacting detail what the experience is of biting into an apple for someone who has never done it.”

    I’ll start that as soon as I finish studying for my next test. I understand that there is one work problem on it and I have 30 minutes. The problem is “Define the universe, and give three examples.”

    When I began my legal career after working as an engineer for 20 years, I clerked for two state appellate judges. The attorneys that provided the most informative briefs that described the issues in a clear manner often prevailed, particularly when technology was involved. I remember assisting in one high tech case of first impression, and what was most helpful was a description, or tutorial, of the technology involved. The judges appreciated that the decision had to rest on its merits and not a bad analogy. Also, I observed when clerking that oral arguments often provided amusement, but rarely swayed the court.

  5. Microsoft did not steal or copy code that was written by AT&T, but rather wrote code that when triggered by a computer, does what AT&T’s ‘580 patent claims and, therefore, infringes the patent. So, by definition, the Microsoft code IS a necessary component needed to practice the ‘580 patent on a computer running Microsoft Windows.

    Microsoft, might have only sent a single copy of the code, but their intent (as demonstrated by their receipt of licensing fees for all the copies that were eventually installed on machines abroad) was to sell multiple copies of the software component.

    Please stop with all the analogies – just read about software and patents on wikipedia.

  6. >>”The reliance on analogies is amazing and amusing.”

    Tom Kulaga,

    Now we are getting into the fun stuff, namely, understanding how the human brain operates and how that leads to observed behavior by various actors.

    You are right. If I’m an elderly judge and I’m not out there, blogging on the Internet, playing the latest video game machines, writing my own software scripts, etc.; then I don’t have much of a grasp as to what this “software” stuff is all about. I have these stuffed shirt lawyers coming into my courtroom and throwing confusing words at me but I still can’t get a handle on it. I need an anchor. I need a connection to something old and familiar so that I can feel I’m on solid ground rather than spinning into a dizzying abyss of unfamiliar terminologies.

    So of course I’m going to demand that the lawyers provide me with an analogy. As judge, I can “demand” that they name that part of the elephant to which software is most akin. Is software more like the trunk of the elephant or more like the tail? I demand that you answer the question as posed with no “if or buts”. If you dare to “if or but” me, I will take it out on you when I write my decision.

    Exercise: Using nothing but the English language describe with exacting detail what the experience is of biting into an apple for someone who has never done it.

  7. SF wrote:
    “ee4law wrote, “This is distinguishable from molds and blueprints, since molds and blueprints are not identical functional copies of the components to be used in assembly.”

    That’s a distinction that isn’t meaningful.”

    The reliance on analogies is amazing and amusing. Technology advances. With new technology, it is not always possible to analogize to previous technology and have a good fit. I agree, partially, with ee4law that blueprints are different than the Golden Master because, unlike blueprints, the Golden Master can be inserted in a computer to make the computer do something. (Get with the times, folks. “Blueprints” have been obsolete for years. I saw some 30 years ago as a young engineer. But, they have pretty much disappeared as an obsolete drawing reproduction method. Most engineers I know are familiar with “prints” and “drawings,” and use computers, not paper, vellum, or mylar, to work with them.) We cannot ignore a distinction just because it doesn’t fit into a particular analogy.

    To further complicate the analogy, drawings (as CADD files) can be fed to a CNC machine. The CADD file causes the CNC machine to produce a product as represented in the drawing. Although a CADD file is not an identical functional copy, it can create one through another machine. But, is a CADD file the same as a blueprint of 50 years ago? Can a CADD file be analogized to a Golden Master? Who cares?

    Technology changes things, creates new capabilities and new problems. My point is that analogies will always be imperfect when used to describe new technology. Maybe instead of trying to make analogies work, we need to gain a better understanding of the new technology.

    After considering the technology, a Golden Master could be considered to be “especially adapted” for use in the invention. Maybe “especially adapted” includes replicating the Golden Master. After all, that is the purpose of the Master.

    I also wish that many of the Supreme Court Justices weren’t so old that they still think blueprints are commonly used. I mean, cars still had crank starters when several of them were born.

  8. Malcolm,

    It appears your punditry is reduced to making an emotional appeal to the football and NASCAR crowd.

    Why don’t you address the substantive issues? Why do they call it IT (Informational Technologies)? Is software an informational component of a computer configured to execute the NetMeeting process? Yes or no? Who caused the informational component to be supplied into the computer?

    Hail Mary prayers are not admissible in a secular court of law. You may be scoring some high fives with your buddies at the drinking pub. However that is a different kind of “bar” than the one that comprises lawyers debating over legal issues. You are posting to the wrong site. Try Sportingly-O.
    link to diamondinfoanalytics.com

  9. AT&T threw a Hail Mary. Unfortunately, Justice Breyer caught the ball, ran it back for a touchdown, spiked it, then took a magic marker out of his back pocket and autographed the goal post.

  10. ee4law wrote, “This is distinguishable from molds and blueprints, since molds and blueprints are not identical functional copies of the components to be used in assembly.”

    That’s a distinction that isn’t meaningful.

    The analogy works because molds, blueprints and the master disc never end up in a patented system. In particular, the blueprints and the master disc are referred to in order to create a patented system, and the mold would be used to create patented molded product. In the absence of claims to the replication system, neither the mold, nor the master disc form part of a patented invention — they’re merely tools that could be used to create a patented invention. But that’s not what the statute covers. The statute covers components “especially made or especially adapted for use **in the invention**” — not other things that could be used to create those patented inventions or components.

  11. >>”so totally against the facts of our history from Thomas Paine onwards that it could only have fallen out of the mouth of a lawyer.”

    soft.viz,

    Actually buddy, we “feel” for you which is why we are responding. If we didn’t “care”, you would be getting the stone wall of silence treatment.

    Truth is, you do not live in a “democratic” country. You live in a representative republic. You vote for an electorate college, they pick a president, the president nominates the Supreme Court judges and the judges decide on appeal cases such as this one.

    Even if one or more of the SCt. justices were to write an opinion that includes a statement like, “software is not patentable”, that would be dicta and pure gobbledygook. You will be hard pressed to find a patent that actually concludes with: I claim: 1. “Software” comprising A, B and C.

  12. May I add that voicing the idea that the arguments of citizens should, by default and except in exceptional cases, be excluded from discourse about the affairs that determine the details of their very lives is simultaneously nauseating, condescending, superior, and so totally against the facts of our history from Thomas Paine onwards that it could only have fallen out of the mouth of a lawyer.

  13. anonymous with link-
    there is nothing there that says MS challenged the validity of the patent. I am begining to think you are bluffing, but I am willing to be proven wrong. Cut N Paste anything to the the contrary with a link to the page. The briefs, which I read earlier, are all about whether an abstract thing can or cannot be a component.

    stepback said:
    The thing that happened yesterday at the US Supreme Court was not a “trial”. No one is going to determine what “software” is, what a “software patent” is, or whether such things should be legally allowed or not.

    Dennis Crouch said on Dec 19, 2006:
    It would be odd for the Court to decide the 101 issue in this case after dismissing LabCorp earlier this year. However, I expect at least one concurring opinion supporting the ideas in this brief.

    So the Court IS free to decide EXACTLY what we are talking about if they so choose.

  14. Look, rebut what is said and never mind the your generalizations of the characters of the people who are saying it. Your argument is what is known as an “argument from authority”, i.e. “I am an authority, you are not, therefore, I am right.” I never use that tactic when rebutting because it’s invalid, and well known to be so. If I see an assertion about technical issue, I treat the assertion directly. Argument by authority in a non argument.

    On the one hand you’re telling participants that they their point of view is naive because they are not lawyers and on the other you also want us to accept that the Courts opinion is naive because they ARE lawyers who don’t understand technology. Isn’t that convenient? Apparently both the Court and the public are desperately in need of that most special breed, the patent lawyers to handle all this thinking *stuff*. In fact, your position is so exalted over both the parties that you need not even argue the issue and offer no argument.

    Oh, you’ll go far before any court you appear before.

  15. to Anonymous with MS district court info:

    link please?

    to Hawkeye:
    It’s annoying (but also informative) that the substantive portions of my arguments are not being rebutted. Substantive == source would have been non infringing in MS v ATT, mere effects are being patented, sw patents run contrary to the very purpose of issuing patents in the first place , computer scientists nearly universally reject software patents etc etc….

  16. There are a lot of technical people commenting here today, many of them apparently being “intrigued” by how the suits, the lawyers, parse out their legal mumbo jumbo and come to a determination of what is “right and wrong” (legal or not legal).

    Let me say this as an ex-engineer and a currently practicing “shyster”. It is all too tempting to take a quick peek behind the curtain and decide instantly that you know what is going on because you see what you see and you believe your lying eyes will always tell you the “truth”.

    The thing that happened yesterday at the US Supreme Court was not a “trial”. No one is going to determine what “software” is, what a “software patent” is, or whether such things should be legally allowed or not.

    What happened yesterday was part of a post-trial appeal process. The legal question presented was that of statutory construction: what do the words in 35 USC 271 (f) mean? More specifically what is the scope and meaning of “component”? And also what is the scope and meaning of “supplying or causing to be supplied”? All the rest is an elaborate circus sideshow.

    It is the same for the law as it is for technology. You need to spend many years studying the law before you can get a grasp of what is going on and how the game is played. It was probably painfully obvious to many a technical person who read the SCt. transcripts that the Justices do not have a clue about how computers work. They think that electrons dance on the hard disks. They circle blindly around the elephant trying to find a convenient analogy –it’s like a snake, it’s like a wall, no it’s a mold, I feel a blueprint here, definitely a blueprint.

    Please don’t make fun of them. You sound equally clueless when arguing “law”.

    The real culprit is the way in which our society becomes ever more “specialized” and out of touch with the interconnections between various fields of endeavor. CS majors laugh at hardware engineers. Hardware engineers laugh at lawyers. Lawyers laugh at judges (when not in court and politely genuflecting in front of them). Judges laugh at the ignoramous software coders who sit all day in their Dilbert cubes grinding out mindless and “abstract” code. And on it goes.

    I do not mean to offend anyone. I’m just asking you to step back and see the full extent of the circus. We are all clowns, each of us sure that our area of speciality is the supreme one.

  17. “First MS did not try to show such a thing”

    This is just getting ridiculous here…

    People without any knowledge whatsoever about the subject being discussed are projecting their false opinions on the rest of ignorant public… just like SCOTUS oral argument…

    MS had no part in developing all those vocoder techniques. They simply didn’t exist at the time…

    Read IEEE Oral History for some education on the history of linear prediction for speech:
    link to ieee.org

    (Page 54)

  18. I expected Waxman to focus his arguments on why the court should not interpret the words “component” and “supplying” by their literal meaning, but should rather read them in a broader sense, with the “intent of Congress” in the backdrop.

    Instead, Waxman was ridiculously arguing against basic axioms of physics.

    The real issue is, what did Congress intend?

    271(f) is a broadened implementation of barring exportation of U.S. made, patented products. The logic behind 271(f) is to include “parts” that are deliberately “made” for use in infringing products broad.

    Software is a “part” that is “made” for deliberate use in hardware.

    Congress inadvertently used the syntactical phrase of “supplying components”, but their intent is unmistaken.

  19. >First MS did not try to show such a thing, which is
    >different than saying they tried and failed. As far
    >as has been shown, they have only claimed they are
    >not liable for infringing on AT&T’s licensing rights
    >because it only supplied the golden disk.
    >
    >If you have information to the contrary, well, do
    >share.

    They tried, and they failed. The case was fought by Microsoft in the District Court, and they capitulated, agreeing to dispute only the foreign sales.

  20. you said : “Microsoft, the most powerful company in the world, the employer of literally thousands of programmers, was unable to show such a thing.”

    First MS did not try to show such a thing, which is different than saying they tried and failed. As far as has been shown, they have only claimed they are not liable for infringing on AT&T’s licensing rights because it only supplied the golden disk.

    If you have information to the contrary, well, do share.

    Your (faulty) assumption is that IF they could have THEN they would have.

    So your perhaps your classes was served from your absences,a s often as they happened because that is a logical fallacy of the following form:

    If Sam was rich, he’d drive a great car
    Sam drives a rotten car
    therefore Sam is not rich.

    MS has it’s own reasons for not going after SW patents, mainly, they want to use them against the rest of the world as Ballmer recently stated. (lawyers are all they have).

    You OTHER FAULTY argument was to assume the consequent- specifically, if ATT got a patent, THEN it MUST have been valid ion the way you described. But that is exactly what is being argued in the first place, so you cannot simply declare it to be so by default then exit stage left. You cannot raise your won hand, declare you’ve won and walk off the court.

    Well you can, but that’s not the way life is played.

    AS to your assertion that “it’s a problem with developers generally” that we do not understand law- so sayeth you. So what?

  21. ee4law

    good points. Great post.

    There’s a forest from the trees issue going on here as I will try to illustrate.

    Here is a potential (though, trivially, not actual in this case) problem with your argument; you said:

    “This is distinguishable from molds and blueprints, since molds and blueprints are not identical functional copies of the components to be used in assembly.”

    MS could have as well, and with the same effect, (specifically the effect of causing the distro of the same or similar program) have transmitted source code and not object code. The subsequent creation of the object code via compilation by machines overseas of the source would have resulted in exactly the same situation, but a finding of non-infringement.

    So how can MS be guilty in the first instance (the actual case) and not guilty in the second, the hypo? It’s a distinction without a (practical) difference.

    This being so, we may as well run the resulting equality the other way and say, since they were not guilty, all agree, in the second case, they are not guilty either in the first.

    If we don’t all agree because source is only “good for one thing” )making object code) , I will be happy to raise the level of abstraction of “that which was provided” until all hold-outs say uncle. Let MS ship *very* highly detailed (i.e. determinate) specifications which are turned into source then object. And so on.

    So if blueprints and molds are exculpatory, what MS is saying is, look, the Scherazadian disk, mea culpa. The others? They’re “especially good renditions” made created, cast from, the original, but of course they are NOT the original because the original is here, your Honor in my right hand before you while the rest are elsewhere and as we all know, matter cannot be in two places at one time.

    So you have the following problems as I see it: one, there is an essential hollowness to your argument which cannot address the underlying issue and, simultaneously, by everyone’s admission, a short and easily traversed path backwards into non-infringement.

    This case is not really, therefore, as clear cut as excellent presentations such as your would, at first blush, make it seem.

  22. >OK, you wrote those words. So you’ve got some reason
    >you aren’t familiar with the fact that software is
    >mathematics. Perhaps you missed class that day / >month /year.

    Well, if I missed class that day, then my computer science students got a day off.

    I agree with others here that the question at issue is not whether software should be patentable. So, I will make one final remark on that subject, and not return to it.

    The “programmer” stated:

    >This is WORLDS away from what ATT is claiming and
    >what all software patents claim.

    This comment reveals a fundamental lack of understanding of patent law in general, and the AT&T patent, in particular. (this problem, by the way, is endemic among many who call themselves programmers) The AT&T patent covers a very specific way to do what it does, and that “way” is very precisely set out in the language of the claims. If anyone can show that those claims cover something that had already been done, or that the combination of elements in the claims would have been obvious to one skilled in the art, then that person is fully empowered by the law to challenge that patent, by filing for a reexam. Microsoft, the most powerful company in the world, the employer of literally thousands of programmers, was unable to show such a thing.

    This fallacy that has become popular, that somehow “bad” patents managed to “slip through the cracks” at the PTO and allowed the patent holders to do something unfair to large and powerful infringers like Microsoft, glosses over the fact that patents like AT&T’s managed to withstand years of attack by those who would be best equipped to attack them, yet they still stand.

  23. “the recipe or manual cannot be patented. It may be copyrighted, just like software, but in all three cases, they are merely instructions to be followed.”

    More clueless remarks….

    The very first U.S. patent was issued in 1790 to Samuel Hopkins on the process for making potash – sounds like a recipe to me…

  24. “Software and business method patents have been rejected not once not twice but three times by the EU”

    Trying to distort facts over and over again ?

    EU failed to harmonize legislation on Computer Implemented inventions, just that, nothing more.
    Some MPs got scared of those idiots wearing “NO software patents” t-shirts who were running in circles around a parliament and shouting “freedom”

    Freaking RMS-lovers…

  25. In the computer controlled robotic arm example, the human gets a manual for how to operate the computer, and the computer gets a manual (software) for how to operate the arm. Neither manual is or should be patentable, but both are and should be copyrightable.

  26. Computer programs (software)are mathematical functions. They are all logic, both on the digital level (1, 0), and the human level (ex: C++ or C# source code).

    The fact the computer is rather useless without software does not make software any greater or more than it is without the computer.

    Going back to the recipe analogy, or even altering slightly and using an instruction manual (assembling a swing set), the recipe or manual cannot be patented. It may be copyrighted, just like software, but in all three cases, they are merely instructions to be followed. Recipes when followed result in food, manuals when followed result in something like a swing set, software when followed results in data processed (which can be any number of things). The result in many cases is patentable, but the process to get there is not.

    Similarly, if you have a computer controlled robotic arm (such as exist in many factories), the system of computer controlled arm may be patentable, but the instructions to operate it, either by human control or computer control are not.

    And, for the record, I do know programming.

  27. Particularly relevant was the “Posted by: anonymous | Feb 21, 2007 at 10:21 PM” comments:

    As framed, I would offer the following;

    First, both sides appear to agree that exporting the Golden Disk is “supplying” under 35 USC 271(f). However, I have to agree with Microsoft, if the Golden Master is never used in a computer then it is not assembled into the invention and thus non-infringing, “standing alone”.

    Second, does replication fit the meaning of “or causing to be supplied”? My understanding is that without the Golden Disk, the replication could not occur. Further, the requested replication appears caused by the supplier of the Golden Disk, and the resulting physical object is practically an identical functional copy of the Golden Master. Additionally, no significant transformation occurs since the Golden Master and the Copy are functionally identical.

    This is distinguishable from molds and blueprints, since molds and blueprints are not identical functional copies of the components to be used in assembly.

    On the second question I have to agree with ATT, because the ultimately assembled components are functional replicas of the supplied Golden Master, who’s sole purpose is to make replicated components, (because as Microsoft asserts, “never used in a computer”). Moreover, since the Golden Master has only one purpose, then the supply of the ultimately assembled components was caused by Microsoft.

    A recap in light of 35 USC 271(f) (2); “Whoever without authority supplies or causes to be supplied in or from the United States any component of a patented invention that is especially made or especially adapted for use in the invention and not a staple article or commodity of commerce suitable for substantial non-infringing use,

    where such component is uncombined in whole or in part,

    knowing that such component is so made or adapted and intending that such component will be combined outside of the United States in a manner that would infringe the patent if such combination occurred within the United States, shall be liable as an infringer.” [separated at commas for clarity]

    I would argue that the result of exporting the asserted single purpose Golden Master and causing replication for assembly is pertinent to the last clause of 271(f)(2), that of “knowing” the Golden Disk is “adapted and” [Microsoft] “intending that such component will be combined”

    I rely here on “combined” to incorporate the aspects of replication that have been the focal point for many. Microsoft argues that the replication is part of the assembly process, then based on Microsoft’s assertion “combining’ the Golden Master in the same way within the U.S. would have resulted in infringement. The Golden Master replication would occur first, then replicas assembled into the patented invention.

    Therefore, based on Microsoft’s own position, and in light of 271(f)(2), the ATT position appears more persuasive.

  28. There you go again, making a false assumption. I’m a programmer and computer scientist, not a lawyer.

    OK, you wrote those words. So you’ve got some reason you aren’t familiar with the fact that software is mathematics. Perhaps you missed class that day / month /year.

    As to your analogy, it could ONLY apply to the actual object code itself. Supposedly, to pass USPTO muster *snicker* there’s a particular WAY it achieves its end that is sufficiently different from all ways previous that its inventor needs to be awarded a 20 year monopoly on its use in order to induce the Nozzle Genius to share with the world his brilliant inventive step, lest the world be forever bereft of the high technology involved, (and here I am only reminding you of the supposed underlying social and rational basis for granting patents in the first place, which is hugely ignored in practice because the CAFC, in open defiance of its mandate to serve the public good has degenerated into little more than a bunch of lawyers assuring fellow lawyers of their continued prosperity at the expense of larger society, but I digress….)

    So we have The Amazing Nozzle. The IP of the Amazing Nozzle does NOT cover every way to do what it does, it covers a specific way. If I changed the steps and achieved the same ends, I would get around the patent. This is WORLDS away from what ATT is claiming and what all software patents claim. ATT is saying “who cares how it’s done, (the code and the corresponding object code being “how it’s done”) it’s that it was done at all”.

    There is no way out or round or through this for the pro-patent crowd. You end up arguing against your own position because patenting ideas is inherently paradoxical, unenforceable and general legal quagmire. Software patents are demonstrably detrimental to the progress of technology. Software patents are universally rejected by all but a tiny tiny minority of software developers and scientists. Software patents have no firm basis in law, except as appears in CASE LAW well know to be the weakest form of law. Software and business method patents have been rejected not once not twice but three times by the EU. Software patents are NOT decided law.

  29. “Most people think of a program as a thing, a file as a thing, a web page as a thing, etc.”

    Most people think of ideas as things, too. I mean, you can count ideas, right? So they must be a thing. And laws of nature are things, too.

    But ideas and laws of nature aren’t patentable. Not every “thing” is patentable (in spite of Supreme Court dicta to the contrary).

    “It appears as though the Justices were just stopping in for a chat at a cafe, rather than asking well-informed questions based on their thorough understanding of the posture and facts of the case. I am very disappointed.”

    My expectations were low, based on past experience. For true disappointment, I await Supreme Court decisions on non-patent cases.

  30. >People who haven’t studied CS deeply don’t know
    >that, but then again, those same people shouldn’t
    >have a hand in writing laws that effect CS either.

    There you go again, making a false assumption. I’m a programmer and computer scientist, not a lawyer.

    But, on the subject of false assumptions, Justice Ginsberg asked: “Can a mold be patented?” This illustrates another false assumption commonly made about software, that it is like a “mold” because it imposes a structure on an otherwise unstructured collection of elements.

    Just because it *molds* doesn’t mean it *is* a mold. A good analogy can be found in the field of hydrodynamics and nozzle design. A nozzle is something that imposes a structure on an otherwise unstructured collection of elements (water molecules) in order to create a hydrodynamic effect. Without the molding effect of the nozzle, the water can take any number of shapes. Does the nozzle mold the water molecules? Certainly.

    In fact, the nozzle is combined with the water in order to create an operative machine. Remove the molding structure of the nozzle, and the remaining elements lose their structure. Is the nozzle a “mold” as that term is being used by Justice Ginsberg? Certainly not.

    Similarly, a computer, before the addition of software, can take any number of forms. Without the software, the computer can be thought of as a collection of unstructured elements, like the water discussed above. Software certainly molds those elements into an operable form, in order to achieve a desired effect, but it must continue to be present in order to mold those elements during the operation of the machine. As soon as the molding effect of the software is removed, the operational structure of the computer is lost, and the machine is unable to achieve the desired effect.

    In this analogy, is the nozzle a mechanism? Indisputably. Is software similarly a mechanism? Indisputably.

  31. “Clearly, legal norms depend on how people perceive things. ”

    No doubt! In particular, it depends on how the drafters of the legislation in question saw things (like “components”) and how the Supreme Court sees what the drafters saw.

    I’m not aware of any convincing evidence that suggests that anyone saw information contained in transmitted signals as a “component.” And by “convincing,” I don’t mean convincing to AT&T and other people who would love to patent signals as compositions of matter. I mean convincing to a judge or jury — you know, the sorts of people who actually get to decide what is convincing and what isn’t.

  32. 2soft-vis:

    I met some CS people like you who claimed that RSA patent was obvious to them…

    I had no further questions for those folks….

    P.S. Copyright is only good for protecting a large chunk of trivial code, of which a good example would be pretty much anything coming from the open source crowd.
    Only patents can protect real inventions from copycats like you….

  33. It’s as if someone were able to patent all light bulbs by inventing one and claiming that you apply electricity on one end and light comes out the other. When a patent that patents a result mentioning “computer” and “magnetic disk” and “information” all the other hand waving that goes on in these patents, it’s like saying “oh, you know, electricity producing light, it’s all the same d*mn thing”….

  34. Want to resolve all this rationally? Want to watch me make all this nonsense go away?

    Watch this!

    ahem.

    “I grant you copyright on all code you author. ”

    There. Done.

    I can’t steal your code. To compete, I have to be at least as clever as you. We compete. Each of working furiously to outdo the other drives value through the roof. The market loves it. Consumers benefit. We develop niches and sustain ourselves serving different communities of users. More people join the fray. The cycle begins again. More choices for users! More innovation! Technology blazes forward !

    A socially beneficial arrangement for all involved.

  35. This is what you don’t understand- the correct presentation you wanted to make was as follows:

    two black boxes as specified. one analog as specified the other a magic genie created in a test tube performs the function.

    Input output signals are the same in both boxes.

    The patenter of analog box is GRANTED injunction against the genie box.

    Why? Because it’s the IDEA of having the input and output related in such and such a way, the result, that is being patented.

    If that last statement we’re true, then I could program a machine differently, in fact, in an infinite different number of ways, and avoid the patent. But I can’t; I am still violating the patent.

    That describes every single software patent that’s ever been issued.

    That is total BS, to use a colloquial.

  36. Can’t read this nonsense anymore..

    There is something all of you should know about this particular patent and technical field before one is allowed to make all kinds of idiotic remarks on “software” patentability.

    The ‘580 patent is for so-called linear prediction technique for speech encoding – essentially a method, or algorithm if you like, to encode speech signal at a low bit rate preserving its essential characteristics for subsequent reconstruction.
    This is called “vocoder”. The first vocoder was invented (and patented) by Homer Dudley of AT&T in early 30’s. At the time it used to be a huge pile of analog circuitry taking at least half of the room. No “software” whatsoever was involved in building it.
    The digital revolution of 60’s and 70’s led to the rise of DSP, or digital signal processing, a whole new technical field to replace all those analog gadgets with digital equivalents. The linear prediction was invented in 70’s by Schroeder and Atal of AT&T and it was a major breakthrough which allowed substantial speech quality improvements for (now digital) vocoders.
    The grandchildren of that first Dudley’s vocoder (based on linear prediction in its various forms , e.g. Code Excited Linear Prediction, or CELP) run on tiny DSP chips in each and every of your cell phones.
    Now, the question for the SCOTUS judges: if something like vocoder was clearly patentable as an analog device back in 30’s, should it be declared unpatentable after all things became digital, you know 1’s and 0’ those idiots are talking about…

    OK, how about a little court presentation: 2 black boxes – each box has a microphone as its input and a speaker as its output.
    One box has a purely analog vocoder implementation inside (vacuum lamps, whatever…) and the other box has A/D converter and a DSP chip executing some software code, followed by D/A converter and a speaker.
    Input and output signals are made identical for these 2 black boxes.

    What’s the difference between two boxes ? Why would anybody think that the first box (analog) should be patentable while the second box (DSP) might be unpatentable ?????.

  37. Absurd:

    1) inconsistent with reason or logic or common sense;

    2) completely devoid of wisdom or good sense;

    so 99.99% of software developers have an absurd notion of their field. Wow the rest of the world is so stupid; we need lawyers to help us understand things we went to school for 10 years to learn.

    The level of narcissism implied in that comment is really disturbing. You disagree for reasons X Y or Z. Your OPINION is divergent. You see things this way. But to call it absurd is really, deeply dysfunctional.

    Software IS just an mathematical algorithm and any software program can be expressed as a mathematical equation. People who haven’t studied CS deeply don’t know that, but then again, those same people shouldn’t have a hand in writing laws that effect CS either.

  38. Trevor hill wrote: “things like prices, mp3 files, electronic documents, etc., are all concrete and tangible objects”

    I’ve never heard anything sillier in my life. Have you looked up “tangible” or “concrete” in a dictionary? If you had, you’d see how silly this claim is.

    The case is trivial: Microsoft is correct. I can think of several different legal theories of the case under which Microsoft is correct; we’ll see which of the five or so the Supreme Court chooses, or whether they go with something incoherent. There are no coherent theories under which AT&T is correct, discounting the “All your base is belong to us” theory. The best theory they could go with (and they didn’t go with this theory) is that Microsoft is inducing third parties to infringe their patent, and Microsoft could probably defeat that based on the publication principle of patents.

  39. Rather a system programmed to do something in particular is

    Yes, to achieve an end through any means at all, heck, we’ve thrown out even the need for any physical embodiment at all (i.e. Lundgren). A bit of a land grab, isn’t it? But it goes on every day, to society’s detriment, as the USPTO.

    This is very serious, and it’s implications for society are very serious. I don’t see that the Court is scientifically challenged; I see that they are not acting like they are beholden to corporate interests or enthralled by bizarre ideologies involving the universal applicability of IP.

  40. >Treat software as mathematics and be done with it.

    That is absurd. Software is not mathematics. It is an encoding of a structured set of operations which is combined with a physical machine in order to make that machine operate.

  41. Philip –
    For purposes of these arguments, whether software, in and of itself, is patentable is off-target. A gear or switch is not patentable, but may be a component of a patentable system.

    Even in this case, the software itself is not patented. Rather a system programmed to do something in particular is. No one would argue that a machine whose gears are set in such a way would be per se unpatentable because gears have been known or are obvious.

  42. The fact that software must be executed by a physical computer in order for either the software or the computer to be useful is irrelevant to the discussion of software patents. The only relevant question is whether or not software ought to be patentable, and as far as I’m concerned, that question does not hinge on the form that software takes.

    Software is just a sequence of instructions to a computer. It is purely functional in nature, although one could argue that there is a certain amount of creativity available to programmers in regards to how one instructions a computer to perform a particular task. Nevertheless, it is not uncommon for programmers to independently come up with similar or identical algorithms to solve the same problems, particularly when you look at software as comprised of smaller and smaller components. This is precisely why allowing software to be patentable is so fraught with difficulties: you may THINK that your algorithm is clever enough to be patentable, but it probably isn’t. It is far more likely that you’re already the nth person in the past 50 years to think of it, where n is somewhere considerably north of 1.

    Most programmers are humble enough to accept this fact. Most software companies and their patent lawyers are not. Generally speaking, the software industry as a whole would be better off if software patents were not allowed. It isn’t fair to most programmers that the code they write, independently of anyone else, should be considered patent infringement simply because the algorithms involved were simple enough to be thought of by more than one person. Sure, it would at least be tolerable if the USPTO employed examiners that were capable of distinguishing between the rare “novel” algorithm from all the rest, but I personally think there are so few of the former than we’d all be better off if software patents were disallowed, period. Treat software as mathematics and be done with it.

  43. >>A more accurate (and more clear) correlation is
    >source code is a recipe or a cookbook.
    >
    >That is incorrect. Computer code, especially object
    >code, must be combined with the machinery of the
    >computer to be used.

    A more accurate correlation would be to an assembly of gears, which is a mechanism to transduce information encoded in the physical shape and arrangement of the gears, via energy transfer (force and motion), from one location in a machine in order to act upon another location in the machine. Similarly, the instructions contained in software, its mechanism, is seqentially transduced as it is read from disk, transferred over wires, stored in RAM, etc., in order to be read and operated on by the CPU. Each transductive step transfers the mechanism embodied by the software, but does not alter the operative structure contained within that mechanism.

    One need only look to history, to Charles Babbage’s Difference Engine (link to en.wikipedia.org) and his Analytical Engine (link to en.wikipedia.org), to see that such an analogy holds, and to prove that software is, in fact, a tangible mechanism.

  44. >A more accurate (and more clear) correlation is source code is a recipe or a cookbook.

    >>That is incorrect.

    No, you are incorrect and he is correct. Object code is just a recipe for the actual thing being protected. the “abstract functioning” of the object code and NOT the specific way it performs it, or the source code or even the object code.

    No one is accusing MS of stealing source, so the object code is NOT the same as ATTs.

    What is *trying* to be protected is not something concrete, not source, not object code but rather something abstract, to wit, what the software achieves and a very very very broad definition of how it achieves it.

    I can write software to achieve the same ends in a large number of ways and you want to claim them all, even those that aren’t written by anyone.

    That is patenting a result achieved you care not how. That is patenting something abstract. That is why we are where we are.

    It’s hilarious to hear lawyers decry the technical ignorance of Court while they themselves completely ignore the opinions of the vast majority of actual scientists software developers, 99.9999% of which are for the total abolition of software patents dismissing them with comments like , they’re not lawyers, they don’t know the law, as though their opinion on technical matters was somehow superior to the experts who’s field they wish to lay claim.

    That’s the only pathetic thing here.

    Have fun listening to the court make Platonic distinctions between the words of Moby Dick and Moby Dick itself. This is exactly what you get when you attempt to patents ideas, which is exactly what this is about, and exactly why society is forced to waste its time and resources on this kind of twaddle.

    One

  45. Very good so far. Now you people (I do mean you people) are starting to dissect the issues more cleanly rather than simply throwing mud (or its functional equivalents) at the Supreme Court Justices and the lawyers who belong to the fair and balanced 3 sides who appeared in court. (BTW Zeke, that is basically my answer also, the lizard, the lizard came first.)

    Speaking of what comes first, have you ever thought that we might be going at this from the wrong direction simply because many of us read only from left to right? What I mean is this: Where do we start dividing things into “components”? Do we start at the manufacturer’s site and follow “components” forward in time as they move towards merger in the potentially infringing computer; or can we start at the other end, with the finished computer, and then start mentally dividing that final assemblage into “components”? For example can we divide a computer into its “informational” components and its information-storing media components and its information “processing” components? What happens then? What happens if we start at the right end of the time line, divide out the “components” there and then trace back in time to see who “supplied” such components and how they did it? Can you say with absolute certainty that one view is more correct than the other?

    Who among us belongs to the chicken-first crowd and who to the egg-first crowd?

    (Personally, I divide the world into 10 kinds of people ….

    those who count in binary and those who don’t :-)

  46. Trevor Hill wrote, “To liken software to a mold to blueprint or… is inaccurate.”

    What about likening the master disk to a mold or blueprint?

  47. >A more accurate (and more clear) correlation is source code is a recipe or a cookbook.

    That is incorrect. Computer code, especially object code, must be combined with the machinery of the computer to be used. The combination must be persistent. The computer is inoperable in the absence of that combination. Remove the physical embodiment of the software from the computer (the electric charges defined by the “1s and 0s”), and your computer is about as operable as a paperweight.

    By the way, this applies to source code, as well as object code. Many languages run directly from source code (perl, python, tcl, etc.), they are interpreted rather than compiled, so let’s not get into playing silly word games with topics such as software “assembly,” etc.

    A recipe, on the other hand, is used once and then discarded. The recipe is not combined with the food elements. You don’t need to use the recipe to use (eat) the food. That is why recipes, blueprints, designs, etc. are nothing like software.

    The error that everyone (including Waxman) made was to accept the false assumption that the information represented by software is “abstract.” It’s not. It’s real and concrete, and exists in physical form. It is not an abstract notion, it is a real and tangible mechanism. The fact that its physical form is easily transferrable, via electrical impulses, photons, etc., doesn’t change this fact. In fact, Bryer’s idiotic analogy of reading the 1s and 0s of a software program over the phone, as a supposed illustration of its abstractness, actually demonstrates nothing other than yet another possible (albeit tedious) method for physically transferring the tangible mechanism of the software from one location to another.

  48. Malcolm, what in the world do you mean by “how many compositions?”

    The only way to answer that question is to ask what we as people consider to be compositions of matter, and that depends on what we consider to be semantically distinct things.

    We certainly think a “radio station” is a thing, but objectively, we needn’t think of it that way. It’s merely a bunch of waves, or a frequency. Actually a range of frequencies. You can’t define these things objectively, and then think that those definitions will easily map onto our human concepts of what constitute separate things.

    Clearly, legal norms depend on how people perceive things. They can work in no other way. They must be informed by technology, but must make distinctions that help implement the law, not necessarily distinctions that simply make sense for ordering physical theories.

    Most people think of a program as a thing, a file as a thing, a web page as a thing, etc. Thus, it makes sense to have our laws work according to these concepts. We certainly consider a radio or TV program to be a legally cognizable thing, which can be regulated, but all it is is wave fluctuations in the electromagnetic spectrum.

    All matter is merely a collection of standing probability waves anyway, as we know now from quantum physics. There’s no sense in making a distinction between electromagnetic waves and baryonic waves. There is sense in a distinction between devices or processes with a concrete, tangible, and useful result, versus those without.

  49. “Rational scientists and physicists understand that electrical signals are generally manifested by drifting electrons and by the electromagnetic force lines that emanate from these real “compositions” of matter.”

    That’s nice, stepback.

    So, if I show a “rational physicist” a movie of an apple, a radio playing music, and a floppy disk and ask her to tell me in ten seconds how many compositions are in the picture, she will answer without hesitation: SIX, the apple, the radio, the floppy disk, the light signals emanating from the objects, the sound signals emanating from the radio, and the radio signals entering the radio.

    Right? Is that how rational scientists think?

  50. “Reading all this nonsense coming from technically clueless people makes me really depressed… ”

    What nonsense? You mean the “nonsense” that a “technical” argument which contradicts common sense and understanding by most people (including most scientists) is being rejected by highly educated people?

    Doesn’t sound like nonsense to me. In fact, it sounds highly predictable. So predictable that … many people predicted it. I even offered people a chance to earn some money by playing the odds — nobody bit.

  51. Software can be looked at in many different ways, but here is the key: the court must look at it in terms of its practical effect, its practical use, and in terms of how people, industry, and the law, interact with that use.

    So it’s useless to simply say that software is a recipe or soemthing. That would imply that selling software is tantamount to selling a recipe, which is hogwash. When one buys software, one thinks of it, and uses it, as though it is in fact a machine, a device. The fact that software must be put into another machine (the computer) in order to use it is merely an afterthought in terms of its actual use. The software is the machine, for all practical purposes. The court needs to recognize this.

    The court also needs to recognize why State Street was decided the way it was — because things like prices, mp3 files, electronic documents, etc., are all concrete and tangible objects that we use every day, regardless of the fact that they are embodied within a computer.

    It is immensely frustrating to see the court grasping for tutelage on what software is and how it works. Oral argument is not the time to be going back to square one. It is also not the time to be throwing out poorly-thought-out extemporaneous hypotheticals with which to flummox counsel. There are so many land-mines in each of their hypos, no counsel could possibly do what they should have, which would be to carefully revisit all assumptions on which the hypos are based, and explain the appropriateness or lack thereof for each one.

    It appears as though the Justices were just stopping in for a chat at a cafe, rather than asking well-informed questions based on their thorough understanding of the posture and facts of the case. I am very disappointed.

    I’m also worried that with Roberts out of the picture, we will get an opinion that everyone knows is wrong, but that nonetheless throws Fed. Cir. jurisprudence completely off-kilter for no sound reasons. I will clasp my hands and simply wish that the court will realize its lack of clear understanding of all the issues, and just restrain itself…

  52. To liken software to a mold to blueprint or… is inaccurate. A more accurate (and more clear) correlation is source code is a recipe or a cookbook. The recipe may be printed on a physical medium (paper, in a book) or it may be distributed without said physical medium (word of mouth, television, internet, email). The recipe itself is only an idea. It becomes reality when the person reading it executes the given instructions in sequence. Furthermore, the instructions should be detailed enough so as to limit the possible outcomes, but broad enough to handle the situations (say the person is allergic to nuts, the recipe in most cases should not list them as a *requirement*).

  53. Reading all this nonsense coming from technically clueless people makes me really depressed…
    Sadly, these people have more influence on the future technical progress in this country than the brightest US scientists and engineers…

    Something is clearly wrong here…

  54. JUSTICE GINSBURG: “Let me ask you about a domestic law question. One side is telling us it’s the component that’s supplied, whether it’s the master disk or the object code. And the other side says this is just like a blueprint, like a mold, like a template. Can a blueprint be patented? Can a mold be patented?”

    In view of Justice Ginsburg’s comments, suppose that Company A patented a plastic structure, but failed to patent the mold or molding system used to make the plastic component. Party B manufactures the mold, exports it and makes the plastic structures. Should the mold be considered an exported component of a patented invention? Should the plastic structures made with the exported mold be considered an exported component of a patented invention? Should the master disc be treated differently? Finally, doesn’t the fact that Company A could have patented the mold and the overall molding system make their cause seem less sympathetic?

  55. step_back wrote, “There is no issue regarding ‘disembodied’ software.”

    In discussing his opponent’s (AT&T’s) positions, Mr. Olson noted, “And then they say [the component's] just the binary code in the abstract, but that in the abstract never becomes a part of the computer.” This reference to binary code that never becomes part of the computer is referring to disembodied software, that is, software that is not embodied on a computer readable medium.

    We all know, of course, that software doesn’t exist unless its embodied on some medium. But that’s AT&T’s problem. To make its “supply” argument, AT&T needs the component to be the abstract, disembodied software — not the original physical embodiment of the software.

    If the component is the original physical embodiment of the software, Microsoft wins big because the original physical embodiment is never used as claimed — it’s merely copied. See p. 7, line 4 (“[The master disk] is being copied.”).

    Think about it. Are any original electrons physically moved from an original source disc to another? No. They’re copied. The originals remain. Are the pits in the golden disc somehow removed from the golden disc and added to another disc? Of course not.

    The golden disc is being used as a template to configure another tangible device. But no portion of the golden disc is physically removed from the golden disc and then added to the other tangible device. The supply issue under the facts at issue necessarily require an analysis of the tangibility question.

  56. ‘And while we are at it, we might want to consider when “life” begins or whether the chicken rather than the egg came first. :-)’

    There are two possibilities. Either the first chicken hatched from a chicken egg laid by a pre-chicken (egg first theory) or a pre-chicken mutated sometime during its lifetime into the first chicken (chicken first theory). Since a hypothetical biologist observing pre-chickens would not likely recognize the spontaneous mutation of a pre-chicken into a chicken (the once a pre-chicken always a pre-chicken assumption), the chicken first theory is out the window. Egg first.

    You might disagree and say that I’m playing games and that the egg laid by the pre-chicken could equally be called a pre-chicken egg. This is what’s known as the chicken first through the backdoor rebuttal. Nice try, but it’s flawed by a logical impossibility, namely, how could a chicken hatch from a non-chicken egg?

    There is actually a third possibility, and that is chickens are actually a disembodied construct (the no chicken theory), but we needn’t discuss that here.

  57. Please allow me to answer your question bluntly. No. There is no issue regarding “disembodied” software.
    I have never observed a computer to work off of “disembodied” software.

    uh, that would be your opinion, but I think the issue is how does SOCTUS see it….

    Breyer said it, and NOT in reference to DISEMBODIED anything, since there would be NO need to say it if he was referring to disembodied anything- the Court has not ruled on software patents, ever. It’s an issue, and an issue to still be decided by SCOTUS.

  58. Malcolm,

    (Warning ad hominen attack coming.) I have scientifically determined that there is a 54% probability you already tried the 120V AC signal experiment at home and that when you awoke afterwards it you did not remember a thing and thus did not learn from the experience.

    Rational scientists and physicists understand that electrical signals are generally manifested by drifting electrons and by the electromagnetic force lines that emanate from these real “compositions” of matter. (Electrons are composed of the more fundamental particles, mesons, leptons or that sort of thing –sorry that theory came after I left school. Therefore they are clearly compositions of matter and not disembodied spirits.) I cannot vouch for “most” English-speaking human beings because certain polls indicate a majority of them believe in supernatural forces. The US patent law speaks in terms of persons skilled in the useful arts and not in terms of “most” English-speaking human beings.

    BTW, don’t you agree that Simon was a bit too mean to some of the girls on American Idol tonight?

  59. “Your reference to the Golden Master as being the only tangible thing out there begs the question and distorts the reality.”

    It’s the only “tangible thing” in this story that most English-speaking human beings — including most scientists, and including most physicists — would fairly characterize as a “composition” or “component.”

    That’s reality.

  60. SF:
    Please allow me to answer your question bluntly. No. There is no issue regarding “disembodied” software.
    I have never observed a computer to work off of “disembodied” software. The only time I have seen computers respond to software is when that software is embodied in a physical signal that is transmittable into the processing bowels of the machine for fetch and execution by its CPU.
    Now some readers out there may resort to use of shamans and other occult practioners in the supernatural arts for whooing ghosts into (or out of, if it’s an exorcism) the hardware shell that they refer to as the “computer”. As for myself and many other practioners in the real physical sciences it is necessary for physical particles to be moved, such as electrons or photons for thereby transferring the signaling energy for real into the corpus of the computer and causing real physical change in the magnetic domains of the computer’s disk and/or physical change in the trapped charge of the computer’s flash memory before that computer responds to the alleged software.

    So in sum, no I do not give this talk about ghosts and “disembodied” software any credence.

    Your reference to the Golden Master as being the only tangible thing out there begs the question and distorts the reality. Real, tangible signals were used to record the signal in the Master Disk and real tangible signals were used to continue the transference of the computer-instructing signals from the Master Disk to the end product computers. If you do not believe that electrical signals (for example) are real and tangible, you might try connecting your human shell across the 120V AC “signal” available at your nearest home outlet so as to determine if it has real, tangible and concrete results. However I would caution against doing so as such experiments may be hazardous to your health.

    The real question remains. Where does the “supplying” process end and where does the assembly process (assembly of a computer that executes NetMeeting) begin?

    And while we are at it, we might want to consider when “life” begins or whether the chicken rather than the egg came first. :-)

  61. step_back: Isn’t there an issue as to whether disembodied, non-tangible software is a component? I thought Microsoft’s position assumed that disembodied, non-tangible software is not a component. Thus, the only tangible component was the master disk and infringement would only occur if that disk was actually used as claimed (that is, instead of being used for mere copying).

  62. Excuse my forthrightness, but I feel compelled to act like Simon of The American Idol Show and to put a halt to the abysmal performance that is evolving here at Patently-O. You all can do better than this.

    It is fairly undisputed that the “software” was wholly manufactured in the USA and that it is an essential component of the NetMeeting system.

    The only question on the table is that of identifying the point in time where the “supplying … from the USA” terminates and the act of assemblying of the claim-matching products begins.

    AT&T’s position (if I understand it correctly) is that replication of the Master Disk into hundreds of copies is an integral and continuous part of the act of “supplying or causing to be supplied” (see 35 USC 271(f)). Their position is that assembly of the product (which assembly would clearly be infringement if it instead occurred in the USA) begins only after the multiple copies have been made.

    By contrast, Microsoft’s position (if I understand it correctly) is that the act of “supplying or causing to be supplied” ends with the arrival of the Master Disk in Europe. Everything after that belongs to the process of assemblying, including the step of replicating data from the Master Disk into hundreds of other disks. Thus according to Microsoft they are liable only at most for the price of one Master Disk, or not at all liable if the Master Disk is itself never installed in a computer (which it is not).

    Now that the real issue is on the table, go ahead and debate the merits of both sides.

  63. Michael Risch wrote, “Seems like that would be obvious, no? Haven’t software replicating systems been around for some time? And even then, if the replication is outside the U.S., how would you enforce it?”

    First, if a software-related invention is novel and non-obvious, the replication of that novel and non-obvious invention is necessarily novel and non-obvious. How could it be obvious to create multiple copies of an initial software program if creating that initial software program isn’t obvious?

    Second, if the patented replication system includes a master disk, then the master disk is a tangible component of the patented system. Thus, this statute would certainly apply.

  64. The object code seperated from a physical, computer-usable embodiment, is nothing more than instructions on making a computer-usable medium that contains a representation of said object code.

    There, I just wrote the majority opinion :)

  65. “That’s why you would write a claim to a replicating system that used the master disk and then try to get royalties based on the individual installations as a convoy to the use of the replicating system.”

    Seems like that would be obvious, no? Haven’t software replicating systems been around for some time? And even then, if the replication is outside the U.S., how would you enforce it?

    Perhaps patents should be written like this:
    I claim:
    1. The invention.

    2. Copies of the invention.

    3. Thinking about the invention.

    4. Every part of the invention, so long as it is used in the invention.

    5. Every part of every part of the invention, so long as it is used in the invention.

    6. Every molecule that is part of every part of every part of the invention, so long as it is used in the invention.

    7. The design for the invention (including this patent, which is by definition enabling).

  66. Well the distinction between blueprints and a software component can be made as large as needed and ATT would still have something to cry over. If you tell me object code is a violation, I will send source. If you tell me source is a violation, I will send UML or whatever I need to to raise the level of abstraction while still specifying the function. It’s clear that if MS had sent source, ATT would have been SOL. Twist and turn as much as you like, the proper and non-paradoxical level of IP protection for software is copyright/trade secret. Every other road leads to an intolerable quandry, as MS is finding out now.

  67. Michael wrote, “That’s really the point – sure, one golden master disk is infringing, but only if that disk is actually put into a computer, and then it’s only one instance of infringement for damages.”

    That’s why you would write a claim to a replicating system that used the master disk and then try to get royalties based on the individual installations as a convoy to the use of the replicating system.

  68. I especially love this:

    JUSTICE KENNEDY: But suppose, suppose you had a machine that makes another machine, and if you ship that machine to Europe — and there’s a patent for the machine that makes it. If you ship it to Europe and it starts making another machine, the statute is not violated; and isn’t that just what’s happening here?

    Actually, that is exactly what happens when source code goes to object code. A machine, the compiler, makes another machine, called object code through the mechanism of reading source code. So Kennedy has, quite accidentally, declared that sending source code overseas , where it is then compiled into objct code and distributed to NOT be violating the statute:

    Waxman is falling all over himself to declare that that is NOT what is going on:

    MR. WAXMAN: No, no, no. This is not a machine tool. The thing that was violated, the machine readable object code, is precisely what is installed on
    the computer and precisely what is moved from one part of the computer to another in different forms as the computer operates and it continually instructs.

    That is just hilarious. So object code is in violation, but source compiled overseas is not. I will keep that in mind. I have to go tell my friends.

    also:

    JUSTICE SCALIA: That, that code is not patentable, you’ve said.

    MR. WAXMAN: The code is not patentable.

    The expression is copyrightable. AT&T has not sought to get a patent on the code. AT&T has a patent on a system that can be practiced, among other ways, through the use of software.

    OK so no more “software is patentable and everyone knows it”… thanks. Just couldn’t resist.

  69. Just in from the SCOTUS in M$ vs ATT:

    I am pleased to present the words of one Justice Steven Breyer; who today said he was afraid to give the ruling to ATT because infringement could then be interperted so broadly that “transmission purely of information” could be prosecuted as a violation.

    “I would be quite frightened of deciding for you and discovering vast numbers of inventions that can be thought of in the way you describe this one,” he said.

    Breyer used the example of someone reading the text of a patent claim over the phone to someone in foreign country who later decided to make the same product, a reading of the law he said “can’t be right,”.

    Which is exactly what I was saying earlier- that merely learning in an academic environment an algorithm then “exporting” yourself back to your country of origin would create grounds for infringement and have the effect of making such a person unemployable anywhere in the world. Of course it would. This is almost something out of a Python routine.

    Moreover, to those in this forum who have asserted that software patents are settled law and “the Supreme Court has already determined that software is patentable” as one poster insisted (hopefully, not a lawyer anyone was giving money to) or “there is no such thing as a “software patent” as another confident poster asserted, it appears that not only does Justice Breyer think there’s such a critter, but the matter has NOT been ruled on:

    “We’re operating under the assumption that software is patentable…but we’ve never held that in this court, ever, What should we do here?”, Breyer said, addressing Joseffer of the U.S. Department of Justice

    A system exists in which every software developer is responsible for every line of code or idea expressed in every claim in every patent ever published known or unknown to him and for ascertaining, through the opinion of a qualified attorney( because, in fact, as a matter of LAW, I, a non-lawyer CANNOT be said to be able to judge whether an invention infringes) whether some line of my. possibly vast source code infringes said claims and FURTHER not only am I responsible for each and every claim, which essentially functions as its own independent patent, but if I attempt in good faith to educate myself about their existence, I can be fined treble damages because, as a matter of law I cannot, as I said above, reach a conclusion one way or the other, however just having looked at some or any or one of the hundreds of thousands of patents I am responsible not to infringe is sufficient to have treble damages awarded to the patent-owner, should I be found guilty of infringement.

    There, Dennis, did I get that all right?

    No offense but you can see why software developers despise software patents and are adamant that if the country continues to act AS IF they they had been enshrined into law, which they have not, that software development will for all intents and purposes be brought to a standstill, or as one observer put it, if patents had been permitted earlier, we would still be in generation one of word processing and spreadsheets.

  70. Software is merely a convenient way of discussing a particular configuration of a computer-readable medium (for example, the pits on a compact disc). It’s the particularly configured computer-readable medium that is patentable — not the disembodied configuration itself. Disembodied configurations (like software) are not components of a patented invention.

  71. I liked the question about reading (or shipping) the patent – that’s how I think about this case. I think the critical admission was this one by Waxman:

    JUSTICE STEVENS: But that they now send to some central point which redistributes them. They send one copy to, say, the wholesaler, who then makes 500 copies that are sent to the retail customers, isn’t that correct?
    MR. WAXMAN: That may be. We don’t —
    JUSTICE STEVENS: Which is exactly what happens in this case?
    MR. WAXMAN: It’s one of the things that happens in this case. We have a stipulated record that is not very detailed, but in that instance there is only one violation.

    That’s really the point – sure, one golden master disk is infringing, but only if that disk is actually put into a computer, and then it’s only one instance of infringement for damages.

  72. When asked if there was an another, non-software industry that might find more “components” if software, in this sense, is a component, I believe they missed an obvious one: nanotech. Could you imagine if individual atoms constituted components? Or photons (they sure liked the word photon in that argument)?

  73. JUSTICE BREYER: But I then would be quite frightened of deciding for you and discovering that all over the world there are vast numbers of inventions that really can be thought of in the same way that you’re thinking of this one, and suddenly all kinds of transmissions of information themselves and alone become components.

    Bingo. And even a caveman will admit that light waves and sound waves and radio waves and water waves can not be meaningfully distinguished when it comes to their ability to carry information and be transmitted.

    I don’t suppose anyone will take my bet now. :(

  74. The problem with this kind of exchange is that there comes a point where the Justices have to go back for 4 more years of intensive college training just to understand what they are talking about: What is “software”? Is it like the “blueprint” tail end of the elephant or like the moldy front end? What is a “component”? Is a signal a real physicallity or is it an abstractionisitc ghost? Is the sending of the Golden Child Disk an integral part of “supplying” or is the Golden thing itself, “the component”? Such hard hard questions. And yes, how many angels do indeed dance on the head of the disk player’s needle?

  75. It was a pretty fascinating oral argument.

    Dennis, I guess I would disagree with you on Waxman’s response to Justice Breyer’s blueprint analogy.

    Even Breyer seemed to accept– in response to Ted Olson’s answer– at least the possibility that Seth Waxman had drawn a sufficient distinction:

    “Justice Breyer: His answer to me is that — it’s sort of misleading to think of this as if it’s just information, because it’s really a method that switches things at a level of detail that impossible to put in a patent application. It’s taking the information in a patent application, it’s transforming it into what we think of as 1s and 0s, but they’re not really even 1’s and 0’s. Whey they are is things that happen with electricity… And it’s putting that on the disk and then it makes other things happen. … It’s putting something on a disk, pits and lands, instructions that are copied from the master disk and then put into either a hard drive or a disk, it’s the same thing. It’s something that is into the computer that will make the computer operate.”

    Who knows if Breyer is truly persuaded, but he obviously saw enough merit in Waxman’s response — even borrowing Waxman’s terminology — to use it against Olson.

Comments are closed.