|
Is the economic cost of patent trolling slightly harder at least because the defendant can maybe get the patent invalidated by a judge rather than a full jury trial?
|
#
?
Jun 19, 2014 23:14
|
|
|
|
| # ? May 9, 2024 18:45 |
|
|
ShadowHawk posted:Is the economic cost of patent trolling slightly harder at least because the defendant can maybe get the patent invalidated by a judge rather than a full jury trial? This decision has nothing to do with that, plaintiffs are always going to request jury trials and invalidity will go to the jury. Very, very rarely you can get a judge to issue summary judgment of invalidity - but it almost never happens. The CBM and IPR processes at the PTO introduced in the AIA have been somewhat helpful in changing the economic calculus, though, and this decision does make CBMs a little easier to win.
|
|
#
?
Jun 20, 2014 00:28
|
|
|
gently caress it I already rejected my first application under Alice Corp because I really don't have good art and our 101 guidance is still shot and this really shouldn't be patented even though it's the 4th continuation reciting basically the same invention as three issued patents but the law finally caught up.
|
|
#
?
Jun 20, 2014 01:14
|
|
|
The big thing I'm wondering about this is how it will affect cryptography and data compression patents, especially public key cryptography, that are not much more than applying an otherwise-unpatentable mathematical formula.
|
|
#
?
Jun 20, 2014 05:55
|
|
|
OneEightHundred posted:The big thing I'm wondering about this is how it will affect cryptography and data compression patents, especially public key cryptography, that are not much more than applying an otherwise-unpatentable mathematical formula. It'll depend on the specifics of the patent, really - algorithm patents were already (theoretically) not allowed, but patents that use an algorithm to do some function (e.g., public key cryptography to exchange secured email) were and likely still will be fine. It was pretty much a status quo decision.
|
|
#
?
Jun 20, 2014 06:38
|
|
|
I was thinking about this on the train tonight, I think there's a strong argument that computer programs are essentially an "artifact of nature" in the same way that, say, the Fibbonacci sequence is. Using Turing machines yields some interesting properties about general computers. Essentially, there are an infinite number of computer programs, but there are only countably many computer programs. Turing machines allow a way to assign numbers to every possible computer program (algorithms, inputs, and resultant outputs) - there is a Turing Machine 1, Turing Machine 2, etc (not all of which halt). Every (computable) computer program, no matter how big, can be transformed into a Turing Machine, and then represented as the corresponding Gödel number. It becomes really lovely for non-trivial programs, but you can do it programmatically. To me at least, that sets the stage for the algorithms themselves to be unpatentable, in the same way you couldn't patent the Fibbonacci sequence. The algorithm itself isn't really important, knowing that it's Turing Machine 38876369 is sufficient to describe it, it's how you apply the algorithm. In the same way, parts of the Fibbonacci sequence probably aren't patentable as such, just the way you're applying the Fibbonacci sequence to solving real-world problems. This is also a rather obvious conclusion of Lambda calculus, in which computer programs are essentially the evaluation of mathematical equations, which is also not patentable. You can't patent an equation, and the method of evaluating it is obvious. There's a third type of machine, a recursive machine, which can also be mathematically evaluated, and the computing power of all three machines (Turing machine, Lambda calculus, and recursive machines) are all equivalent (Church-Turing Thesis), they're just different ways of processing the same result. The "general computer/specialized computer" framework is clearly awkward as poo poo and SCOTUS doesn't seem to like it. It really clearly becomes absurd if you apply computational theory or number theory to it. So far as I can see it is literally impossible to run a program on a general computer, they exist only in theory, which is incredibly  if you think about it. As soon as you load something into memory it becomes a specialized computer. They exist as an abstract class of computational machine within our legal framework. It's a totally valid thing in computer science but it's really weird to see it used legally.e: Shifty Pony posted:The giant elephant in the room is the question of if a generic computer programmed a particular way can impart structural novelty to the invention. This, and I think "no", clearly. e2: The big thing is that Universal Turing Machines also require a input/program (operating memory and inputs are mixed together on a single tape), but there are extensions with separate memory/input/output tapes, the computing power is equivalent. Paul MaudDib fucked around with this message at 08:22 on Jun 20, 2014 |
|
#
?
Jun 20, 2014 06:41
|
|
|
Paul MaudDib posted:This, and I think "no", clearly.
|
|
#
?
Jun 20, 2014 07:08
|
|
|
Paul MaudDib posted:I was thinking about this on the train tonight, I think there's a strong argument that computer programs are essentially an "artifact of nature" in the same way that, say, the Fibbonacci sequence is. "Artifact of nature" doesn't mean what you seem to think it means.
|
|
#
?
Jun 20, 2014 08:16
|
|
|
Kalman posted:"Artifact of nature" doesn't mean what you seem to think it means. You'll have to be more specific, because stuff like equations definitely aren't patentable, and computer programs are nothing more than an equation that is evaluated. quote:What Are Abstract Ideas? These two ideas are fundamentally at odds. Computer programs are pure mathematics, a formula to be evaluated, nothing more. This is a thing that computational theory tells us. Furthermore we can enumerate every possible computer program that could ever be written (countably infinite), so if anything they are even less inventive than the set of all possible mathematical formulas, where there are uncountably many possible formulas. These statements may not make sense, but they are provably true. In the lambda-calculus view, the algorithm itself is just a mathematical formula. The computer follows standardized rules to manipulate the formula and produce an output. The turing computer model is the inverse of lambda calculus. The computer is just a standardized black box that takes a number and some input and produces an output. Specifically, since the output can also be encoded, the computer is really just a mathematical function with two parameters (program number and an encoded input). It really doesn't matter what function we use to evaluate, as long as it's Turing complete and everyone uses the same function. From this perspective, probably the only inventive/patentable thing is what input you provide the program and how you utilize the output. On its face this argument only attacks the individual steps in an algorithm, but combinations of the sub-steps are just another program and therefore represent a different Godel number that satisfies a different set of criteria, the program number and input are evaluated by the same black-box function to produce a different result. Patenting an algorithm is very closely akin to patenting a prime number. It's neat, it fits certain criteria that could be useful in certain applications, but that shouldn't make it patentable separate from the application. Paul MaudDib fucked around with this message at 09:09 on Jun 20, 2014 |
|
#
?
Jun 20, 2014 08:24
|
|
|
Paul MaudDib posted:These two ideas are fundamentally at odds. Computer programs are pure mathematics, a formula to be evaluated, nothing more. This is a thing that computational theory tells us. Furthermore we can enumerate every possible computer program that could ever be written (countably infinite), so if anything they are even less inventive than the set of all possible mathematical formulas, where there are uncountably many possible formulas.
|
|
#
?
Jun 20, 2014 09:33
|
|
|
The law does not give a gently caress about computational theory. Nor should it. The "all programs are algorithms" argument hasn't succeeded yet in court, and it probably never will. (And it shouldn't.)
|
|
#
?
Jun 20, 2014 10:02
|
|
|
Kalman posted:It'll depend on the specifics of the patent, really - algorithm patents were already (theoretically) not allowed, but patents that use an algorithm to do some function (e.g., public key cryptography to exchange secured email) were and likely still will be fine. It was pretty much a status quo decision. Preempting every practical application of an abstract idea
|
|
#
?
Jun 20, 2014 12:11
|
|
|
Paul MaudDib posted:You'll have to be more specific, because stuff like equations definitely aren't patentable, and computer programs are nothing more than an equation that is evaluated. You've had some CS education (but not much math I'm guessing) mentioning stuff like lambda calculus and turing computers but then you go on to make no sense talking about the equivalency of algorithms and prime numbers...
|
|
#
?
Jun 20, 2014 12:15
|
|
|
Kalman posted:The law does not give a gently caress about computational theory. Nor should it. Right. While an interested line of thinking, I believe this isn't a productive one for the courts to explore. Yes computers can be represented as an abstract algorithm, but in the same way a house or car can be represented as a mass of statics equations. Where I think modern patent claim interpretation goes wrong is in the weight given to the arrangement of electrons or magnetic fields in non-volatile storage. If I grab two identical system-on-a-chip devices and put two different control programs onto them with different sensor feedback coefficients, as far as present patent interpretation goes they are now no longer the same. This is, IMO, stupid as poo poo. A person of skill in just about every field would not ever consider such an interpretation a reasonable one. In my view this is the same situation as what was decided in In re Ngai: a claim should be able to claim printed instructions (a program stored in a ROM) and to anticipate the prior art should also have instructions, but the CONTENT of the instructions should not be able to patentably differentiate the claim from the prior art. Even were that not the case, should the programming of an integrated soc or controller in any manner which operates the various subsystems which it is connected to to merely perform what they are recognized to be used for (such as a variable speed motor being speed-controlled, an electrically triggerable fuel injector being triggered, a display being driven to display an image, a touch screen monitored for inputs, etc) be presumed to be simply a matter of routine optimization and skill, absent some unexpected synergy or complete lack of indication that anybody ever considered it something to be optimized? The entire point of such cpus/controllers is to be programmed in any way to user desires, shouldn't the presumption be that merely doing that is just routine engineering? To go back to the house analogy: would anyone consider the placement of the windows and interior walls in a house anything but a matter of routine optimization of the building to the purpose in which it is intended to be used? This would also have the benefit of leaving open the door to the possibility that someone comes up with a way to use a device in a completely new way contrary to accepted thought at the time.
|
|
#
?
Jun 20, 2014 13:44
|
|
|
Yeah the patent office doesn't look fondly on no-reference 103 rejections. I know; I've tried.
|
|
#
?
Jun 20, 2014 14:15
|
|
|
I submitted a 1-reference optimization 103 to my sig panel because I'm insane like that. Didn't hear a peep.
|
|
#
?
Jun 20, 2014 14:37
|
|
|
I've failed sig twice because of poo poo like that so welp
|
|
#
?
Jun 20, 2014 14:44
|
|
|
Commentary from Robert Merges, who wrote a lot of the IP casebooks:quote:Those of us who sweat in the clammy gymnasia of patent law have been waiting with a mix of excitement, dread, and cynical disregard for the Alice v. CLS Bank decision. The idea was, when the Court took the case, that we would finally have an answer to the question whether software can be patented under U.S. law. To say we did not get an answer is to miss the depth of the non-answer we did get.
|
|
#
?
Jun 20, 2014 18:14
|
|
|
He then immediately quotes Hitchhiker's Guide.  I have a new mancrush.
|
|
#
?
Jun 20, 2014 20:27
|
|
|
Now that the forums are back - I don't get Aereo, at all. That for the end-user it's similar to a cable provider is irrelevant - it's also similar to going to a rent2own store and renting the antenna and Tivo, and that's explicitly legal. Is it just a case of ~~internet~~ being a magic word in the way a really long video cable wouldn't have been? Edit: The worst part is that I have to agree with Scalia on this one, I just don't see how it qualifies as a performance without so widely stretching that term that it makes the entirety of cloud storage illegal. Harik fucked around with this message at 02:20 on Jun 26, 2014 |
|
#
?
Jun 26, 2014 02:16
|
|
|
Yay 4th Amendment.
|
|
#
?
Jun 26, 2014 02:20
|
|
|
ayn rand hand job posted:Yay 4th Amendment. I'll take a Good 4th Amendment Ruling over Aereo any day.
|
|
#
?
Jun 26, 2014 02:21
|
|
|
Harik posted:Now that the forums are back - I don't get Aereo, at all. That for the end-user it's similar to a cable provider is irrelevant - it's also similar to going to a rent2own store and renting the antenna and Tivo, and that's explicitly legal. Is it just a case of ~~internet~~ being a magic word in the way a really long video cable wouldn't have been? The opinion amounts to "This is a community antenna service that operates over the internet instead of over coax. Congress explicitly regulates CATV services in a specific way and having a bunch of little antennas in one location instead of one big antenna at one location is irrelevant."
|
|
#
?
Jun 26, 2014 02:23
|
|
|
hobbesmaster posted:The opinion amounts to "This is a community antenna service that operates over the internet instead of over coax. Congress explicitly regulates CATV services in a specific way and having a bunch of little antennas in one location instead of one big antenna at one location is irrelevant." Yeah, that's the catch with this case vs cloud service providers. Congress passed a law that more or less specifically deals with this situation.
|
|
#
?
Jun 26, 2014 02:24
|
|
|
hobbesmaster posted:The opinion amounts to "This is a community antenna service that operates over the internet instead of over coax. Congress explicitly regulates CATV services in a specific way and having a bunch of little antennas in one location instead of one big antenna at one location is irrelevant." I'm still not seeing it. If I just leased space that had good reception and provided internet access and SOLD antennas to people, would that be legal? Now they own the antenna, but they're renting the place to put it and the internet access to get the signal back to you. Because at that point Aereo just says "Buy an antenna and get one month free!" and nothing has changed technically, but it's legally a completely different scenario.
|
|
#
?
Jun 26, 2014 02:33
|
|
|
Harik posted:I'm still not seeing it. If I just leased space that had good reception and provided internet access and SOLD antennas to people, would that be legal? Now they own the antenna, but they're renting the place to put it and the internet access to get the signal back to you. The providing a cable to you for a fee is the problem thats what makes it cable TV.
|
|
#
?
Jun 26, 2014 02:35
|
|
|
duz posted:Yeah, that's the catch with this case vs cloud service providers. Congress passed a law that more or less specifically deals with this situation. In other words, any organization that offers group access would be illegal, including renting physical space and utilities in a room where everything else is customer provided (antenna, OTA-to-IP translation, etc.)? In other words, it would seem that the legal space for digital streaming of OTA broadcasts is about as wide as the anti-circumvention exceptions in the DMCA. Just as you can break CSS to get clips for commentary or criticism purposes, but it's still illegal to obtain the software to do so (i.e. everyone seeking to do so legally would also have to write DeCSS software on their own), you're allowed to send OTA signals from yourself to yourself, but you can't get help from anyone to do so, because any organized attempts (even renting space and utilities and leaving the rest, like installing the antenna and setting up a streaming box, to the consumer) would count as community-access...? EDIT: I mean I suppose if it was a general usage rental (e.g. rackspace for any equipment without specifically advertising use for streaming OTA signals) it might theoretically work because then it's multi-purpose and not actively seeking to be community-access television but just a general "hacker space" so to speak... ComradeCosmobot fucked around with this message at 02:51 on Jun 26, 2014 |
|
#
?
Jun 26, 2014 02:42
|
|
|
Yes. Congress literally passed a copyright act amendment that said "we want to make it so that anyone who is passing programming they capture OTA along a wire to customers has to pay the broadcasters." Any attempts to try and weasel around that by pretending that isn't what's going on are going to have to deal with the clear Congressional intent not to allow that. In contrast, cloud services, time shifting, etc. are not within narrow Congressional categories and won't have the same issues as a result. E: the legal space for digital streaming of OTA broadcasters is exactly as wide as the legal space for almost all works - you pay the rights holder and you can do it.
|
|
#
?
Jun 26, 2014 03:08
|
|
|
They are really going to announce their Hobby Lobby decision at the last possible minute, aren't they? You're killing me, Roberts Court.
|
|
#
?
Jun 26, 2014 03:19
|
|
|
ComradeCosmobot posted:In other words, any organization that offers group access would be illegal, including renting physical space and utilities in a room where everything else is customer provided (antenna, OTA-to-IP translation, etc.)? Private storage units are generally considered that -private. I think if you tried to set up a seed box at a library it'd probably be illegal but they'd probably also kick you out.
|
|
#
?
Jun 26, 2014 03:26
|
|
|
McDowell posted:They are really going to announce their Hobby Lobby decision at the last possible minute, aren't they? You're killing me, Roberts Court.
|
|
#
?
Jun 26, 2014 03:30
|
|
|
McDowell posted:They are really going to announce their Hobby Lobby decision at the last possible minute, aren't they? You're killing me, Roberts Court. I got a sandwich and a beer riding on this Roberts DON'T gently caress ME Also the Aereo case just makes me question why the Cable Companies get to use public airwaves for OTA broadcasts for free, they should have to pay for that poo poo if they're making Aereo do it. I think i'm just misunderstanding part of the argument though.
|
|
#
?
Jun 26, 2014 03:32
|
|
|
Grapplejack posted:I got a sandwich and a beer riding on this Roberts DON'T gently caress ME ... The cable companies do have to pay to retransmit OTA broadcasts, though. E: I can't find the post from my phone but I am 99% sure that I posted that Aereo was going to come down as, essentially, a retransmission consent case way back when. Kalman fucked around with this message at 03:38 on Jun 26, 2014 |
|
#
?
Jun 26, 2014 03:36
|
|
|
Grapplejack posted:
They don't, the networks do. The way it works is that the Networks (ABC, CBS, FOX, NBC) get permission from the FCC to transmit their programming OTA. However, they still reserve the right to keep other people from rebroadcasting those signals (so, copyright) unless you pay them. Cable companies work through a cable (self explanatory), so they don't need the go ahead from the FCC but they do pay the networks for their content so that they can offer it.
|
|
#
?
Jun 26, 2014 03:36
|
|
|
Grapplejack posted:I got a sandwich and a beer riding on this Roberts DON'T gently caress ME Don't forget that Canning is still outstanding in case you were betting on whether all NLRB regs for the past four years are null and void and whether almost all Obama recess appointees would be kicked out of office.  quote:Also the Aereo case just makes me question why the Cable Companies get to use public airwaves for OTA broadcasts for free, they should have to pay for that poo poo if they're making Aereo do it. I think i'm just misunderstanding part of the argument though. They do have to pay. That's why they opposed Aereo.
|
|
#
?
Jun 26, 2014 03:38
|
|
|
ayn rand hand job posted:Yay 4th Amendment.
|
|
#
?
Jun 26, 2014 03:38
|
|
|
Grapplejack posted:I got a sandwich and a beer riding on this Roberts DON'T gently caress ME Cable companies already pay OTA broadcasters for retransmission.
|
|
#
?
Jun 26, 2014 03:38
|
|
|
computer parts posted:They don't, the networks do. Ohh, okay. I confused the OTA regulations with something else. Whoops!
|
|
#
?
Jun 26, 2014 03:39
|
|
|
Grapplejack posted:Also the Aereo case just makes me question why the Cable Companies get to use public airwaves for OTA broadcasts for free, they should have to pay for that poo poo if they're making Aereo do it. I think i'm just misunderstanding part of the argument though. They do pay for it. It's why you'd occasionally hear things where cable companies would be in dispute with the networks, and so for a period you could get that network's programming if you hook up an antenna and get it over the air, but you couldn't watch it via your cable system. There was a big blowup over this last year, when Time Warner and CBS went at it and there was a month or so where any Time Warner subscriber had their local CBS networks blacked out coming from the cable provider.
|
|
#
?
Jun 26, 2014 03:40
|
|
|
|
| # ? May 9, 2024 18:45 |
|
|
Is there a consensus yet on the likelihood of the NLRB ruling being one way or another?
|
|
#
?
Jun 26, 2014 03:41
|
|