It's for cases like these I miss Groklaw. The reporting that we get now usually is either shilling/biased or too low on details to get a clear picture of what went down, as patents and lawyerese usually result in a vague mess most of the time.
I hope Google knocks 'em out, but sadly the US legal system has 'surprised' me a bit too often to feel confident in the outcome.
I have to disagree, I find Groklaw extremely biased as well. It just happens that the bias seems to fit with your personal biases as well, so it appears like good coverage. This was especially true during Samsung vs. Apple, or with anything to do with Microsoft, or with defending Motorola's abuse of FRAND patents.
Groklaw had a clear bias and that could be corrected for. Their clear explanation of legal minutiae such that even a layman could understand it, and their attention to detail is why I follower their coverage.
I do not believe there is a completely unbiased source of news, No true Scotsman and all that...
Groklaw may have had a bias but you could always tell where Paula (?) was coming from and she doesn't appear to have taken money from anyone involved in the cases.
Other bloggers on the other hand seem very biased, are paid to consult on this stuff and don't disclose who's paying them for what.
I'm not sure how accurate Groklaw was during the SCO case, but as someone who understands patents more than the average geek: Groklaw was horrible when it came to patent matters. Let alone the obvious bias and snark, PJ was frequently either deliberately misleading or, if you want to be generous, completely confused.
I personally would not be so generous because I saw her use underhanded methods too many times, like quoting things out of context or cherrypicking sentences or even blatantly saying something completely unsupported by the very documents she refered to. And of course, when she could not twist things to match her agenda, she'd make subtle insinuations of corruption. Like the time when a judge refused to rule against Microsoft in some case, "Isn't it weird this judge flew out to take a case that's not in his jurisdiction?" -- conveniently overlooking the fact that the judge in question had 1) more experience in the issues at hand and 2) ruled against Microsoft in a previous case.
I always knew that patent trolls were a problem, and that the US patent system had big issues, but these shows made it oh-so-clear just how bad things are.
Every time someone links to those episodes I feel compelled to chime in and note that they (well, the first one, at least) are a terrible piece of reporting about patents.
1. Read through the transcript of the first episode. Note that the word "claims" shows up not even once.
1. a) If you don't know why that's a problem, you're in no shape to judge the patent system either.
2. All the patents at the M-CAM guy says are "on the same thing" are actually not. I know because I read them all.
3. There is no such thing as "a patent on toast". It's a downright shame that the authors didn't think to double check that with someone who, y'know, real understands how patents work. But oh look, amongst other things, M-CAM sells "patent risk management" services! No, not a conflict of interest there at all.
4. They talk to engineers (like many you'll find on HN) to get their views on patents. And like most here, none of them seems to understand how patents work. (Again, see point 1. a) above.) What do you expect will happen when you ask a Blub programmer his views on Haskell?
5. In the first episode, they insinuate that IV doesn't pay inventors like they claim. In the second episode, turns out that they do pay quite well, but oh look! that inventor is totally a schmuck for ripping off his coinventors! (Is he really the kind of inventor that IV is paying?!)
>"4. They talk to engineers (like many you'll find on HN) to get their views on patents. And like most here, none of them seems to understand how patents work. (Again, see point 1. a) above.) What do you expect will happen when you ask a Blub programmer his views on Haskell?"
The engineers understand how patents work, and how the methods work, they are puzzled by the patents' language. This is significant because a patent is supposed to describe a method; when an expert cannot use a patent to learn the method, a large part of the patent (everything except the claims) is useless and wasted.
>"3. There is no such thing as "a patent on toast". It's a downright shame that the authors didn't think to double check that with someone who, y'know, real understands how patents work. But oh look, amongst other things, M-CAM sells "patent risk management" services! No, not a conflict of interest there at all."
Sounds like a toaster where you move the bread away from the element rather than turn the element off. E.g. what happens in thousands of bread/toast factories around the world.
And not having looked at the spec, it is possible that the claims might be construed in a way that won't cover how you or I would usually make "toast."
>> Note that the word "claims" shows up not even once.
That word actually shows up exactly once in each episode.
>> In the first episode, they insinuate that IV doesn't pay inventors like they claim. In the second episode, turns out that they do pay quite well, but oh look! that inventor is totally a schmuck for ripping off his coinventors! (Is he really the kind of inventor that IV is paying?!)
Yes, IV paid Chris Crawford -- the one example given by IV of an "inventor" who made money -- for a stake in a patent. That's what IV does. They buy patents or percentages of patents. TAL never claimed otherwise. Crawford retained a percentage of interest in ongoing revenue obtained in court, which was distributed accordingly.
As to whether Chris Crawford is a schmuck and ripped off others, or whether the invention in question was even his to sell or was his idea to begin with, I think this court transcript from a case that he lost, speaks for itself:
---
Attorney: So the first paragraph of this document reads, quote, "This proposal, it is in response to Jack Byrd's idea to provide automated offsite backup services for PC users." And aren't you in the second sentence saying that Jack Byrd had an idea to provide automated offsite backup services for PC users?
Chris Crawford: No. What I'm saying is that Jack Byrd had an idea of me pursuing the automated offsite backup services that I had--
Alex Blumberg: Chris Crawford then launches into a rather surprising explanation for how the sentencing that this business was, quote, "Jack Byrd's idea," doesn't actually mean that the business was Jack Byrd's idea. His explanation? He was using the apostrophe S incorrectly.
Chris Crawford: As far as the apostrophe S-- and I'm still not clear if you're trying to derive a specific meaning with respect to the apostrophe S--
Alex Blumberg: Of course, what would a sentence reading, "in response to Jack Byrds idea" even mean if the S was plural?
Attorney: I'm asking you, though-- you certainly know what the use of an apostrophe S means, do you not?
Chris Crawford: [SNIFFING] As I've written documents over the years, there are times when I use an apostrophe S, and it seems like I'm supposed to use an apostrophe S. But I have to say that my grammar is not strong enough to tell you right now with clarity when an apostrophe S is used.
Hmm, I'm been unable to respond because I get a "you're submitting too fast" error. Maybe I tripped some sort of flame detector. To avoid multiple replies, I'm consolidating responses into one mega-reply.
davesims:
> The word ['claims'] appears exactly once...
Yes, I worded my argument poorly. The word "claims" does occur, but it's not in the context of a patent's claims, but rather the claims made by a plaintiff. They discuss multiple patents in the episode, but not once do they talk about any patent's claims, and no discussion of a patent can begin without the claims. In fact, the journalists seem to make the very common mistake of interpreting a patent's scope based on what the abstract and other sections mention.
> TAL never claimed otherwise.
Oh, they very much insinuated it by emphasizing how difficult it was to verify instances of inventors getting paid.
Also, yes, the inventor was a schmuck, but that should have no bearing on IV or other trolls who claim to help inventors. The vast majority of inventors don't try to rip off their co-inventors. (Glass half-full: because most people are decent human beings. Cynically: incorrect inventorship, as in this case, can automatically invalidate a patent. If you search askpatents.com, you'll see this issue come up a few times.)
nickff:
> Bread refreshing method US 6080436 A
The main claim in that patent, as pointed out by belorn, requires a temperature of 2500 F to 4500 F. Toasters typically operate at temperatures at 310 F. I don't know what you get at 2500 - 4500 F, but it's not toast. The thing is, you don't even need to be a patent lawyer or an engineer to fact-check this little but.
> The engineers understand how patents work...
This, in my experience with multiple online forums and countless personal interactions over the course of 7 years, is rarely true. Pretty much nobody, especially the tech media making the most noise about patents, even knows what claims are.
I agree that patentese seems esoteric at first glance. But that is because it has this particular structure for legal and historic reasons. It's easy enough to learn, though. If you've had to debug C++ STL compile errors, patentese is a breeze. This was the intent behind my Blub/Haskell comparison: complaining about something wihout understanding it is really not contributing much to the discussion.
reitzensteinm: > Lodsys
In my eyes, Lodsys is clearly a bad actor. They are preying on solo developers who have no way to afford any kind of defense. Many trolls exploit the extreme cost asymmetries of mounting a legal defense, but by targeting such small players, Lodsys is taking it to new lows. In that sense I find IV to be more palatable because at least they pick on "someone their own size". (Lodsys has been linked to IV, but until some time ago, Lodsys had a page on their site explicitly disavowing any relation to IV. No idea why it's been taken down.)
Also, a general note about TFA: I don't see what's so landmark about this trial. Generally, trials are "landmark" if they break new ground, and while interesting because the defendant is formidable, this is a pretty run-of-the-mill lawsuit of IV vs another tech giant.
I think that the TAL episodes were great at exposing the fact that there are patent trolls, and that there are problems with the current patent system. But your descriptions and analysis do point to some serious flaws with their reporting and explanations.
And yes, I remember first hearing the episode about the patent invalidation, in which everything hinged on someone's handwritten notes that happened to be written down. That didn't point to the fundamental issues here.
I think that a lot of engineers, especially HN readers (and including myself), are interested in the world of software patents, but lack the legalistic training to think through things the way you described.
The claims are that the heating element is set between 2500F and 4500F (claim 1b). The bread itself is held away from the element (sheets 2, 3), separated by air (which is not a great heat conductor), for between 3 and 90 seconds (claim 1c). Sheet 4 indicates that the 'cut fuzz & peaks' of the exposed face of the bread becomes toasted.
This patent describes a method of toasting bread using an electronically-controlled fast toaster.
Instead of hand-waving about the conductivity of air, what would have helped is a citation that clearly states how hot the heating element of toasters themselves get. (Really. Because I could not find a good cite either :-P)
The best I coud do was wiki answers: The heating element of toasters reaches 1100 - 1200 F [1].
If the answers.com link is not trustworthy, consider this: The most common alloy used for the heating elements in toasters is Nichrome [2]. Nichrome is used because it has a "high" melting point of 2550 F [3].
Not only does the temperature range required by this patent not make toast, it would melt the heating element in most toasters. The spec of the patent itself requires some kind of halogen lamp heaters.
It's pretty clearly not a "patent on toast". "Burnt to a crisp" toast, maybe, but not toast.
Before 1990, people weren't suing each other over software patents. And the software industry did just fine - it was a period of huge innovation and rapid progress. We know what the software development landscape looks like both with and without patents.
It might not have been the booming industry it is today but patent trolls existed in the 80s, too. Here's John Walker, one of the founders of AutoDesk, writing in 1993:
“Ever since Autodesk had to pay $25,000 to ``license'' a patent which claimed the invention of XOR-draw for screen cursors (the patent was filed years after everybody in computer graphics was already using that trick), at the risk of delaying or cancelling our Initial Public Offering in 1985, I've been convinced that software patents are not only a terrible idea, but one of the principal threats to the software industry.”
I believe the XOR patent was the first major one, and certainly the most blatant troll – it hit most of the industry for a technique which had been described a decade before filing – but the software patent mess started arguably in 1981. Wikipedia has timeline of the evolving caselaw:
Right, those were copyright claims, not patent claims. Apple also sued Microsoft over look & feel, again a copyright claim.
Ironically, Lotus 123 was more or less a clone of VisiCalc's ideas. Lotus 123 would never have been able to exist if software patents had been there at the time.
Those were copyright claims, true, but that doesn't change the fact that companies still sought to protect what they felt were their innovations. There were many other cases involving many other companies, such as Sega vs Accolade. The only thing that has changed is companies have realized patents are a better means to that end because copyright does not, in their eyes, afford sufficient protection.
>Lotus 123 would never have been able to exist if software patents had been there at the time.
It's impossible to say that without considering what patents they'd have been able to secure in the first place, and even then it's almost certain they'd have existed. Software patents exist today and there is no shortage of clones in the market, even for products that had been heavily patented.
> The only thing that has changed is companies have realized patents are a better means to that end because copyright does not, in their eyes, afford sufficient protection
No, what changed was that the Federal Circuit invented software patents.[1]
You make it sound like companies woke up one day and realized that patents on software and business methods were a way better idea than everything they had tried before. What actually happened was that the few software and business method patents that were granted were never upheld. Until the Federal Circuit was founded.
The economic rationale for patents just isn't as strong when applied to software. Patents exist to protect capital intensive R&D from free riding. Software designs are just not very capital intensive. Plus, software has a layer of protection on the basis of copyrights. And that state of affairs manifests itself in the empirical observation: software worked find before widespread software patents.
The real problem is that the Patent Act isn't formulated in terms of the economic rationale for patent protection (it predates much of that economic theory). Thus, its hard to draw legal lines using the language of the patent act that are obvious when expressed in the language of economics. And now that the legal mechanics of patent protection are becoming unmoored from the economic underpinnings, the schism is noticeable to everyone. Note that in fields where there isn't this gap between law and economics (fields where R&D is very capital intensive), there isn't this same feeling among practicing engineers that patents are a problem.
Probably one of the most pernicious aspects of software patents is anyone writing code has no idea if they are violating patents or not. Like the old Roman practice of decimation, you can be randomly destroyed and there's nothing you can do to protect yourself against it.
Yes exactly, that is just an indication that the patent trolls hadn't completely caught up to the industry in 1990. People have been doing patent trolling for a long, long, loooong time. Like in the 19th century, you had patent trolls suing ranchers over using barbed wire that they supposedly patented. It'll take awhile before the law catches up and stomps out the craziness going on now.
I fully understand the sentiment, but let me try to hack your comment, and play the devil’s advocate:
* Law works in mysterious ways: a victory for blatant reasons might be expressed with a rationale from the judge that triggers worst outcome. I hate the laws because of that (or rather, lawyers for being so clueless about economic assumptions behind recommendations) but that makes every nuanced decision so much more enjoyable.
* A victory from the big, profitable company against a new business model (OK, I might have forgotten a moral decrepitude from one party in that description; thing is: law often does, too) means that jurisprudence might end up appearing a lot more in favor of winners, i.e. against competition and for established monopolies. Might: let's see what gives, but please remember that having a vivid tech scene does rely on a handful of counter-intuitive decisions that were made against allowing Standard Oil to offer cheap gas, against Microsoft for offering a simple, working solution, against Google simplifying their privacy statements. Competition law is counter-intuitive and unsatisfying.
Of Microsoft's simple solution was "working" (something like the original Chrome before Google made it into IE:The Next Generation with their proprietary pile-ons, the gnashing and wailing would not have happened. It was their proprietary crap defacto monopoly browser that led to the complaints.
Technically, the quality of the browser was immaterial to the EU Competition authority investigation, and a good browser (as it was at the time the investigation was suggested, an eon ego by tech standards) might have made things worst. The political support for the investigation came later, wasn't neutral to the outcome and was then triggered by the quality of the software. Administration works in mysterious ways, and the European Commission’s are strangers than Cthulhu's.
I fully appreciate your intention to moderate the belligerent tone with reason and a Kantian admiration for applying the law as it is. However, I believe that the contention with Intellectual Ventures is that they hold patents that many say were awarded in spite of what was considered at the time (and far more so now) common concepts. It seems that voiding those patents is unreasonably hard with the law as it stands. Therefore, I hope that the judge considers not just the case, but also the industry-wide consequences, and accepts that what is commonly accepted in tech, for the greater good, are close imitation of previous designs. If I could influence the decision, I would gladly point at Haute-cuisine and their original idea-defending mechanisms.
This is a big test for IV, but I am uncertain that it will individually change the course of affairs for the general patent question. The second This American Life episode cited by Reuven includes a good indication of why. These cases seem often to be examined not on underlying principle, but on the direct evidence. The 'network update' patent from TAL was busted not because the patent itself was found to be ambiguous or over-broad but because its authorship was incorrectly assigned to only one of its 'inventors'. Super narrow and no principle other than "check your application more carefully next time".
In my opinion, this is how it should be. Courts should enforce the law, not change it (excepting the Supreme Court's right to reject the unconstitutional). For that, we need to work through Congress to modernize the definition of patents in respect to software.
These cases play an important role in that process, which is to create a body of guidance on how software patents are being used and how they are being challenged, for statutory reform and/or legal appeal.
[[As is probably obvious, I am not a lawyer, so I apologize for any legal inaccuracies in the above]]
Courts are to interpret the law. The legislature says "all horses are animals." The courts have to answer "what about baby horses? dead horses? are things that are not horses also animals or can only horses be animals? is a zebra a horse? what about a horse embryo? who can decide what is a horse or not? etc etc." The courts have to answer these questions in such a way as to make life as least shitty as possible. So yes, the validity of shitty patents is totally up for the courts, but of course the legislature can weigh in any time.
The suit in question (11-908 in Delaware federal district court [0]) was filed in October of 2011. It is coming to trial in 2014 and the participants have probably spent well over a million dollars each getting to this point. That alone tells you all you need to know about the chances that a startup company has in getting justice or even just surviving against a garbage patent abuser in court.
Google did not own Motorola when the case was filed and Google is not a named defendant, contrary to the deliberately misleading headline.
The Reuters article doesn't say anything about what the six patents claim or even give their numbers. Those numbers are 7,810,144 (the '144 patent), 6,412,953 (the '953 patent), 7,409,450, 7,120,462, 6,557,054, and 6,658,464.
The Reuters article does mention that the accused devices are older Moto phones and accessories. Newer devices than 2011 are, of course, not in the original complaint and usually are not added at this late stage.
The '144 patent is for a "File Transfer System for Direct Transfer Between Computers" -- describing a subset of SMS or IRC style messaging -- filed in 1997 and repeatedly rejected until the PTO folded under seven continued amended petitions in 2010. There is no limit to the patent amendments an applicant can make and no such thing as a final rejection at the PTO. Applicants are encouraged to reword claims to cover technology invented by others since the original application. [1]
If you'd like to be infuriated about how the patent system abuses software and computer networks, read the linked [0] opinion starting on page 14 where claim construction begins.
The '953 patent was filed in 1998 vaguely describing long established LCD backlighting practice. The '054 patent, "Method and System for Distributing Updates by Presenting Directory of Software Available for User Installation That Is Not Already Installed on User Station," was filed in 2000. The '464 patent, "User Station Software That Controls Transport, Storage, and Presentation of Content from a Remote Source," is like iBooks or Archie and filed in 1994.
The '450 patent, "Transmission Control Protocol/Internet Protocol (TCP/IP) Packet-Centric Wireless Point to Multi-Point (PTMP) Transmission System Architecture," was filed in 1998. It describes the use of any QoS scheme that includes packet inspection over a wireless network.
The '462 patent, "Portable Computing, Communication and Entertainment Device with Central Processor Carried in a Detachable Handset," filed in 1999 describes handheld computers with a docking station. It does not describe how to build such a device, of course, just that one could exist.
> and repeatedly rejected until the PTO folded under seven continued amended petitions in 2010
Also called "prosecution". This is, literally, how every patent is prosecuted at the USPTO: the applicants file claims as broad as they dare, examiners reject using prior art (or other statutory reasons), after which the applicants amend claims to sidestep cited reasons, and rinse-and-repeat, until the applicant finally presents a set of claims that the examiner cannot reject. It is not "folding", it is being procedurally unable to issue a rejection.
The only catch is that examiner time is not free, so each attempt costs some thousand USD, and gives you one non-final rejection followed by a final rejection -- which is, of course, not final, because the applicant can just pay the requisite fees to continue examination. It's not like the applicant gets multiple bites at the apple for free.
> Applicants are encouraged to reword claims to cover technology invented by others since the original application.
Funny, but no. The examiner would like nothing more than getting rid of a case for once and for all, and they can't make frivolous allowances either. (Which is why you get cases like the one above where the applicant had to try seven times to get an allowance.) They can very much make frivolous rejections, though, and they do. There really is no feasible recourse for those other than coughing up more fees and forging on.
The reality of continuations is, applicants have the right to reword claims as long as the original supports the claims, and many applicants exercise that right. Because, if they find something out there that is described in their patent but sidesteps their claims, they'd be outta luck. The Doctrine of Equivalents goes only so far, especially when juries are involved.
Continuations can be abused, but it does not mean they always are. Unless new matter is added, the new claims must still be constrained by the spec of the original patent. (If the new claims refer any new material, their priority dates effectively change to the date the new matter was introduced, automatically bring into play all the new prior art introduced in the meantime as well.) So the claims in continuations are still judged in context of the prior art present at the time the spec was filed.
>The '462 patent ... It does not describe how to build such a device, of course, just that one could exist
I dunno, the figures look pretty descriptive to me:
I agree, however, that it is very instructive to read through the claim construction arguments, just to get an idea of how these things shake out. It is all technical and reasonably easy to grasp, so if you have a few minutes free, go through a few pages of this. (Any more and you run the risk of dozing off.) I couldn't find examples of "abuse of software and networks", though.
Software patents: taking very broad, simple, and obvious ideas, expressing those ideas in technical language to impress the patent examiners, suing everybody who independently discovers and uses the idea.
I hope Google knocks 'em out, but sadly the US legal system has 'surprised' me a bit too often to feel confident in the outcome.