TE
科技回声
首页24小时热榜最新最佳问答展示工作
GitHubTwitter
首页

科技回声

基于 Next.js 构建的科技新闻平台,提供全球科技新闻和讨论内容。

GitHubTwitter

首页

首页最新最佳问答展示工作

资源链接

HackerNews API原版 HackerNewsNext.js

© 2025 科技回声. 版权所有。

SparkFun Hooks a Patent Troll

803 点作者 mkeeter将近 4 年前

32 条评论

kens将近 4 年前
I like SparkFun and I don&#x27;t like patent trolls, but I have to say that SparkFun&#x27;s mockery of the patent doesn&#x27;t make sense. It is pretty clear that this patent was written by an engineer describing their actual system in great detail followed by one fairly narrow claim. I&#x27;ve seen many patents that are mostly written by lawyers, and I don&#x27;t see any influence of lawyers in this patent.<p>SparkFun&#x27;s criticism is kind of painful to read. The diagram that SparkFun claims is SPI is not SPI. The discussion that SparkFun claims is malloc is entirely unrelated to malloc. SparkFun claims that when the patent discusses processing multiple streams of data that it&#x27;s discussing variable length arrays, but it&#x27;s not. They claim that a MAU doesn&#x27;t exist, but a MAU is a multiplier-accumulator unit. SparkFun seems to be entirely confusing the motivation of the engineer filing the patent in 2008 with the motivation of the patent trolls. I&#x27;ve seen many patents that are fluff and nonsense, but this one is a solid patent.<p>The infringement claim is probably bogus; I don&#x27;t see any evidence that the pcDuino infringes the one narrow claim in the patent. I hope that SparkFun can quickly get rid of this lawsuit. I don&#x27;t get any enjoyment out of criticizing SparkFun&#x27;s post, but having spent the time to look over the patent, I feel that I should comment.
评论 #27695769 未加载
评论 #27697306 未加载
评论 #27699060 未加载
评论 #27699011 未加载
评论 #27698070 未加载
评论 #27696224 未加载
ohazi将近 4 年前
&gt; &gt; The cost is further reduced by employing only static or dynamic ram as a means for holding the state of the system. [...]<p>&gt; Dynamic ram? Can I poke fun at them for not knowing what RAM is? Yes I can.<p>&gt; The above paragraph succinctly describes malloc(), which is perhaps one of the most fundamental advancements in modern computing. The problem is that Al didn’t invent it. Memory management (the ability to take RAM and then give it back when your function is done) goes back to the dawn of computing.<p>No, this is wrong.<p>Nathan is referring to <i>dynamic memory allocation</i>, which is what malloc does, but that isn&#x27;t what this line in the lawsuit is talking about.<p>Dynamic RAM, or DRAM is RAM that stores state in a capacitor. This is what your laptop uses -- DDR4 SDRAM is &quot;Double data rate (version 4) synchronous <i>dynamic RAM</i>&quot;. In DRAM, the charge on the capacitors slowly dissipates, so DRAM needs to be &quot;refreshed&quot; periodically, which is why it&#x27;s referred to a &quot;dynamic&quot; RAM.<p><i>Static</i> RAM, or SRAM, is RAM that stores state in a register-like structure (usually cross-coupled inverters). It&#x27;s &quot;static&quot; in that it doesn&#x27;t need to be refreshed -- as long as the supply voltage is maintained, the bits stored in an SRAM will remain intact.<p>I agree that this patent lawsuit is bullshit, and I really hate to derail the conversation, but I&#x27;m surprised by Nathan&#x27;s misunderstanding here, given that SparkFun makes embedded boards that use all kinds of memory. SRAM&#x2F;DRAM isn&#x27;t unusual terminology. The distinction doesn&#x27;t even matter in the context of this line in the lawsuit, but if you&#x27;re going to make a point of nitpicking about something, you should really make sure that you&#x27;re actually right about it first...
评论 #27695167 未加载
评论 #27694784 未加载
评论 #27694802 未加载
AlbertCory将近 4 年前
Me: software engr, retired; also patent agent who worked in Google Patent Litigation.<p>A rookie mistake is reading the TEXT of a patent. A pro looks only at the claims. The text is only there to &quot;support&quot; the claims, which boils down to using the same words. Cynical, I know.<p>Claims unfortunately look like gibberish if you&#x27;re unschooled, but &quot;construing&quot; them is an early step in a trial, and often if you get a bad claim construction you give up. I worked on tons of claim construction briefs.<p>I&#x27;ll read those claims a little later this evening.
评论 #27696001 未加载
评论 #27696391 未加载
评论 #27695613 未加载
Taniwha将近 4 年前
OK - so I vaguely skipped through the patent looking for claims - mostly it&#x27;s a description of some guys possibly novel media processor architecture, remember the description is not the patent, the claims are. I could only find a short claims section at the end (1.5 pages out of 56).<p>This doesn&#x27;t read like most patents which tend to be lower on the details, and longer on the claims, I suspect that this was either written directly by the engineer, or written to effectively publish the details of the design to avoid others patenting their stuff out from under them (or both). Probably the original company went under and the patent was onsold to the troll.<p>The actual claims have nothing to do with SRAM vs. DRAM (or even malloc) - my (lay) reading of these claims are that you have to have:<p>- memory containing instructions and data - more than one processor containing a multiplier, an ALU, a logic unit, a bit processing unit - all the media processors running in parallel<p>this does describe pretty much any modern multicore chip, or graphics chip, and perhaps more closely Sony&#x27;s Cells for example.<p>As a sometimes computer architect having a bunch of CPUs execute code and data out of shared memory is &#x27;obvious&#x27; (and was at the time, I was building media processors before 1997, and frankly was 30 years before). As far as prior art - I&#x27;d consider any multi-cpu mainframe starting in the 1960&#x27;s onwards to be valid prior art here<p>If I were Sparkfun I&#x27;d pull in Arm, Intel, AMD, nVidia, Sony, IBM .... and nuke this thing from orbit
评论 #27696027 未加载
ww520将近 4 年前
(Note that none of the following is legal advice.)<p>One tip on reading patent - ignore most of the diagrams and description, and go straight to the claims. The claims are the only valid part of the patent. In this case, all the static&#x2F;dynamic RAM, signals, memory cache, and malloc stuffs are irrelevant and misdirection.<p>This patent has only 1 claim [1]. The claim has a series of elements making a number of conditions for the patent. Notice that semicolons separate each claim elements. That means these are AND-conditions. All these conditions need to be met to fit the patent. Also notice the colons. These list the sub-elements or sub-conditions.<p>Reading the claim. It seems to be a system with a number of processors of different kinds. There&#x27;s some memory with input&#x2F;output ports. The memory holds the instructions and data. The input&#x2F;output ports of the processors connect (coupled) to the input&#x2F;output ports of the memory to read the instructions and data. Do computation on the data based on the instruction. Spit out the result. The kinds of processors are: multiplier, arithmetic unit, arithmetic logic unit, and a bit manipulation unit. The processors run simultaneously.<p>To defeat this patent, you can list prior arts. This is the most damaging because it invalidates the whole patent. Or you can go by the claim conditions. You can either claim your product doesn&#x27;t have any one of the kind of the processors (no bit manipulation), or the processor doesn&#x27;t read instruction&#x2F;data from memory (instructions are hard-coded or a processor gets its input from another processor), or the processor doesn&#x27;t connect to the memory directly (via a bus or via a data switch or via special registers), or the processor doesn&#x27;t do computation according to the instruction (e.g. a doubler always double the data), or your processors run serially in some fashion, etc.<p>And then there&#x27;s the legal definition of the terms in the claim, such as what is a multiplier, an arithmetic unit, an ALU, what&#x27;s a bit manipulation unit, and what&#x27;s &quot;coupling.&quot; This process is called claim construction and done by the court via precedents and the intents described in the description of the patent. Many times a patent lawsuit falls apart during the claim construction phase because the definition of the terms don&#x27;t apply to the sued product.<p>[1] <a href="https:&#x2F;&#x2F;patents.google.com&#x2F;patent&#x2F;US6289434B1&#x2F;en" rel="nofollow">https:&#x2F;&#x2F;patents.google.com&#x2F;patent&#x2F;US6289434B1&#x2F;en</a>
评论 #27695978 未加载
评论 #27695629 未加载
评论 #27695653 未加载
评论 #27696729 未加载
BeetleB将近 4 年前
BTW, for anyone who hasn&#x27;t heard it, the This American Life episode linked to is absolutely worth listening to (or just read the transcript).<p>They should link to the update to that story:<p>Transcript: <a href="https:&#x2F;&#x2F;www.thisamericanlife.org&#x2F;496&#x2F;transcript" rel="nofollow">https:&#x2F;&#x2F;www.thisamericanlife.org&#x2F;496&#x2F;transcript</a><p>Audio: <a href="https:&#x2F;&#x2F;www.thisamericanlife.org&#x2F;496&#x2F;when-patents-attack-part-two" rel="nofollow">https:&#x2F;&#x2F;www.thisamericanlife.org&#x2F;496&#x2F;when-patents-attack-par...</a><p>One key takeaway: If you yield to a patent troll and agree to give them some percentage of your earnings, and later that patent gets invalidated - you still have to keep paying.
评论 #27695340 未加载
cycomanic将近 4 年前
I&#x27;ve said it before, but if you support the patent system, but say patent trolls should be abolished you are essentially saying that patents are only for big corporations to protect their turf from new incomers.<p>One argument is often patent trolls don&#x27;t make anything themselves. So someone who makes a truly groundbreaking invention but for some reason (e.g. it requires huge capital) does not want to create a company for manufacturing it, should not be able to use patents? Also if they would actually make a company to build it and then try to go after a big player who infringes on their patent, if the big player doesn&#x27;t want to pay the big player would most certainly win. They would just bring out lots of their patents that the small player supposedly infringes and thus make the suit to expensive to win. Essentially, if being an actual &quot;builder&quot; becomes a requirement, only big players will profit, because they have the pockets to kill off any small competitor through costs.<p>The other argument is patents are to bogus&#x2F;general. Which is true, but what is the alternative, make the patent investigation even more elaborate and costly (it already is very expensive), so again only the big players gain.<p>The problems people point to with patent trolls, are problems inherent in the patent system. Patent trolls are just a convenient boogey man to blame so the big players can keep a system that disproportionately be if its them. There is virtually no benifit to the patent system overall and it should be abolished.
评论 #27698487 未加载
评论 #27698900 未加载
评论 #27698530 未加载
评论 #27699706 未加载
评论 #27698808 未加载
fossuser将近 4 年前
I wonder if it&#x27;s possible to make some sort of patent troll insurance company.<p>Basically a company that attempts to collect as many patents as possible to weaponize against trolls and then provides legal protection to clients from trolls. The clients can share their patents with the insurance company too to help them become more powerful.<p>Then when sued by a troll the insurance company goes scorched earth to destroy them as a matter of policy. So their clients get the benefit of also deterring suits just because they&#x27;re represented by this insurance company.
评论 #27695199 未加载
评论 #27695858 未加载
评论 #27696014 未加载
评论 #27694759 未加载
评论 #27694889 未加载
评论 #27694795 未加载
评论 #27695677 未加载
评论 #27694746 未加载
评论 #27735210 未加载
评论 #27695466 未加载
评论 #27696315 未加载
评论 #27694970 未加载
cormacrelf将近 4 年前
Altair Logic actually sued Asus as well in 2019, after the patent had expired in 2018, alleging that Asus sold products infringing exactly this patent 6,289,434 before then. Asus tried, unsuccessfully, to get it dismissed, so we&#x27;ll see where it goes later.<p>But the dismissal judgment is enlightening about what happens when you sit on a patent for 20 years and then start to sue people. The &#x27;marking requirement&#x27; referred to is that you stamp products with a US patent number.<p>&gt; The &#x27;434 Patent was valid for 20 years from the filing date, meaning that it expired on February 27, 2018. See 35 U.S.C. § 154(a)(2). Because Altair did not bring suit until after the patent had expired, it is entitled to relief only if it can ultimately establish notice of infringement prior to the patent&#x27;s expiration date. Here, Altair would need to prove either that it satisfied the marking requirement or that neither it nor any of its licensees ever made a product which needed to be marked.<p>It seems the fact of never having licensed your patent to anyone allows you to avoid having to give notice to infringers. I&#x27;m guessing the damages limitations around non-marking are drafted to avoid a situation where someone sees a product, sees it and copies it thinking that because it has no patent number, it hasn&#x27;t been patented. There appear to be other &#x27;constructive notice&#x27; provisions. But it seems a little perverse that you&#x27;re at an advantage, notice-wise, as a patent troll if you avoid the marking requirements by never attempting to license the patent to anyone in 20 years, and separately nobody takes a second look when searching for patents applicable to their products. Normally both those things point to the patent not being useful or valuable from a licensee perspective.<p><a href="https:&#x2F;&#x2F;casetext.com&#x2F;case&#x2F;altair-logix-llc-v-asus-computer-intl" rel="nofollow">https:&#x2F;&#x2F;casetext.com&#x2F;case&#x2F;altair-logix-llc-v-asus-computer-i...</a><p>(Morbid detail: the patent was issued on September 11, 2001. Somehow the USPTO worked through that day.)
civilized将近 4 年前
I read the patent claim and ignored the rest, as the pros in these threads have been recommending that we do. I paid careful attention to the colons and semi-colons. There is one claim and it is short and not hard to understand. Take a look for yourself, you should be able to get the gist in well under 5 minutes: <a href="https:&#x2F;&#x2F;patents.google.com&#x2F;patent&#x2F;US6289434B1&#x2F;en" rel="nofollow">https:&#x2F;&#x2F;patents.google.com&#x2F;patent&#x2F;US6289434B1&#x2F;en</a>.<p>To me, it looks like the patent is for.... &quot;hooking up memory together with several processors in a system, and having that system process &#x27;media&#x27; data in parallel&quot; with &quot;processor&quot; defined as &quot;multiplier unit + arithmetic unit + arithmetic logic unit + bit manipulation unit&quot; (quotes mine, not from the patent text). There&#x27;s also a bit of stuff about how the arithmetic logic unit and the bit manipulation unit should be able to &quot;operate concurrently&quot; with some of the other units. It seems like this patent is basically for the idea of parallel processing on modern computers in general. If so, most of the computing industry infringes this patent today.<p>I&#x27;m not a hardware expert, though, and it&#x27;s possible I&#x27;m missing something and the claim is more narrow. It depends on the exact definition of those processor components and whether this ability of some CPU sub-units to &quot;operate concurrently&quot; with others is common on modern computers.<p>All that aside, the patent seems to have no intellectual value whatsoever. I have no idea what the history of parallel processing is, but it definitely predates the writing of this patent in 1998. The only thing that seems even possibly new is a tiny bit of text specifying which units should be able to operate concurrently with other units. Is that supposed to be the novelty that we&#x27;re protecting with the power of the law for 20 years?<p>The point of patents is that the public benefits from novel invention ideas being published, in exchange for the publisher getting exclusive rights to the idea for a few years. The public can build upon the idea sooner if it&#x27;s published earlier. In what universe do we imagine that someone reads this patent claim and is enlightened and inspired to build upon it?<p>How is this patent anything but yet another insane, horrific abuse of the system?
评论 #27698715 未加载
JumpCrisscross将近 4 年前
I wonder if one could make trolling patent trolls self-sustaining.<p>For example, let’s say you get hit with a demand letter. I review it and agree to finance your case. Using my resources, I prepare to invalidate the troll’s patents <i>unless</i> they settle for some amount less than their expected lifetime value from said patents. This gets split between you and me (after litigation costs), with maybe some of my share reserved for the eventual holdout who fights and gets their patents invalidated.<p>Less emotionally satisfying than invalidation. But far more scalable and thus threatening to the model.
评论 #27694764 未加载
评论 #27694742 未加载
评论 #27700929 未加载
评论 #27695868 未加载
croo将近 4 年前
The comments are about the contradictions between the patent and the blog post and the technical discussion are great. I would point out another angle.<p>I had a loud and irritating neighbour who patronized me (everyone) and I had hard time coping until I gave the guy a stupid nickname. When I was able to laugh the whole ordeal went away and no longer stressed me. This blog post is the same: it helps SparkFun shifting the seriousness of the threat into something laughable so it can be handled in a calm manner as it should be and to not loose another night of sleep over it. How exact or rational are the snarky remarks about the pattern doesn&#x27;t really matter.
W-Stool将近 4 年前
Did I miss in the article what the author plans to do about this? It is one thing to poke fun at patent trolls, but when you are the target of the troll legally you need to do something. What&#x27;s their response going to be?
评论 #27696545 未加载
评论 #27695280 未加载
desine将近 4 年前
Interesting read. Much more in-depth bullshittery than the usual patent troll. I love sparkfun, I&#x27;ve spent far too much money there building small trinkets, but I like supporting a local independent shop, and getting quality builds, rather than waiting and risking lower quality direct from Shenzen. I&#x27;m almost glad they got hit by the troll, because they&#x27;re one of the companies with the wherewithal and life to blow them out of the water.
评论 #27695179 未加载
ncphil将近 4 年前
The terms for both patent and copyright are way too long. Instead of increasing over the years as we advanced further into the Digital Age, they should have been gradually decreasing. The whole point of the US patent and copyright system is not to &quot;give a special reward&quot; to inventors and authors, but to &quot;promote the Progress of Science and the useful Arts&quot; by building up the public domain commons of inventions and art. The current law, with its lenghthy terms and slipshod processes that invite abuse by trolls and encourage the smothering oppression by monopoly power, is at odds with those original goals. It needs to be torn down completely and rebuilt for this not-so-new century.
评论 #27700511 未加载
postmeta将近 4 年前
Do the people that work there know it&#x27;s a troll company I wonder? Maybe like working at an MLM company, you&#x27;ve rationalized it in your own mind as protecting inventors and creative people! <a href="https:&#x2F;&#x2F;www.linkedin.com&#x2F;company&#x2F;ip-edge-llc" rel="nofollow">https:&#x2F;&#x2F;www.linkedin.com&#x2F;company&#x2F;ip-edge-llc</a>
评论 #27694974 未加载
ajnin将近 4 年前
Don&#x27;t patents examiners receive bonuses based on the number of patent applications they process? And isn&#x27;t accepting a patent basically checking a box, while rejecting one requires an argumented memoir and a lengthy back and forth discussion with the applicant and an appeal process? The patent system is skewed by design towards accepting patents too easily. I don&#x27;t believe that patents are entirely harmful, but I think they actually hinder progress in 95% of the cases. The system works like this by design, and a lot of interests rely on it. It will be very hard to fix.
评论 #27699507 未加载
评论 #27698638 未加载
AlbertCory将近 4 年前
The patent is indeed expired (there&#x27;s a website that will tell you that date, but I forget its URL), but they can still sue for infringement that took place before the expiration date.<p>If the total damages are small (as it appears), then you are probably costing them more than the suit is worth. So I&#x27;d expect them to offer to settle for a token amount, to avoid having an official loss on the record. But IANAL and this is not legal advice.
评论 #27696559 未加载
FpUser将近 4 年前
I do not want to go in the details of particular case. My opinion is that in general the paten system had outlived its purpose since some time and is now nothing but a detriment to society. It only benefits big corps at the expense of everybody else.<p>I can say that it is not possible for a regular person to do anything without breaking some patent. Mostly obvious and not deserving being granted in a first place. The only reason said person is not always sued is because of being too small of a fish. The minute a couple of bucks is made all of the vultures come down.
kevin_thibedeau将近 4 年前
I was under the impression that modern patents had to have a clear list of claims but this one has nothing but technobabble and &quot;1 Claim, 64 Drawing Sheets&quot;. How was this approved?
评论 #27695043 未加载
评论 #27695121 未加载
评论 #27695019 未加载
评论 #27695106 未加载
DarkmSparks将近 4 年前
Austin Meyer of XPlane fame did a good documentary on this a few years back.<p><a href="https:&#x2F;&#x2F;www.google.com&#x2F;amp&#x2F;s&#x2F;www.forbes.com&#x2F;sites&#x2F;johngreathouse&#x2F;2019&#x2F;11&#x2F;08&#x2F;this-entrepreneur-turned-a-nasty-patent-fight-into-a-compelling-film&#x2F;amp&#x2F;" rel="nofollow">https:&#x2F;&#x2F;www.google.com&#x2F;amp&#x2F;s&#x2F;www.forbes.com&#x2F;sites&#x2F;johngreath...</a><p>Basically Laminar got sued for having an application on the apple mobile store.<p>You Americans are crazy.
legel将近 4 年前
I have personally enjoyed community SparkFun events in Boulder, Colorado, which were hosted at a level that was both cool and fun for kids, adults, and experts alike. A really creative and intelligent company. Thanks Nathan for the courage to publish this, and helping other entrepreneurs inventing the future, to not be afraid to call fraud (and prior art) exactly what it is.
nkssy将近 4 年前
It might help if the patent dispute involved or required actual instantiations of the content of the patent.<p>It should be a valid argument that if no device exists for a category of patents then the patent is unenforceable. There must be a prototype. Set a threshold. After filing, there has to be some kind of prototype to demonstrate as an example. Otherwise what was the point of the patent anyway? Especially if its only good for 20 years. It should manifest within 5 years in the least otherwise where was the benefit to society of granting an effective monopoly?
ladyattis将近 4 年前
These kinds of patents really prove that we need to boost the budget for the USPTO since they&#x27;re always having things like this slip through despite the prior art out there. It&#x27;s much like the FTC in this situation where Congress knows that if the FTC is well funded then nonsense that businesses do won&#x27;t get by so easily so they starve the beast just enough to keep it ineffective. I swear it&#x27;s the same kind of setup for the patent office.
silexia将近 4 年前
Patent trolls are pure evil.<p>The solution is to dismantle the system scummy patent attorneys built and use to steal from good companies.<p>The entire patent system only serves to prevent competition through a government enforced monopoly. Why is there only one big aircraft manufacturer (Boeing) left in the US? Patents. Why is there only one major DB maker (Oracle)? Patents. Get rid of patents and you will greatly increased innovation and competition and the consumer will win.
评论 #27702004 未加载
评论 #27700974 未加载
civilized将近 4 年前
Why not just abolish patents? Who or what would oppose this?
评论 #27696886 未加载
AHappyCamper将近 4 年前
Reading that article makes me not want to open a business in the USA, and I really want to open a business in the USA...
av3csr将近 4 年前
I think MAUs are multiply-accumulate units.
评论 #27695122 未加载
评论 #27695284 未加载
_pmf_将近 4 年前
&gt; So imagine you’re a small business<p>Is SparkFun considered to be a small business?
keithnz将近 4 年前
I wonder whether I could get GPT3 to write patents for me so I can be lazy patent troll? hmmm
P0l83q4p1Hw3Ul将近 4 年前
Why not just kill patent trolls? That should solve this problem real quick.
评论 #27695535 未加载
FridayoLeary将近 4 年前
&gt; Patent trolls are Americans that are destroying American manufacturing and small business<p>Why should the identity of their victims matter? Also, are they?
评论 #27694809 未加载