Slashdot Mirror
Chrome's Sandbox Feature Infringes On Three Patents So Google Must Now Pay $20 Million (bleepingcomputer.com)
An anonymous reader writes: After five years of litigation at various levels of the U.S. legal system, today, following the conclusion of a jury trial, Google was ordered to pay $20 million to two developers after a jury ruled that Google had infringed on three patents when it designed Chrome's sandboxing feature. Litigation had been going on since 2012, with Google winning the original verdict, but then losing the appeal. After the Supreme Court refused to listen to Google's petition, they sent the case back for a retrial in the U.S. District Court in Eastern Texas, the home of all patent trolls. As expected, Google lost the case and must now pay $20 million in damages, in the form of rolling royalties, which means the company stands to pay more money as Chrome becomes more popular in the future.
Troll's Get Them!
This is another shining example of why software patents need to be abolished.
I'm now anxiously awaiting expert legal opinions by people who didn't even read the article, much less the patent.
"This is a shining example", you said. Which of the three patents is a shining example and why? What issue do you see in whichever patent you're talking about?
Patents in general. Medical too. Yes. Really.
I couldn't find the patent numbers being spoken of here, at least in the first hundred pages of Googles 350+ page brief, and the article itself is pretty useless when it comes to details.
But to answer your question in general, software patents break the entire purpose and intent of the patent system as a whole. That's what makes nearly all of them worthless and impossible.
Patents are intended to describe an implementation of something, previously an implementation of a machine or process.
If one chooses not to design their own machine, they can look for a patent describing a machine that does what they want and license it. At that point you are allowed to build the machine as described in detail in the patent and typically sell it.
That's the entire purpose of licensing a patent in the first place, to save you the time of designing something to perform that function when that work has already been done by someone else.
Software patents however have no such requirement, and thus almost never actually describe any form of working machine or process or anything.
If I want a machine to package my widgets automatically, I could find a patent on a machine to package my particular widget, license it, and use the description of that machine to build a widget packager. If the cost of the patent license is cheaper than doing my own R&D, it's still a win.
But if I want a program to customize my widgets, despite being patents that describe "a process to customize a widget", there is generally no description of any form of software that would do that.
So no matter how much cheaper it would be to license said patent than do my own R&D and programming, actually licensing the patent does not benefit me in any way shape or form since it does not provide any form of software or a design of software that would accomplish that.
Instead the trolls get a patent on the concept and idea of customizing widgets, and then use that to sue me when I do my own R&D and programming work to write a widget customizing program all on my own without their assistance.
That aids no one but the patent trolls, and that aid comes to them for exactly zero effort or work that benefits literally no one.
That is why software patents are wrong and should not exist.
Don't try to put words in other people's mouths. StormReaver said nothing about any specific one of the patents being a shining example, and was clearly referirng to this incident rather than to any particular patent.
they know that when they go off the jury pool, they will get to go to the nice local stores and skating rinks sponsored by patent troll companies.
Well, you don't need to wholly ban software patents.
Just make it a requirement for them to be valid, that the patent application contain enough detail to fully build the machine being described, from scratch, without infringing other patents or doing significant work other than what's described specifically in the patent application.
Make the wording of the law such that no software patent can really be patentable without working code, and you're done. Bye bye trolls.
im curious why they owe $20 million on a product which is given away for free? and what were the damages of infringing on something that isn't used by the person owning the patent?
It's not a typo if you understood the meaning!
Don't forget to pay your $9B licensing fines, you cock-smoking teabaggers!
Maybe the billion dollar corporation should have worked something out before violating the patents. It would have probably saved them a lot of grief.
“Common sense is not so common.” — Voltaire
Are you talking about the list that the Obama administration created or the deportation policies that Trump also carried over from Obama?
A Pirate and a Puritan look the same on a balance sheet.
These are developers who never gained profit on their own so they look to a big company that can provide them with that. Of course you can find lawyers willing to file suit for their own paychecks. This is why the courts are filled with this kind of frivolous patent or ideal lawsuits. Especially with software its a murky argument that you copied someone else.
All I need to hear, Trolls Win
substantially, as a patentable idea that is, than java applet sandboxes of 1995 vintage?
This kind of litigation would seem to rely on the profound, deep, deep technical ignorance of most in the legal profession.
That's one reason this kind of patent trolling is so despicable.
Where are we going and why are we in a handbasket?
You would think that all the major tech companies would take a small percentage of the billions$$ they make every year to do some serious lobbying for sane patent laws. But that never happens, no matter how many times these companies get hit by patent trolls.
Instead, patents, especially software patents, are treated like nuclear weapons. Publicly, everyone says they are bad and should be eliminated, but privately, nobody really wants to get rid of them because there's always a chance that some day you might want to use a patent to nuke a competitor. Nobody will ever admit it, of course, but it's the truth.
Since a few million dollars here and there is nothing more than pocket change to companies like Microsoft and Google, and just another "cost of doing business" nothing ever changes. The patent trolls get rich and are encouraged to file more lawsuits and companies who don't have deep pockets can be driven out of business by bogus patents that never should have been allowed in the first place.
From the article:
At the heart of both decisions was the phrase "web browser process," which Google said it was too generic to describe the anti-malware features it included in Chrome.
Perhaps you can find the three patent numbers and post them, but if trolls have now moved from "on a computer" to "web browser process" then Google should prevail. If the patents don't actually include code, then Google should prevail.
Our patent system is broken. It is obvious that software patents are stifling innovation.
I think he's referring to the 3M+ people Obama deported over 8 years. But Trump's focus is mainly criminal elements, with the occasional "innocent" illegal caught in the net, whereas Obama was focused on deporting families in order to not upset the crime-loving sanctuary cities.
So how do they count the damage and the profit from a free product? The services are the products, after all. The article speaks of web browser process as if the naming of the process would trigger a patent protection. Better again to rename the e10s content process to a plugin-container like it used to be, Mozilla.
If not then the patents should be voided. Otherwise the patent system gets trolled into garbage disrepute.
My vote would be we should amend software licenses such as he MIT, Apache and GNU to have a blacklist provision for software patent trolls.
Not being able to use most of the internet would get the point across.
Agreed.
Software is a written work much like a book. The author can assert copyright protection and control its distribution. Patent protection for software is akin to patenting mathematics or logic (which are not patentable).
It doesn't need to be any more complicated than this. Software copyrights, at this point, destroy more value than they create.
That is not the intent of patent protection.
Reissue serial numbers: USRE43500; USRE43528; and USRE43529. Searchable on google.
Uncle Tom = ((corporate slave) || (engineer)) && (hate patents)
This really is an example of why the patent system in the US is completely broken.
The patent authors amended the patents years after Google Chrome implemented sandboxing in order to specifically make the patents apply more specifically to Chrome.
Lets pay the lawyers instead!
You are being ripped off every second of every day, so that advertisers can help rip you off even more tomorrow.
This is already a requirement for a valid patent. Search "enabling disclosure" sometime rather than continue to float your theories that the system accounted for literally centuries ago.
Amen.
https://xkcd.com/435/
You have the right to remain sentient. If you give up the right to remain sentient, you will be elected to public office
Which of the three patents is a shining example and why?
ALL of them.
1) They are all overly broad.
2) They all describe a mathematical process which is not patentable.
3) They are all obvious.
I got called away before I was finished, so I posted prematurely.
4) They all have invalidating prior art, which the patent office and the patent court uniformly ignore.
Do you create software? Have you read a software patent?
Anyone who has done both will know that software patents, as the grandparent post said, are completely worthless to someone creating a working implementation of software which would be covered under the patent. Software patents describe a concept. They do not provide the details of a working implementation (because there is no working implementation) that would allow one to re-create the software.
Software parents are akin to copyrighting e.g. stories using certain plot devices. Imagine if one author patented revealing the killer at the start, another revealing him at the end, another patenting revealing the doer-of-crime two thirds of the way in. Certainly abolishing software patents would not inhibit progress in software development, and without them more progress would be made.
John_Chalisque
https://www.sandboxie.com/
sounds like its about the same thing
The people who actually read the patents, multiple juries, disagree with your claims of fact, 1, 3, and 4. If you'd care to point to even one *possible* instance of prior art, an interesting discussion might be possible. If you just keep saying random stuff, with absolutely no idea what the patents even cover - well that's just boring. We'd be watching Trump if we wanted to hear someone say random stuff from their ass. (Though he at least gets a five minute briefing first, most of the time.)
On 2 (math), you've been trolled. The law is:
--
The laws of nature, including the laws of physics, chemistry, and mathematics, aren't patentable [because they aren't invented].
--
You can't patent gravity, you can patent a new elevator design.
You can't patent momentum, you can patent a newly invented braking system.
You can't patent multiplication, you can patent a time machine which happens to use gears (multiplication), levers (more multiplication), etc.
You can't patent addition, you can patent a newly invented system for detecting spam, which happens to use addition and other things.
What's not patentable, and I'm quoting the law, are "the laws of nature".
I've read a lot of patents, and written a lot of software (I've spent far more time programming though).
I've read some *extremely* specific "software patents"*, and some very general ones. The company I work for has a patent so specific that it would be hard to infringe it. If you tried, you'dv probably accidentally do something slightly differently, so your implementation wouldn't be covered by our patent.
There are about 40 MILLION commercial airplane flights each year. Of those 40 million, about 2 have fatal crashes. So 39,999,998 safe flights, 2 crashes. We all know which flights end up all over the news.
Patents are similar. Bad ones end up on the news, often being the subject of extensive litigation. If someone didn't know anything about commercial aviation, they'd never been on a plane or at at airport, just watching the news they might reasonably be very afraid of flying. Every flight they've ever heard about crashed. If you know something about aviation, you know that's not even a million-to-one chance, it's a twenty-million-to-one chance. If you've never had reason to read patents, and never been in any kind of court trial, you might reasonably think most of them are like the ones you hear about in the news.
There ARE some patents that are overly broad, which sucks. There are some plane flights that crash, which also sucks.
Actually even the patents you hear about in the news are frequently *not* actually like what you hear in the news, especially on Slashdot - patent law is a big clickbait item on Slashdot. Most of us know that half the tech-related headlines and summaries we see on Slashdot are basically BS. The same is true of patents.
Some company will apply for a patent related to new type of fire extinguisher for cars that uses a new compound they've developed, which requires a new kind of valve they've invented to handle it. They've *applied for* a patent to cover "toroidal double monkey valve inverted for use in automotive fire extinguishers using dry dimethyl carbonate extinguishing agent". The Slashdot headline will most assuredly be "Company Patents Fire Extinguishers". The summary will mention fire extinguishers in cars, but won't mention the newly invented valve which is the thing they are actually trying to patent. Slashdot commenters go wild posting about "prior art - Nascar". You can't really blame them, the summary didn't even mention the patent is for a special valve the company invented, to use with a brand new type of extingushing agent. Commenters react to the headline, not having any way to know it's BS unless they take time to research it.
* There's no such thing as a "software patent", but that's a topic for another day.
Ps, my apologies to the chemists here, who know that dimethyl carbonate would *not* actually be a very good extinguishing agent. I wasn't going to *actually* invent a brand-new extinguishing agent just to make my post, so I named a random chemical that most people aren't familar with.
I've read through a number of recent patents in a variety of areas and the problem wasn't that they used terms that were too technical, but words that were too vague for anything whatsoever to be constructed from them. They shouldn't have been granted, but the legal requirements for a valid patent isn't the problem, but that law means nothing these days... and not just patent law.
No true Scotsman's what I hear whenever I read "There's no such thing as a...
There's no such thing as a leprechaun.
There's no such thing as a unicorn.
There's no such thing as a gaseous solid.
There's no such thing as a dollarback.
There's no such thing as a purple patent.
There's no such thing as a software patent.
You could also make some headway by enforcing the novelty requirement for patents. If a software troll can make piles of money suing 32 companies who accidentally independently created software that infringes the patent... I think you'd have to question the novelty. If enough other people have done it by themselves without your help or even a decent description of how you did it (as most software patents are written) then I don't think it would meet any sane person's idea of a novel idea.
How do you get away with this? She HATES it when I get any kitty litter in there.
Sleep your way to a whiter smile...date a dentist!
First, there ARE software patents. They describe ideas that can be implemented on a general purpose computer without any specialized hardware.
Second, pretty much NONE of software patents are useful. Most of them are filed for defensive purposes or to show "value" for investors and are never asserted or licensed. During the recent years 20 years or so the cost of patent litigation exceeded the licensing revenue for software patents ( https://arstechnica.com/inform... ).
Third, there ARE useful patents. Pretty much none of them are purely software ones. A novel agent for extinguishers would be a great invention. However, in software patent speech it would be described as: "An agent consisting of chemical matter that utilizes heat-sensitive decomposition reactions to rapidly terminate the free-radical heat-driven chains reactions", - without any actual chemical structure specified in the patent.
Softwares is algorithms. Point in case.
Alice.
Chew on this:
https://www.techdirt.com/artic...
On the scale of sandbox quality, Chrome should dump their model and adopt the SSH techniques - the rendering engine should be chroot() to /var/empty. That improves the software and kills the patent violation in one stroke.
http://undeadly.org/cgi?action...
"First of all, on the positive side, privileges separation, chrooting and the message passing design have proven fairly efficient at protecting us from a complete disaster. [The] Worst attacks resulted in [the] unprivileged process being compromised, the privileged process remained untouched, so did the queue process which runs as a separate user too, preventing data loss... This is good news, we're not perfect and bugs will creep in, but we know that these lines of defense work, and they do reduce considerably how we will suffer from a bug, turning a bug into a nuisance rather than a full catastrophe. No root were harmed during this audit as far as we know."
The fundamentals behind sandboxing and hardware virtualisation date back from the time of big iron mainframes.
The only novel thing is that this is applied to a personal computer/workstation (to a newer type of hardware that didn't exist back then, but is basically just smaller).
"Sufficiently advanced satire is indistinguishable from reality." - [Tips: 1DrYakQDKCQ6y52z6QbnkxHXAocMZJE61o ]
java applet sandboxing is (in theory) granted by the virtual environment provided by the JVM.
In theory, all the applet could be running in the same process.
chromium's sandboxing is hardware segragation provided by the CPU hardware itself (memory protection, and similar bread and butter of multi-processoring)
from the CPU 's perpesctive, each tab sandbox is an entirely different process.
and if you check the links of patent mentioned here around on this /. thread, this is exactly what is covered : using hardware multi-processing to isolate tasks.
so the prior art isn't Java in 1995, but much older big iron mainframes of past era.
BUT, I suppose that, because the patent says "...but on a home computer/workstation" (in claim 1) suddenly all the prior art on mainframes and minicomps doesn't count.
(Kind of like all the "...but on the internet !" business patents).
Case in point that the whole idea of software patent is completely b0rked.
(Happy to live in a european country !)
"Sufficiently advanced satire is indistinguishable from reality." - [Tips: 1DrYakQDKCQ6y52z6QbnkxHXAocMZJE61o ]
Nike Free Run 17 Etats membres ont atteint l'objectif exprimé en énergie finale et 19 l'objectif exprimé en énergie primaire, explique la Commission. La France ne figure dans aucune des deux listes du rapport dédié à l'efficacité énergétique. S'agissant des secteurs, l'industrie affiche la plus forte baisse de consommation d'énergie finale en dix ans (-16%), suivie de près par le résidentiel (-15%). Toutefois, certains pays ont vu la consommation énergétique industrielle repartir fortement à la hausse entre 2014 et 2015, sous l'effet de la reprise économique et un regain d'activité de l'industrie cimentière. C'est notamment le cas de Chypre (+19%). La Commission précise que le taux de rénovation énergétique des bâtiments est de 1% par an, soulignant que l'Allemagne et la France affichent des taux légèrement supérieurs, de l'ordre de 1,75% et 1,5%, respectivement.
Here's what the court ruled in Alice:
--
mere recitation of a generic computer cannot transform a *patent-ineligible* idea into a patent-eligible invention
--
In other words, if you have something that's not patentable in the first place, adding the words "with a computer" doesn't magically make it patentable. It's either patentable or not, saying "with a computer" doesn't change that, the court ruled.
By the same reasoning, if you start with something that *is* patentable, having a computer involved doesn't make it magically unpatentable.
One can make a machine out of transistors which compresses sound in a new and useful way, and if it's new and useful it's patentable. We'll call this machine "Machine A". If you invented Machine A and I sold copies without your permission, I would be in violation of your patent. That's true even if the copies I'm selling are "some assemby required". If assembling it according to the instructions I provide results in violating your patent, you could sue me.
An electronic machine like Machine A can be built by hand, or with tools. I can instruct another machine called a "pick and place" machine to build "Machine A" by picking up transistors and putting them on a circuit board. Either way I end up with Machine A, whether I build it by hand or use a pick and place machine to build it. If I'm building your machine without a license, I violate your patent whether I build it by hand or I use a pick-and-place to build it. I end up with your patented Machine A.
I can also build Machine A by starting with a huge grid of transistors and REMOVING the ones I don't want. I'll still end up with the same Machine A. That's called a PGA - it's building electronic devices by carving, destroying the parts I don't want. That's another method of building Machine A and violating your patent. To build Machine A from a PGA, I type up a precise specification of what Machine A to be, then the PGA machine builds me a copy of machine A by disconnecting all the transistors that aren't needed for Machine A. An ASIC is similar - it builds electronic machines from a set of instructions.
Here's the interesting bit - to specify what ASIC or PGA I want built, I can use a language called C. Basically, I build Machine A in C, then my PGA "printer" makes a hardcopy, with transistors. That's the same C language that we use to build the Linux kernel, and most system software. The same machine, specified using the same language, can be "printed out" as a physical manifestation, or run internally to the computer. Either way, it's still Machine A, your invention. If your invention is patentable and patented, it's patented whether I build it by hand, with a pick-and-place, with an ASIC, or with a compiler.
From google's reply:
This case concerns three patents that seek to protect computers "from malicious software", or malware. According to the patent's common specification, malware programs often succeded in infecting computers because the computers' "resources" were "shared by programs simultaneously, giving a malware program a conduit to access and corrupt other programs.". [...] The proposed solution was to eliminate that conduit by separating the computer's memory into distinct areas, such that a malicious program could not access or infect other programs. [...]
Because that simple concept had long been known, respondents secured allowance of their claims only by limiting their scope. Among other things, the patent claims were limited to segregating components of a computer's hardware (as opposed to software). The computer's key system files would operate on a processor with access to one memory region, while any network-interface software would operate on a second processor with access only to a second memory region. Those claims issued as US patent No 7,484,247. [...]
After failing to license or sell that patented invention, [...] respondents surrendered the original patent in 2010 and filed reissue applications with different claims. [...]
Armed with publicly available information about Google's Chrome web browser [...], the applicants drafted claims geared toward browsers. Instead of separating hardware components, the proposed claims used discrete software processes. The computer would execute trusted processes in the main memory area, while "isolating" potentially dangerous processes "from the main computer system" in a second area. The proposed claims referred to the trusted process and the potentially dangerous process as the first and second "browser process[es]". [...]
The examiner rejected the claims because they were not new.
Then it continues that some earlier patent "Narin" already had such a concept, and apparently the patent got issued anyway because it was limited to "web browser process" and not "browsing program", as that earlier patent.
Whether to abolish software patents or not (I think they should), this is clearly an abuse of the system.
Its also shocking that simple obvious ideas like the concept to confine processes into a sandbox for security purposes can be patented when you just make your claims to a specific enough subset.
Thanks for the research. Much appreciated.
> From google's reply: ...
> Its also shocking
Don't be too shocked that according to Google's lawyers, Google is right. Maybe they are, maybe they aren't, but what's quoted is Google's argument. The other side may very well reply "Wtf? Did you forget to read the entire first half of the patent?" :)
Most of the time when a patent story shows up on Slashdot I read the patent to see what it actually claims. I haven't done the research on these. So I'm taking no position on who might be right or wrong. I'm only pointing that it's a bit silly to take a position on a patent without knowing what the patent says, or anything more than the very general field the patent is related to, but that's common around here.
I couldn't find the patent numbers being spoken of here, at least in the first hundred pages of Googles 350+ page brief, and the article itself is pretty useless when it comes to details.
But to answer your question in general, software patents break the entire purpose and intent of the patent system as a whole. That's what makes nearly all of them worthless and impossible.
Patents are intended to describe an implementation of something, previously an implementation of a machine or process.
If one chooses not to design their own machine, they can look for a patent describing a machine that does what they want and license it. At that point you are allowed to build the machine as described in detail in the patent and typically sell it.
That's the entire purpose of licensing a patent in the first place, to save you the time of designing something to perform that function when that work has already been done by someone else.
Software patents however have no such requirement, and thus almost never actually describe any form of working machine or process or anything.
If I want a machine to package my widgets automatically, I could find a patent on a machine to package my particular widget, license it, and use the description of that machine to build a widget packager. If the cost of the patent license is cheaper than doing my own R&D, it's still a win.
But if I want a program to customize my widgets, despite being patents that describe "a process to customize a widget", there is generally no description of any form of software that would do that.
So no matter how much cheaper it would be to license said patent than do my own R&D and programming, actually licensing the patent does not benefit me in any way shape or form since it does not provide any form of software or a design of software that would accomplish that.
Instead the trolls get a patent on the concept and idea of customizing widgets, and then use that to sue me when I do my own R&D and programming work to write a widget customizing program all on my own without their assistance.
That aids no one but the patent trolls, and that aid comes to them for exactly zero effort or work that benefits literally no one.
That is why software patents are wrong and should not exist.
Best to setup a software company in Nigeria, Libia, or anywhere where software patents are illegal. That is what is happening to the American Software Industry. (India, Malaysia and middle east, here we come).
Leslie Satenstein Montreal Quebec Canada
There ARE some patents that are overly broad,
Irrelevant. Many patent proponents like to claim a narrow patent somehow makes it okay. They're wrong,
All patents are obviously harmful due to their huge cost of implementation, whether the cost of running the patent office, the cost of litigation, the opportunity cost due to literally billions of people being blocked from using useful idea or even just the application fees.
The onus is on you to prove that the huge cost of patents is outweighed by their purported benefit. And the purported benefit is nowhere near as clear as proponents like to claim as many areas of business and public life work just fine without patents e.g. I have the idea of opening a business in a growing town. Nobody else has thought of doing this here. It's very specific. Why shouldn't I be able to patent that idea and stop any competition for a while? Yes, I'm well aware there are arbitrary patent office rules that mean that's not possible however that's the point - they're arbitrary. Unfortunately, software patents are just a particularly clear example of how broken patents are.
There are about 40 MILLION commercial airplane flights each year. Of those 40 million, about 2 have fatal crashes. So 39,999,998 safe flights, 2 crashes. We all know which flights end up all over the news.
Patents are similar. Bad ones end up on the news, often being the subject of extensive litigation. If someone didn't know anything about commercial aviation, they'd never been on a plane or at at airport, just watching the news they might reasonably be very afraid of flying. Every flight they've ever heard about crashed. If you know something about aviation, you know that's not even a million-to-one chance, it's a twenty-million-to-one chance. If you've never had reason to read patents, and never been in any kind of court trial, you might reasonably think most of them are like the ones you hear about in the news.
Myth. There are lots of good reasons to suspect the majority of software patents - the League for Programming Freedom has had a position paper out on this for decades. Further, there are massive legal ethics issues with the patent system - and some serious government ethics issues - both of which have been discussed at length on this forum.
The majority of those patents examined by the community have problems.
In short, the hypothesis that most patents are bad - especially software patents - has been repeatedly been confirmed.
We call this process of developing a hypothesis and then getting evidence science.
It beats bad analogies hands down, any day.
I've read a lot of patents, and written a lot of software (I've spent far more time programming though).
I call BS. No matter how much programming experience you claim to have, it isn't enough. You can't possibly know about everything that has been done and published.
And it's very likely - despite what you may think - that your experience isn't broad enough for you to really know what is "novel" and "not obvious" - but there are plenty of people with such experience. Read the discussions of patents here, and you'll find all kinds of prior art pointed out, things that most people wouldn't know about.
Your views seem to be pure hubris.
That's also the problem with the patent system: arrogance and hubris, combined with a lack of integrity, by the majority of people making something from the system - and not willing to examine it closely enough.
Interesting idea. Unfortunately, ideas cannot be patented. Which was more or less the whole basis for the opinion of that judge (that, in general, software patents are not valid patents).
Your argument would seem, then, merely to show that saying that an idea can be implemented in hardware also doesn't make it patentable.
I think what is confusing you here is that any specific implementation of your idea in hardware might be patentable (assuming your idea has been implemented in an original, innovative fashion) --- it is the implementation which is patentable, not the idea.