Google Asks USPTO To Reexamine Four Oracle Patents 122
An anonymous reader writes "Google leaves no stone unturned in its defense against Oracle's patent and copyright infringement allegations. eWEEK reports on the latest development: Google has asked the USPTO to reexamine four of the seven patents asserted by Oracle. Patent watcher and skeptic Florian Mueller believes 'the world would be a better place without those virtual machine patents,' which he considers excessively broad and not really technical inventions. He also reports on a Google letter to the court, asking for permission to file a motion to throw out Oracle's copyright infringement allegations as soon as possible, without further discovery."
Obvious things (Score:1, Insightful)
like computer code, etc, shouldn't be able to be patented. Period! It would be better for Google to ask USPTO to re-evaluate more than just those 4 and throw them out and deny all the new coming.
Re: (Score:3, Informative)
Especially since the computer code has copyright protection anyway.
Re: (Score:2)
Especially since the computer code has copyright protection anyway.
Why, lacking patent protection would someone feel encouraged to give you the source to a valuable algorithm? The natural response from those losing wanted patent protection is going to be increased secrecy.
Sure, you can eventually reverse engineer most things, but I fail to see how that advances the state of the art more than full disclosure. I can see how the patent systems needs some reforms, but not protecting software at all is going to evolve valuable software into "hardware", appliances and such.
And
Re:Obvious things (Score:4, Insightful)
>valuable algorithm
Algorithms are unpatentable in the entire world.
>more than full disclosure.
Have you read some "computer-implemented method" loophole patents? They usually don't even include a reference implementation.
As for secrecy, there's no way for algorithms to be kept secret. If it's important (and most are not), someone will "reverse-engineer" (i.e. read) it, or just rediscover the same algorithm by himself. It's just mathematics.
And nowadays reverse compilers get you almost the original code back, automatically.
As for putting stuff into hardware, fine, vote with your wallet (also, it's still in there and can be read).
Patent protection for computer-implemented inventions are worse than useless. There's copyright already and the pace of the computer industry is so fast you don't need patents (which is why we don't have them, I guess). If your competitors want to copy something, let them. In the time they take to copy it you are at version 93953539 and they are at version 0.1. How does this even become an issue?
On the other hand, some weirdos who think they invented fourier transform (only a few hundred years late) would be making signal processing a minefield.
And don't forget that pure software shops are the minority of all programming jobs, why have special protection for 0.1% of the programmers (and of these, they would mostly block each other with trivialities), blocking the other 99.9% from doing their work (since everything would be a patent minefield)?
I really wonder where this damn elitism comes from, it's not that there's a lot of missing stuff in computer science anyhow. We've figured out how to program computers, now we program computers.
Re: (Score:1)
Algorithms are unpatentable. All the current patents are about the application of a particular algorithm to a particular application.
A singular value disposition is just math. You may no patent a natural law., but a singular value disposition using variable names related to dating is patent #6,734,568.
It's pretty inane, and this usually any software patent is.
Now compared to rockets. The idea of thrust is not patentable, just as the idea of a singular value disposition is not patentable.
Now a parti
Re: (Score:2)
Algorithms are unpatentable. All the current patents are about the application of a particular algorithm to a particular application.
That's a meaningless statement, in that the same applies to a mechanism in the physical world. It has to do something useful. So for all practical purposes, algorithms (like RSA, for instance) are patentable.
Re: (Score:3, Interesting)
I agree with you.
People getting software patents should publish their source code, and not get copyright on it. After 15 years it's public domain. A very good point.
Re: (Score:1)
Also copyright. Google shouldn't be allowed to hoard all its search code, since it's the commercialisation of an open academic effort and its systems build on so many millennia of scientific progress.
Re: (Score:2)
Google should certainly be able to not release the code and keep it a secret, but Google shouldn't be able to take legal action if someone reverse-engineers code, decompiles the code, if Google releases it themselves, or if someone else releases the code (Google sho
Re: (Score:2)
No employee at "bring information to everyone" Google has decided that freedom of information is more important than the risk of being sued. IOW, none has simply released Google source, knowing full well that no amount of legal force would reverse the effect on the marketplace. This says so much about the homogeneity of the working environment at Google.
Re:Obvious things (Score:5, Insightful)
Re: (Score:2, Insightful)
The world is full of martyrs. I fear most the institutions which lack them.
Re: (Score:3)
Explain how releasing Google's source code benefits the world as a whole. It doesn't. Sure, it might be -neat- and sure, Microsoft, Yahoo, and the rest of them want the information but it doesn't benefit the world as a whole. This isn't some lifesaving cancer drug, this isn't stopping a war, this isn't deposing of a dictator. This is a search algorithm. The most it would do if released would benefi
Re: (Score:3)
Except it's search code would be worthless if publically available. It would be trivial to game it.
Re: (Score:2)
Nope, SEO outfits are doing a fine job of that already - mostly because PageRank is (imho) a fucking awful and easily exploitable way of judging usefulness.
Re: (Score:2)
Re: (Score:2, Insightful)
This is always dumb. A computer implementation of existing inventions should never be patentable, for example most of the code that goes into graphing calculators is just mathematics. However, a new encoding algorithm is a novel invention. Specific new methods in MP3 and AAC were new inventions when created: new methods of doing things. A patent is a set of instructions: make this bore, bend this sheet, combine these materials, bolt here, and you have an engine. If you designed a novel mechanical proc
Re: (Score:2)
No it does not indicate that it was new and non-obvious. Only that it was new. Maybe lots of people had come up with the idea and just not had time to set pen to paper. The fact that they are waiting until the patent expires only means they do not want to be sued.
Re:Obvious things (Score:5, Insightful)
Software or an algorithm shouldn't be patentable because they aren't inventions, they are ideas or mathematics. Inventions are patentable, ideas are not.
Re: (Score:2, Insightful)
Of course, doing something USEFUL with the mathematics is what is patented, not the math itself. You may as well say that all mechanical or electronic devices are 'just physics', drugs are 'just chemistry', etc.
Re: (Score:2)
That's exactly right. The idea of a laser is not patentable. An actual design of a laser is.
It all comes down to whether or not you think an algorithm is a device. Reasonable people can disagree about this.
Re: (Score:3)
Algorithms patents usually are written as device patents, usually starting with "A System and Method to ...". The disctinction is a bit arbitrary to begin with.
Is "a computer programmed to do X" a device? If I have a cool new algorithm for the shift plan for an automatic transmission, is a non-computerized implementation fundamentally different than a computerized one?
Re: (Score:2)
I don't see the fundamental difference between the arrangement of cogs, springs and rods (say), and the arrangement of conditionals and loops.
Re:Obvious things (Score:4, Insightful)
I don't see the fundamental difference between the arrangement of cogs, springs and rods (say), and the arrangement of conditionals and loops.
Artificial scarcity anyone? One is copyable/usable by 6,900,000,000+ [census.gov] people at little or no cost. The other isn't.
The onus is on any patent proponent to show why billions of people should be blocked from using an idea, their free speech, to give a small number of people additional profit, for every single area of technology. Not for anybody else to prove that patents are not applicable.
Patents are a damaging, unnatural monopoly and by default should not exist unless there is compelling, scientific evidence, not anecdotes, that patents actually help rather than hinder in every single area of technology where they may be applied.
Particularly since there are vast swathes of ideas that have no patent protection and those areas of technology work just fine. Everything from cooking recipes to math to locating businesses to business organization to artistic ideas to political ideas to ...
The whole patent edifice is based on handwaving about what it means to say two ideas are the same or different (fundamental to deciding whether something is original or not). They can't even cope with the difference between inventing new words and inventing new ideas. The PTO's depth of thinking is atrocious.
---
"It is difficult to get a man to understand something when his job depends on not understanding it." - Upton Sinclair
Re: (Score:2)
Artificial scarcity anyone? One is copyable/usable by 6,900,000,000+ [census.gov] people at little or no cost. The other isn't.
Um, what? Both are ideas. Idea easily copyable. That's why we have patents in the first place.
In any case, you seem to be anti-patent in general. I don't see much point in debating that angle - there's too little common ground. I was responding to GP, who was contrasting algorithm (software) patents against device patents and trying to demonstrate that there is some fundamental difference that makes algorithms unpatentable.
Re: (Score:2)
The scarcity is only artificial if you think the original magically fell out of the sky.
Little or no marginal cost.
You talk, in pseudo-economic jargon, about the cost of copying as if it was the act of creation. It isn't. Symphonies, stories and software don't write themselves. If someone didn't put in the initial labour, there'd be nothing to copy.
Only somebody who's never created anything could possibly tal
Re: (Score:2)
I think the difference between a device and an idea is the specificity of it. You can't (or shouldn't be able to) patent the idea of a device that catches mice. However, you should be able to patent a particular arrangement of springs and levers that constitute your particular device that catches mice.
For computer code, you don't patent your source or object code as it's covered by copyright. Trying the patent the idea that is manifested in that code is akin to patenting the idea of a mouse trap.
Even if you
Re: (Score:2)
You can't (or shouldn't be able to) patent the idea of a device that catches mice. However, you should be able to patent a particular arrangement of springs and levers that constitute your particular device that catches mice.
Patents are by definition about protecting ideas. Consequently, you're not required to be overly particular - for example, you don't have to restrict the patent to a very precise placement of components, or specify exactly what material they're made of. Just so long as your description is for a clearly separate and distinct idea. For example, Watt had effectively patented the use of a separate condenser in his steam engine.
Patenting very specific things is essentially useless, since then anyone could replac
Re: (Score:2)
This [ipwatchdog.com] patent attorney disagrees with you.
Re: (Score:2)
Frankly, this is just legalese. An invention is an idea, fundamentally. Yes, you need an actual implementation to patent it rather than a "mere idea", but it's the formalized idea that is the core of any invention. Simply put, an invention is an idea that is implemented and works. And the novelty requirement (at least when properly applied) essentially requires it to be a new idea.
Re: (Score:3)
No, no, no. An idea on its own is worth no more than a wish or a prayer. "Wouldn't it be great if..." requires no effort and deserves no reward.
A horseless carriage is an idea. You can't patent that.
But an Otto cycle engine is an implementation of an idea, and that is patentable.
Re: (Score:1)
Re: (Score:2)
You might be technically correct, but practically you are not.
Try writing and distributing a video compressor without getting sued by MPEG-LA. If I can write software that infringes on a patent, then software is effectively patentable.
Re:Obvious things (Score:4, Interesting)
Much of the patented work on MP3 and related psycho-acoustic modeling technology were well predated by work on signal processing for cochlear implants. If the USPTO had vaguely qualified examiners with enough time ti actually examine the patents, they would have realized that.
Re: (Score:2)
Much of the patented work on MP3 and related psycho-acoustic modeling technology were well predated by work on signal processing for cochlear implants.
As I understand it, it is possible to patent a really new application of known principles. For example, if you invent a new machine that includes a rotating axle, it is not a problem that axles are a known technology.
Some of the principles behind MP3 were known since the 1940s and even earlier. But nobody had ever applied them to the problem of reducing the
Re: (Score:3)
I'm not claiming that there was no innovation at all in MP3, it was some nice work but that it was hardly so non-obvious and novel that it deserved a 20 year legal monopoly plus the thicket of follow on patents extending the monopoly. The very same techniques were already in use for years at the time in an even more extreme form.
The objective of the processor in the cochlear implant was to take the entire human auditory range in and drop it down to a signal in 8 to 16 frequency bands (later 32) such that it
Re: (Score:2)
There was NO question that lossy compression based on a transform to the frequency domain and dropping information that a listener couldn't perceive would work.
But isn't MP3, essentially, a specific method of determining "information that a listener couldn't perceive"? I'm not familiar with details, so excuse my ignorance, but doesn't Vorbis also does what you describe, without touching MP3 patents? If so, the latter would seem to be pretty specific, as a good patent should be.
Re: (Score:3)
Actually, it's a huge mess [wikipedia.org]. Many entities have made many claims on pretty much anything that conforms to the standard file structure and plays audio. Every possible abuse of patents has been tried in some form or another in the area of MP3.
Re: (Score:2)
Re: (Score:2)
Re:Obvious things (Score:5, Insightful)
i suspect they chose the 4 weakest patents that they know are bunk and have the highest chance of being overturned.
this will set a very good example for all those other pesky patents they're being sued over.
TFA misses a very important point of strategy - Google aren't filing patents like mad because that's not a game they are playing. they are massive innovators, but their business model is not technology so much as advertising. the patents aren't as valuable to them as they are to Oracle, and if Google lose 576 patents it'll be nothing to the many thousands that other companies stand to lose if they are ever challenged.
Google can afford many lawyers, and these 4 patents might just be the tip of the wedge that they plan to drive into ALL software patents.
here's hoping. the MPEG-LA are already shitting bricks at Google's VP8 e-penis.
Re: (Score:2)
TFA misses a very important point of strategy - Google aren't filing patents like mad because that's not a game they are playing. they are massive innovators, but their business model is not technology so much as advertising. the patents aren't as valuable to them as they are to Oracle, and if Google lose 576 patents it'll be nothing to the many thousands that other companies stand to lose if they are ever challenged.
You might want to check the date of Google's patents. More than half of them are less than a year old. Google is filing patents like mad, to try to catch up with its competitors that already have an armoury of them.
Patently Mad Re:Obvious things (Score:1)
Is 250 patents per year really filing like mad? IBM received over 5000 patents in 2010 [ibm.com]. It took them more than 50 years to get their first 5000...
So if they got 5000, how many did they file (actually the way the USPTO works it was probably close to the same). Still, I suspect that the GP was closer to right than the parent. They both raise valid points. Certainly Google is smart to get some patents as armour. Still, I suspect their primary tactic is going to remain simply out innovating the competition so f
Re: (Score:3)
Anyone working in software R&D for a big company knows that such patent filling numbers are absurd. Sure great research is done at IBM, but 5000 patents... This becomes just another metric for management. Employees have objectives and rewards to fill maximum number of patent idea.
Off course quality of these patents is really low. I have seen patent on stuff everybody already do for years.
I've even seen someone patenting a design failure. We had to cope with his faulty software, because to save time for
Re: (Score:1)
Re: (Score:3)
I think skippy has it right [skippyslist.com]:
Patent Reform:
Patents Are For Inventors Act
You can only file a patent on a physical object. You must have a working prototype, or a set of schematics that can be followed to build a working prototype. Specifically illegal to attempt to patent anything that was not created yourself, or by an authorized representative of you. Attempting to patent something that you did not invent will now be a felony.
No more patenting software, game mechanics, or ideas on how to accomplish things.
Patent Trolling will result in Federal Prison.
Re: (Score:2)
Patenting software is like patenting math.
Re: (Score:1)
Quoting Florian (Score:2)
Re: (Score:2)
Not going to happen. (Score:2)
Re: (Score:2, Insightful)
It's lobbying game now, actually. The one who pulls the right strings wins.
Re:Not going to happen. (Score:4, Interesting)
Why, yes, there has been changes to patent laws since then.
Several court rules, including a couple from SCOTUS have changed the rules. You might think these don't apply retroactively, but a review is not strictly a retroactive action. They would have to apply the current standards for any review.
Things like Prior Art, Obviousness, overly broad patents have all had reviews in the courts, as any google search will reveal.
Patents are challenged almost weekly, and a significant number of these are invalidated.
Re: (Score:2)
Re: (Score:2)
I'm pretty sure if they granted them the first time around, they're not changing their mind. It's not like there's been some kind of change to the patent laws between now and then, and changing the decision would really amount to admitting they screwed up the first time.
Don't bet on that - the USPTO does grant reexaminations, but regularly changes or invalidates patents based on reexaminations.
In the case of 3rd party requests for reexamination such as this, if the reexam is granted, there's a very good chance the patent will either be invalidated, or some of the claims will be removed or changed (about 71% according to this (somewhat old) report [sonnabendlaw.com]).
Re: (Score:1)
I would say that Zork has them beat by quite a few years on VM design. Now I agree that you could say that it wasn't a general VM but it obviously is prior art.
http://en.wikipedia.org/wiki/Z-machine [wikipedia.org]
Re: (Score:2)
Eh? The courts most certainly have invalidated patents before...
A lot of granted patents probably wouldn't stand up to the higher level of scrutiny given by a court.
Re: (Score:2)
The key is that a re-exam is granted when the requestor presents evidence (i.e., prior art) indicating a "substantial new question of patentability". Usually, this means prior art that the PTO didn't have ready access to, such as something not published online.
Patent examination is not a perfect process. It's essentially an attempt to prove a negative (i.e., the prior art in question doesn't exist). The examiner can search for an increasingly long period of time, but the return on that increased time dim
That's not cricket grommit (Score:4, Funny)
Re: (Score:1)
Doesn't work with GPL code.
Unless they only allow distribution in areas that the patents are invalid (there is an exception in the GPL allowing that restriction to be placed on redistribution).
This could be interesting... (Score:2, Interesting)
...because if any of the patents are found to be unenforceable, because of either prior art or obviousness, then Oracle's suit falls apart. Take a look at RE38,104 [google.com]. This appears to be prior art.
Interesting times lie ahead.
Disclaimer:
I am rooting for Google.
Re: (Score:2)
I looked at it and it appears to be owned by Oracle. Was there something else we were supposed to look at?
Re: (Score:2)
Have a look over here (Score:2)
Sorry for the late response but a read over here [theserverside.com] can tell you a lot about what I was talking about.
Re: (Score:1)
...because if any of the patents are found to be unenforceable, because of either prior art or obviousness, then Oracle's suit falls apart.
It's important to understand that in reevaluation, Oracle's patents would not be considered unenforceable, but simply not patentable as being anticipated or obvious. In reexamination, there is no patent
Re: (Score:1)
Re:This could be interesting... (Score:5, Informative)
How did this get modded "Insightful?"
Here's a quote from the original lawsuit filing from August last year:
(emphasis added)
In other words, this "prior art" you posted is one of the patents that Oracle is suing Google over, that Oracle acquired when they bought Sun Microsystems, Inc.
Re: (Score:2)
Um, so? Who owns the patent that's prior art doesn't exclude it from being prior art, does it?
Of course, it can't be considered prior art for itself, but it can very well invalidate the other patents by the same patent holder.
If I patent "generic drinking vessel made from glass", I can't very well patent "generic vessel for drinking made from crystallized silicates" a couple of years later, because my own prior art would exempt it. Else one could abuse the system to prolong a patent ad infinitum.
Re: (Score:1)
A better source... (Score:3, Informative)
Some choice parts excluded from the OP's articles:
> materials identified by Oracle as infringing in fact created by a third party and released into the public domain
> the only two files allegedly containing "copied" code were created by Google and provided to Oracle for use in open source releases of Java
Hypocrisy (Score:2)
I am one to agree that software patents should not exist (and non software ones should be only for exceptional inventions, not for tiny progress in a field) but I hate hypocrisy even more.
Google, go for revision requests the day you stop requesting patents left and right, and volunteer your own collection to be invalidated.
It won't be until some of the big patent holders themselves complain to the government about the horrible use of patents that things may change.
As long as they try to play the patent gam
Re: (Score:2)
Re: (Score:2)
Re: (Score:1)
Re: (Score:2)
Re: (Score:1)
Re: (Score:2)
Patents simply monopolize progress, and the worst part is very often is a monopoly only used to lock a door, not even to actually implement said progress.
What are you relying on when you say this?
What you yourself said (bolded)
Copyrights and patents function very differently. One difference is the so-called independent creation, where if two people write the same code, neither having seen the others, they both have rights to use that code and obtain copyright in it. This protection is very different from the monopoly granted by a patent , which prevents all second-comers, regardless whether they knew of the patent, from practicing the invention. So when comparing copyright and patent law, one does not supplant the other, and frequently they address different issues. Boiling that down, patents protect the idea/function achieved by the code (performed by a machine, of course), while copyright protects the expression of that idea/function (the individual lines of code).
If you refer to the "door lock" part, I refer to the flood of patent trolls that do just that.
That paragraph also happens to very clearly state some of the reasons I dislike patents (for one it is becoming more expensive to research if whatever you are doing is already patented than it is to develop anything in the software world.) I should not be able to redistribute some one else's code, I should not be able to clone some one else's software to the T, but I should have the
Re: (Score:1)
but I should have the freedom to come up with things on my own that happen to work the same way some other guy somewhere thought off first.
I know this goes against everything Slashdot discussion board stands for, I disagree with your position, but understand and respect it.
My only question is where does this line of reasoning end? Does it only apply to software? If so, why? Does it have to do with the generally lower cost of development when compared to other industries? Or is it something else? I've never been quite sure why. Another reason I have heard is that an invention patented in a "software patent" is usually obvious. If so, sho
Re: (Score:1)
Re: (Score:2)
Ummm, it's the fault of the freeloader benefitting for free from the R&D that the inventor paid for.
Well said. We all know that when a new invention comes o
Re: (Score:2)
Re: (Score:1)
Intellectual theft does not exist, so we can ignore that part
To me, that suggests you do not believe there is any such thing as intellectual property, since the wrongful taking of property is usually regarded as theft. With how easily you summarily dismiss the notion of intellectual theft, I don't think I will bother to point out the folly in your position.
and the point of the patent system is not to reward innovation, but to encourage it
While I recognize and appreciate the distinction you have pointed out, I think that the method in which innovation is encouraged in the U.S. and the rest of the world could fairly be characterized as a reward. Pa
Re: (Score:2)
I acknowledge that patents and copyrights exist, but even if one compares copyright and patents to physical property, what happens is infringement, not theft. It's actually much more like trespassing (use without
Re: (Score:1)
Re: (Score:2)
Oh yeah? Well how about I patent your FACE with my FIST?
Re: (Score:2)
Unfortunately, the patent system is being abused and the USPTO appears ill-equipped to evaluate the patent worthiness of software.
If word processing was patented then Wang would have prevented most of home and small business computing in the 80's. It could've happened.
I'm not commenting on the validity of Oracle's patents that they purchased from Sun.
Re: (Score:1)
Re: (Score:2)
Patent-trolling (Score:3, Insightful)
Re: (Score:2)
Nah, not patent trolling. Oracle cares about control, and they'll get control any way they can. If they can leverage patents to get control they will.
Summary Time (Score:5, Informative)
Article summary from this and Groklaw:
Court: You're slamming us with paperwork so if you want to load our asses up with any more you have to summarize it first and give the other guy a chance to write a short summary of why you're full of it.
Google: Summary time then. You can't patent byte code and virtual machines. We're asking the US Patent office to take a second look at these dumb patents, and we're also asking that we put the whole trial on hold till we get an answer.
Court: You're still on the hook because Oracle says you have bunch of copied source files in Android. That's copyrights not patents.
Google: All but 12 of those source files are interfaces and we've got 3 previous cases that show that you can copy interfaces because they're interfaces and don't contain any actual code.
Google: Also two of the 12 files that they're suing us for were written by Google employees and given to Sun under an open source license.
Google: The remaining 10 files that we're totally not admitting that we straight stole from Java comprises less than one percent of the Android code base so we can get away with it because of "de minimus", which basically means we legally get to copy your crap so long as we don't do it too much.
Google: So basically Oracle copyright claim is full of crap and we should dismiss this whole case till the USPTO gets a chance to look at the patents.
Oracle: We've got two days to come up with reasons that you're full of crap, so expect a formal response soon.
Re: (Score:2)
As I understand it, the 10 files aren't even part of Android - just a test suite. Sure, they should've been more careful, but this is perhaps the least significant part of the code base - a small and unnecessary (could be replaced with custom code) part of a *test suite*.
And as soon as it was brought to their attention, they deleted the files. Of course, it's source control, but the files are already available as part of the Java testing kit - their source is available, but it's not open-source.
I think the
Re: (Score:2)
Anyone serious about this topic should most of the above. Mr. Mueller is knowledgeable but his opinions must be taken with a large grain of salt.