Amazon Owes $525 Million In Cloud-Storage Patent Fight, US Jury Says (reuters.com) 38
A federal jury in Illinois on Wednesday said Amazon Web Services owes tech company Kove $525 million for violating three patents relating to its data-storage technology. From the report: The jury determined (PDF) that AWS infringed three Kove patents covering technology that Kove said had become "essential" to the ability of Amazon's cloud-computing arm to "store and retrieve massive amounts of data." An Amazon spokesperson said the company disagrees with the verdict and intends to appeal. Kove's lead attorney Courtland Reichman called the verdict "a testament to the power of innovation and the importance of protecting IP (intellectual property) rights for start-up companies against tech giants." Kove also sued Google last year for infringing the same three patents in a separate Illinois lawsuit that is still ongoing.
I hate Amazon and software patents (Score:3)
I am not sure who to bag on here!?!?! Haha
Re:I hate Amazon and software patents (Score:5, Insightful)
You should put principles above your hate for the specific victim.
Amazon is in the right here. They didn't steal the ideas from the patent holder. Amazon independently came up with the same ideas, as did Google.
When at least three independent entities come up with the same ideas, they don't pass the "non-obviousness" test and shouldn't be patentable.
The patent holder is a parasite, contributing nothing of value. They patent an idea, sit on it until someone comes up with a similar idea, and then sic the lawyers on their prey. If the verdict is allowed to stand, the money will be spent on more lawyers and more lawsuits, stifling innovation and draining vitality out of our economy.
Re: (Score:2)
And maybe many recognize the need, so there are likely many parallel efforts to solve the need
And if several come up with the same idea, then really you're just trying to patent "recognizing the need." You're doing something new by just combining and adapting existing ideas.
Just look at all the dot com boom patents of "X, but on the Internet." If there is nothing new but the need part ("on the Internet") then you're just trying to patent recognizing the need.
Re: (Score:2)
Thanks, just came here to say this same thing. While I was at Amazon I saw the same thing happen with several other technologies (none of them worth half a billion, though). The company has hundreds if not thousands of teams of teams worldwide working to create new products, it would be amazing if other people somewhere weren't working on creating some of the same ideas. Somehow lawyers manage to convince prospective clients that the only way Amazon could possibly have the same idea is by stealing their
Re: (Score:2)
Re: (Score:2)
"sit on it until someone comes up with a similar idea"
Kove sells actual products. That does not prove that their specific patents would hold up to scrutiny, but it does undermine your claim that they merely sit on patents until they find something to litigate over.
Nuicance patents (Score:4, Insightful)
The article is light on details, and doing further research it *seems* to be , as best as I can tell that this has something to do with "software defined memory", ie defining how much memory a VM can alllocate. That seems to be the gist of Koves patents
I dont understand how this is parentable. "Let VMs have a specified amount of RAM" seems such an uninventable obvious thing. But if it that is what the patent in question is about, that seems to be something that threatens the whole of modern large scale computing.
I hope Amazon appeals this, not for amazons sake but for the whole damn industries sake, its potentially as big a threat as those stupid SCO lawsuits where, had they succeeded.
Re: (Score:2)
I am sure there is an expert out there who can explain why this isn't as simple as that (especially with 3 different patents involved)
Re: (Score:2)
I did some more googling and found the patent numbers. 7,233,978, 7,103,640 and 7,814,170 seem to be the 3 patents in question and they seem to have something to do with distributing a database across a bunch of machines on a network and how you find and retrieve data from that database which seems like something Amazon would be doing with their cloud operations.
Re: (Score:2)
Without reading the patent to minimize the risk of somehow getting any ideas from it and being tainted in the future... how is that not just sharding (a concept that dates back to at least Ultima Online in 1997, almost a decade before their company was founded)?
Re:Nuisance patents (Score:2)
You really should read the patents. That's the whole point of patents - in return for legally-enforceable royalties, the patented ideas become public knowledge.
And more to the point, everyone is "tainted" already, deemed to know what's in the patent, 'cos it's been published, and therefore liable to pay royalties if they use the idea.
Re: (Score:2)
And more to the point, everyone is "tainted" already, deemed to know what's in the patent, 'cos it's been published, and therefore liable to pay royalties if they use the idea.
Nope. At least in the United States, there's this little thing called treble damages for willful infringement. There are very legitimate reasons to avoid reading patents held by other companies until after they expire unless you're a lawyer trying to get those patents overturned.
Re: (Score:2)
I replied to GP post with my reading of the patents but I'm no expert. It's more like distributed non-hierarchical DNS combined with standard filesystem fragmentation management but where the next fragment might be a lot further jump away than another disk cluster. It sure doesn't look like there's anything unique - if someone asked me to invent S3 storage, I would probably come up with a worse version of exactly this. And then when it performed terribly, the first thing the next guy would do to optimize
Re: (Score:2)
All in all, it's of course a complex system. But it's basically built with standard computer "Legos."
7,233,978 reads like BGP (or distributed DNS, see below) but for filesystem metadata.
7,814,170 is just a duplicating how filesystem fragmentation works at the disk level but with a bigger scope. It avoids additional fragmentation by having the server send the data chunk to be stored to whichever server has the previous block so it can prevent fragmentation. It's not quite "x but on the Internet"
Re:Nuicance patents (Score:4, Interesting)
Skimming the page, it looks to be things like using a hash in a distributed database, and redirecting if a resource isn't found.
There could be some interesting details I didn't catch, but it seems like the kind of things a competent software engineer would come up with if tasked with scaling up some servers.
Re:Nuicance patents (Score:4, Insightful)
Not an expert, but there seem to be more details here [casetext.com]
Skimming the page, it looks to be things like using a hash in a distributed database, and redirecting if a resource isn't found.
There could be some interesting details I didn't catch, but it seems like the kind of things a competent software engineer would come up with if tasked with scaling up some servers.
I think this is the disconnect.
People think the patent system should be After many great labours I've made a fantastic new discovery and deserve some reward for my efforts!
Too often, in software, it ends up being I was here first! Someone throw up a few fences behind me!!
I don't think that's right, being first is it's own reward, if you want fences behind you then you better have cut through some serious barriers to get there.
Re:Nuicance patents (Score:5, Interesting)
Too often, in software, it ends up being I was here first! Someone throw up a few fences behind me!!
I think you're lumping together with two of the worst categories, which are "first to do it on a computer" or "not first to do, only first to be bold enough to claim doing it on a computer was patentable". There were far, FAR too many patents that were essentially something being done (or something that could have been done) in a pencil-and-paper business world, but because they either had a computer doing it or sprinkled in some internet mumbo jumbo / ecommerce they got a patent issued, or worse, past a court (east texas comes to mind). I was happy when the Supreme Court finally undercut that but it still seems to happen too often.
Once you eliminate those two abusive categories, there's nothing wrong with coming up with something non obvious and being the first to do it. I'm talking about things like RSA and Diffie-Hellman, as worthy examples. I note the following site: https://www.eff.org/issues/stu... [eff.org] A quick look tells me there seems to be patents on 'matchmaking' and 'clocking in when starting work'.
Re: (Score:2)
This isn't my area of expertise, but these patents all read like "x but on the Internet." They took existing ideas from other computing domains and directly plugged them into each other with minimal glue and called it a new invention. And Amazon and Google did the same but didn't think it worthy of patenting - and if anyone would patent something minimally patentable it would be either of them.
This is just a smaller company sad that they tried to be first to market with something that takes a lot of time
Re: (Score:2)
> Too often, in software, it ends up being I was here first! Someone throw up a few fences behind me!!
Why do you think I got Patent #1 for my time machine! Boy, was that a wild shootout at the patent office! EVERYBODY came!
I guess amazon has to delete all cloud data now (Score:3)
Not the way I thought The Great Oops would happen.
"and intends to appeal"... (Score:1)
Re: (Score:2)
Because that's how the legal system works.
Ask yourself why such things beyond the ken of mere mortals are patentable. How does this benefit society?
Re: (Score:2)
About the jury thing, I know that's how the legal system works, I
Re: (Score:2)
That's what patents were claimed to be supposed to be. What they have been if baffle-gab lawyer speak. I've looked at a few that should have been intelligible to me, and they weren't. I understood how to do what they claimed to be about, but I couldn't understand the patent.
Re: (Score:2)
Re: (Score:2)
I might be misreading, but I think they said the language of the patent was designed to obscure the simplicity of the "invention"
You don't need someone skilled in the art of the patentable thing as much as you need a patent lawyer who can read through all the fluff to find the actual meat of the invention.
Re: (Score:2)
patents (Score:1)
https://patents.google.com/pat... [google.com]
https://patents.google.com/pat... [google.com]
https://patents.google.com/pat... [google.com]
Re: (Score:2)
"Kove is run out of a loft filled with other businesses."
So?
Does valid work only occur in garages?
Totally not a patent troll (Score:1)
/s
The new frontier in patents (Score:2)
Re: (Score:2)
Next it will be "X with AI"
Links to patents in case (Score:3)
US 7,814,180 "Domain Name Service Server": https://ppubs.uspto.gov/dirsea... [uspto.gov]
US 7,233,978 "Method and apparatus for managing location information in a network separate from the data to which the location information pertains": https://ppubs.uspto.gov/dirsea... [uspto.gov]
US 7,103,640 "Network distributed tracking wire transfer protocol": https://ppubs.uspto.gov/dirsea... [uspto.gov]
Remember when discussing the validity of a patent that what the abstract/summary/description says is not relevant. It is the Claims you have to look at.
Re: (Score:2)
All three software patents should have never been granted, as they don't pass the test of being tied to a particular computer (at the very least).