Slashdot Mirror
Apple Ordered To Pay $506 Million In Damages For Processor Patent Infringement (hothardware.com)
MojoKid writes from a report via Hot Hardware: Apple has been ordered to feed a recognized patent troll hundreds of millions of dollars for infringing on a patent that has to do with technology built into its A-series mobile processors. Initially Apple was on the hook for $234 million, owed to the University of Wisconsin-Madison Alumni Research Foundation (WARF) after it won a patent dispute against the Cupertino tech giant. However, a judge this week more than doubled the fine by tacking on an additional $272 million. U.S District Judge William Conley in Madison ruled that Apple owed additional damages plus interest because it continued to infringe on the patent all the way up until it expired in 2016. WARF is reportedly a non-practicing entity that exists only currently by defending its patents in litigation. The lawsuit filed in 2014 involves U.S. Patent No. 5,871,752, which describes the use of a predictor circuit that can help processors run more efficiently. WARF claimed the technology was used in Apple's A7, A8, and A8X processors that power the iPhone 5s, iPhone 6, iPhone 6 Plus, and various iterations of the iPad. Apple is not commenting on the matter, though it's being reported that Apple plans to fight and appeal the ruling.
...and they still didn't see it coming!
Having said that, is this WARF really a patent troll? My (foreigner's) understanding is that in the US, universities often expect some returns on their research in form of patent royalties.
Ezekiel 23:20
Way back in the 90s my little company was targeted by a patent troll. We were small potatoes but the co-defendents included Intel, IBM, and Digital. It seems some law firm was convinced that Ethernet infringed on some arbitration mechanism ARCnet used and they got the patent for it (from Datapoint I recall) so it was off to the horizon. They saw dollar signs.
Nothing ever came of it of course but it was still annoying to say the least to get served papers while we ere still trying to get one of our first products established.
I wish we could come up with a way to slap down patent trolls without making it harder for legitimate patent holders to defend their IP. I can't think of any.
Ordered to pay the University of Wisconsin. By a judge who: Bachelor's degree from University of Wisconsin - Check Juris Doctorate from University of Wisconsin - Check This judge should have recluse himself from the case.
Patent trolls take questionable patents and go after companies without the resources to put up a good defense, hoping they'll decide it's cheaper to settle. They want to avoid litigation at all costs, because a loss threatens their business model.
A non-practicing entity _could_ be a patent troll, but not here. Apple has vast resources to defend itself, did so in court, and lost because the court determined that they were infringing a valid patent.
"National Security is the chief cause of national insecurity." - Celine's First Law
Automative companies still receive new patents for what are largely tweaks or just different ways of doing the same thing and that's on technology that's been around for over one hundred years at this point. There's typically more than one way to accomplish something and Apple was free to use methods for which the patents had expired or to develop their own approach. If they think the patent should be invalid, they had ample chance to demonstrate that during the trial.
WARF is reportedly a non-practicing entity that exists only currently by defending its patents in litigation.
My first question would be does anyone else license this patent from them?
Actually it looks like they do. http://www.businessinsider.com... (the single page link does not seem to work)
I'm not sure troll applies to this company anymore than it does to Apple itself.
I did like one of the comments on the linked article from 2014 though;
"So apple get sued for patent infringement is patent trolling but apple suing for rectangles with rounded corners or unlocking by sliding your finger (on a touchscreen!) Is fine...."
If they are interested in making money on patents, then they should either make a product using those patents, or sell them to someone who can use them.
It represents a university. Universities don't make products, they conduct research. Some of that research has practical usage, some of it doesn't (that's part of the idea of a university: they can conduct basic research, not just product development). The technology that is potentially useful can be licensed to fund more basic research, which is the idea behind WARF. It's a non-profit that handles the licensing and feeds the money back into funding research. WARF may not make a product, but they certainly don't follow classical patent-troll patterns (not least of which is, you know, being a non-profit, which kinda undermines the entire motivation behind patent trolling).
A branch predictor circuit? Those have been around for decades.
Coincidentally, so has the patent: it was filed 1996. I don't know enough about the technology to say much, but it certainly doesn't sound like a troll patent.
"None can love freedom heartily, but good men; the rest love not freedom, but license." --John Milton
There's typically more than one way to accomplish something and Apple was free to use methods for which the patents had expired or to develop their own approach.
C'mon, we all know how this works. Apple probably did figure out a new way of doing it, and it happened to be the same method WARF has a patent for. Because, looking cursorily at WARF's patent, it's a tweak on a global branch prediction circuit.
As for the trial, yeah, try explaining to a judge how a particular tweak to a global branch predictor circuit is obvious. Or how you added further tweaks that make your design work differently enough to not encroach on a particular patent.
My Other Computer Is A Data General Nova III.
I personally studied with Prof. Sohi in the nineties when he and his students initially filed the patent, which they did essentially to get beer money. At Wisconsin the alumni foundation will give the inventors 20% of all royalties and WARF takes the other 80% for the risk involved in filing and prosecuting anyone who is dumb enough to mess with them. The inventors get to split $2000 that is given to them upfront. The first people to try and steal their idea was actually Intel, who actually sent researchers over from their Israeli research lab to sit in on the talks by Sohi and his students after they initially published their ideas in an academic paper. At the time they told Intel they would gladly license it to them for cheap. Intel told them to fuck off, telling them they had no IP since they published an academic paper on the subject. Intel actually based their Centrino line off this patented idea. Today Intel doesn't market the Centrino brand because around 2010 they settled out of court, two days before the trial was to begin, for an undisclosed sum. They did so because WARF and Wisconsin could easily prove that they had stolen the idea. The one caveat Wisconsin requested was that Intel not market the Centrino brand anymore. The same people who robbed Sohi at Intel were hired by Apple and now the same thing essentially happened, fat money being sent to Sohi and his crew but his time out in the open so dipshits who know nothing of the history can talk shit about one of the top academics in chip design in the US. It's not trolling it's a multi billion dollar company who can normally shit on the litttle guys getting their comeuopance.
Apple, IBM, and Motorola did some pretty significant branch prediction work on PowerPC, too, e.g. US5659752. I'm not saying that any of their patents are necessarily prior art, but some of them probably are. IBM, in particular, filed a *lot* of patents.
Check out my sci-fi/humor trilogy at PatriotsBooks.