Legal attack dog Rambus has suffered another blow to its memory technology licensing business after a US patent office appeals board declared three Barth patents invalid. Reuters reports that the three Barth patents refer to PC memory controllers and have been used by Rambus to collect millions of dollars in licensing fees from …
One Dreds to think
What Rambus thinks is patentable in LED lighting that no 10 year old electronics hobbiest didn't think of 20 years ago.
And you think that little problem would deter Rambus how exactly?
LED PWM colour mixing is patented...
That's right - using PWM to do LED colour mixing is patented in the US. A company called Color Kinetics beat Rambus to the punch there. Philips bought out Color Kinetics a while back, so they're now the alpha patent troll in this area.
...can companies ask for their royalty payments back?
Re: royalty payments
I hope not. That would create an awful precedent whereby any revenue obtained by licensing patents came with the risk that you might be asked to hand it all back at any future time.
It wouldn't be safe to spend it.
So you might as well not have it.
So the patent would be worthless.
refund, what refund?
"...can companies ask for their royalty payments back?"
It'll depend on the terms of the licensing ageements because Rambus and its licensees. It's typical for there to be no provision permitting a refund.
"So the patent would be worthless."
And the problem with that, certainly with regards to business method and software patents, would be what, precisely?
Perhaps it *should* be the case. I for one am getting mighty sick of suspect patents which should never have been granted in the first place being successfully used to create an income because it's cheaper to pay the licensing fee than to fight it in court and risk having a judge not realize how crappy the patent is.
Notice how the patent trolls typically go after the little guys first, when possible. They're much more likely to roll over; better to pay the troll than risk being beaten to death and thrown off the bridge. Once enough of them have settled and pay licensing it gives some sense of validity to the patent for when they face potential licensees who *can* afford to get into a legal battle. "Why yes our patent is good yer 'onor, lookit all these companies which agreed to pay us for its use". It becomes much harder to prove the invalidity of the bad patent.
That IMO is an even worse precedent.
In a system where a bad patent being struck down leaves the patent holder on the hook for refunding license fees, patent trolls and abusers would have a hell of a time continuing their practices knowing that any one of their shonky patents could spell disaster.
Of course the real solution to this would be to make sure that only patent-worthy ideas actually get patents granted. Then it would become reasonable to assume good intent by the patent holder, as opposed to how things are now where there is clearly a great deal of malicious intent involved in patenting.
noop ... a patent is validate until it is invalidated or expires, so all payments made between the moment a patent was issued and that is was invalidated are completely "legal" or "justified" ...
However, there is no clear jurisprudence on if payments that were due during the patent's validity but have not been paid at the moment a patent was invalidated (and not expired) have to be paid. I'm sure somebody will have an answer to that one ...
Re: cheaper to pay the licensing fee
"Perhaps it *should* be the case. I for one am getting mighty sick of suspect patents which should never have been granted in the first place being successfully used to create an income because it's cheaper to pay the licensing fee than to fight it in court and risk having a judge not realize how crappy the patent is."
Well if you are considering what *should* be the case, why not just say it *shouldn't* be cheaper to pay the licensing fee than fight it in court. Maybe it's the legal system that needs the attention more than the patent system. Right now, abusing the legal process seems to be fairly low risk with potentially big payouts if you get away with it. Perhaps politicians and judges should wake up to this fact, and the former empower the latter to do something about it.
The alternative, in the long term, is that society gets fed up with the fact that neither politicians nor judges seems to be serving them (society) like they are supposed to. History teaches us that *when* society finally loses patience with its politicians and judges, it ain't pretty to look at. So those two groups have (in the long term) a fairly good incentive to fix the problem before it gets out of hand.
"So the patent would be worthless."
No news here, surely? Aren't they all?
Dunno. Since I don't associate Rambus with business method or software patents, I didn't consider it relevant.
Only took four years. That's an improvement!
Now can they please reach the same decision with those MS patents?
Why stop there? all patents need to be destroyed, the only people who would lose money would be lawyers and the patent office.
Actually, if all patents were destroyed, the people that would loose the most money are the people and companies who INVENTED things. While often misused, the misused patents form only a very small percentage of the total patent group, much less than 1% of the total - and those are primarily "software patents" (which, In my opinion, as a class probably should be eliminated, as they aren't products or true business processes).
Patents do have a valid purpose - unless you feel that you should be able to use someone else's idea and work without compensation, and that you should be able to take a product that someone else spent thousands of hours and millions of dollars to develop, duplicate it, and sell it as your own. If that is the case, you have other problems that are unrelated to patents.
Physical design patents have a legitimate use as they encourage invention. That being said, the terms involved with patents and copyrights need to be seriously re-examined in light of today's age of information and rapid development capabilities.
Not really. I have many patents to my name and let me tell you that (typically) patents only benefit companies to whom the inventor transfers the ownership at the time of filing. The effect is that companies do not have to care about the inventive employees after the filing is done. The money thus saved is spent on lawyers instead of engineers :)
All I can say to RAMBUS about leaving the memory arena is "Don't let the door hit your ass on the way out!" best thing that could happen to the memory industry is for RAMBUS to do one.
Prior art on LED color mixing
I remember an article in a 1970's issue of Model Railroader magazine about using a combined red/green LED to model a 3 color searchlight signal. The red and green aspects were naturally generated by lighting the respective elements of the LED, the yellow aspect was simulated by rapidly alternating between red and green. It would seem obvious to anyone skilled in the art that PWM could be used to control the level of mixing.
That is why red/green LEDs are (or were) frequently called tri-color LEDs.
Since when has prior art had anything to do with it? Any consideration of prior art would be an offense under the Attorneys Full Employment Act.
Actually, my favorite prior-art patents are US6280318, granted in 2001, which patents using a fan to cool electronic equipment, and US7499276, granted in 2009 to a competitor of the first company, which claims the exact same thing except that the fan is specified to extract hot air rather than blow in cool air. These stand out to me because in 1998 I wrote a paper lamenting the patent mania and predicting that one day someone would try this, though I never expected it would be granted.
Re: Prior Art
In fairness, I think we can reasonably assume that the second company only did it to take the piss.
Taking the piss
No, that would be the water-cooling patent with the pump downstream of the heat source.