Patents and (possible) Plagiarism, Anyone ever been in a similar situation?

Could Lucent come after you for using your own circuit? Of course the answer is "yes." The more significant question is whether or not Lucent would have a significant chance of winning a patent infringement suit.

If you can demonstrate that your circuit was published or sold more than 1 year before the priority date of Lucent's patent application, then it would be difficult for Lucent to prevail in court.

Generally it is difficult to estimate the value of a patent. In some cases patent rights are sold, so for those patents there is an established value. In other situations an infringement case is won by the patent owner, or a settlement is agreed to; again some dollar value is determined.

But in many situations a company has a portfolio of patents that keep competitors at bay. Consider the situation where your company develops and patents technology "A", and improvements A1, A2, etc. If these patent allow you to achieve a monopoly they can be quite valuable.

But perhaps I am you competitor and to avoid your patents and to stay in the marketI develop and patent B, B1, B2, etc.

Within a few years we are developing parallel, but distinct technologies. There may be some advantages to the A series, and there may be different advantages to the B series, but it is hard to put a dollar value on each of our technologies. Our patents are important to our businesses, the help us maintain a douopoly, and they may prevent the other party from implementing certain improvements. But how much are they worth?

Richard Tanzer

John -

You mention the patent describes the Median Filter, but does it actually claim it in the claims section? A patent can describe anything, but the last section of the patent (the claims) is where the patent actually claims what the invention is. If it isn't in the claims, it is not being patented.

John Providenza

Since there are 54 patents with "Inventor Name" including "Jiang" and "Assignee Name" including "Lucent" I'm guessing this guy gets paid based on how many patents he churns out.

I would contact the Lucent General Counsel (the lawyers) and *ask* them if they would like to know of a patent based on plagiarized ideas. Most of the claims are for "The method" which you appear to have developed. You can supply the general counsel with supporting documentation.

It's sad to see plagiarism. It's ridiculous to have someone make a living off it.

It's not always entirely the "inventor's" fault. About 30 years ago I was playing around with a Shannon-Rack for audio d/a conversion.

My then boss was insistent that I should patent it. The fact that it was called the Shannon rack didn't seem to be enough of a clue that it had been invented by Shannon.

Things got quite heated for a few days before he relented.

I would be annoyed as well, but I expect I would let it go. Thanks for the posting. I have often found it curious that there are so many posters with very specific questions but no fundamental knowledge in the area. Now we know one reason why. Online patent phishing sweatshops.

It might be worth an appointment with an attorney to determine your best move, but I think the risk is low. Large corporations collect patents to defend the cash-cows of the moment from other corporations.

I think patents are a type of insurance policy for large corporations. Very few lawyers or engineers have become rich or famous because of a patent. A patent citation on a resume doesn't look bad, but a list of publications looks even better to me.

-- Mike Treseler

Mike, I agree wholeheartedly. Generally speaking, a patent is only as good as your resources available to defend it. For an individual, the cost sunk into obtaining a patent will likely never be recovered unless it has bulletproof claims, is important enough to be essential and interesting to a company that might buy it, and you've got enough financial wherewithall to defend your patent if someone does infringe it.

I also agree, publications look better on a Resume than patents. Both are attractive, but the publications carry more weight with me.

yeh, seems to me its is usually somthing like; BigCompanyA has alot of (more or less) valid patents so that when BigCompanyB claim they infringe on one of their patents, they can counter sue claiming they infringe on one of their patents, eventually they reach a settlement and agree to go after everyone else instead.

-Lasse

Having had a little more time, I've read through the patent in excruciating detail now, and it turns out that despite being identical in the first stages and the last stage, Mr. Jiang (umm, Dr. Jiang?) does something very slightly different in the middle. Instead of using

4 comparisons to sort out guaranteed non-median values, he uses a control vector based on ths second stage sort of the first stage medians. This control vector presumably configures 3:1 muxes (he glosses over that, using only the word "locate" in his diagram), and allows using only 2 comparisons to my 4. So he's outdone my circuit! (at least in terms of total comparisons; the two 3:1 muxes actually use more FPGA LUT fabric than the two comparisons, because the carry-logic is free in this context). I humbly apologise to Hong Jiang for any untoward remarks.

Regards all, Just John

No. You can destroy their patent by showing where you published it earlier. Of course, they COULD, actually sue you for using their (invalid) patent, but your defense should be so easy, it would be a bad idea for them to do so.

Some patents are worth billions of $. See some of the patent wrangling and licensing between Intel and some of their competitors. I think TI has had some big patent cases in the past couple years, too. There was a big Xilinx patent that was finally decided this year, IIRC.

The kind of thing you describe is actually quite common, as in many cases the earlier work may not be clearly published. If you have an XCell applications note available on the web, or on print form, that sounds like sufficiently clear publication to me.

If somebody with deep pockets doesn't come along to infringe your patent or offer to license it, then you will never know whether your patent is valuable or not! If a billion $ corporation infringes your patent, and you are successful in suing them for that, then it could be worth a LOT of dough. But, the guys who invented the intermitent windshield wiper and the capillary blood sample tube both went to their graves after 20+ years of litlgation, only to win the cases AFTER they died!

I don't recommend any individual waste their time patenting anything unless it is a blockbuster, and they have the pockets to defend the thing against all comers.

Jon

There are too many could and should's here, when dealing with lawyers :)

A good recent (and still alive, AFAIK) example of the first class sillyness of (too many) patents, and the decisions to litigate, is Microchip (who should know better) suing Zilog (who are quite small) for making Microcontrollers with fewer IO pins, than internal BUS width. A slew of questions arise out of this, none of which has anything to do with innovation.

What next - cars with fewer doors than cylinders patented ?

A sea of prior art exists, going way, way back to the Intel 4004, plus patents are supposed to not be obvious to onw skilled in the art.

Maybe tiny Zilog really does have Microchip scared - but I would have thought Microchip could safely ignore Zilog, and find many better areas for their energies and funds ?

Examples like these serve to confirm that patents are simply feeding troughs for lawyers.

-jg

Hi John, After I read your paper in Xilinx Tip Design, I searched the USPTO with your name and found no result. It really surprised me.

Your result was worth filing a patent absolutely!

Try to get your company boss's support and you have the right to invalidate the Lucent patent with rock solid prove.

Good luck.

Weng

Hi John, Jiang's provisional patent was filed on Mar. 27, 2000, more than 4 years earlier than the patent approval date: July 6, 2004.

What is the published date of your paper?

That is the most important factor deciding who is the first person to get the circuit. It is not only a circuit, but a fundamental method covering all similar circuits.

Weng

Hi Richard, I am asking a question: "If you can demonstrate that your circuit was published or sold more than 1 year before the priority date of Lucent's patent application, then it would be difficult for Lucent to prevail in court. "

Why one year before the priority date of Lucent's patent application?

In this case, John published his paper 5 years ago, Jiang's priority date is January 16, 2001. It seems that the patent was filed to be within one year of John's paper. Doesn't it mean their paper and patent application has no relationship? Could you please give more explanation?

Weng

JustJohn schrieb:

IBM estimates that patents do them 10x as much harm as they earn from their patent portfolio.

This means that the main use for patents is to defend yourself against patent trolls.

Kolja Sulimma

Zilog is actually bigger than a lot of people realize. They are not terribly big in microprocessors, (they once were) but they sell a LOT of network interface chips and such stuff.

Yeah, I used to know what the legal fees racked up in these cases were, and they were ASTRONOMICAL! A complete travesty. There are THOUSANDS of wildly invalid patents on all sorts of things, including perpetual motion machines cleverly disguised under different names.

Jon

In U.S. patent law there are at least two sections that are applicable in this situation:

35 USC 102: A person shall be entitled to a patent unless - (a) the invention was known or used by others in this country, or patented or described in a printed publication in this or a foreign country, before the invention thereof by the applicant for patent, or (b) the invention was patented or described in a printed publication in this or a foreign country or in public use or on sale in this country, more than one year prior to the date of the application or patent in the United States, or ...

To invalidate a patent a patent under 102(a) requires comparing the date that the applicant [Lucent] invented the circuit to the date that John's circuit became publicly known. In this case we don't know when Lucent invented the circuit unless the Lucent chooses to divulge the information or they are forced to do so in court. So the patent may be invalid under 102(a), but we have no way of knowing at this point.

To invalidate a patent under 102(b) we would need to show that John's circuit became publicly known more than a year before Lucent's application. Since we know the date of Lucent's application, that is an easier task.

In either 102(a) or 102(b) the date that John's invention became publicly known is important, not the date he invented the circuit.

- Rich

Hi Rich, Thank you for your response.

Can you please give me a date example to explain the 102(b). I really have difficulty understanding the item.

After I joined this discussion, I re-read the patent 6,760,737 again and found the following several new points that conflict with my previous views posted in this discussion.

1. The patent 6,760,737 has not repeated what John had written in his paper. In other words, my point of view is the patent 6,760,737 has nothing to do with John's paper.

1. It deals with how to find a median data among N data; where N is any odd number greater than 1. Its implementation and its claims are described as more a software operations than a hardware operations.

For example, N data is divided into 3 groups and each group should be sorted before doing any further operations.

1. For N = 9 situation, its implementation is slightly different from what John had done. John did it in the minimum comparison operation, but patent 6,760,737 does it like a software operation that is less optimized than John method, but it can be best done by a software.

Any further comments are welcome.

Weng

The Jiang patent (6,760,737) is based on application 09/760,923, filed on 16 January 2001. Application 09/760,923 claims priority to provisional application 60/192,294, filed 27 March 2000. Any claims in

6,760,737 that are supported by the provisional application may take advantage of the 27 March 2000 priority date.

35 USC 102(b) states that "A person shall be entitled to a patent UNLESS the invention was patented or described in a printed publication in this or a foreign country or in public use or on sale in this country, more than one year prior to the date of the application for patent in the United States..."

Now - inserting the specifics of the Jiang patent, 102(b) is essentially:

"Jiang is entitled to a patent UNLESS the invention was publicly disclosed prior to 27 March 1999."

I'm not going to try to compare John's disclosure to the claims in

6,760,737. My point is simply that if John published his paper on or after 27 March 1999, it is probably not relevant to the validity of 6,760,737.

Rich

Hi Rich, What you said is causing my confusion the most way: " if John published his paper on or after 27 March 1999, it is probably not relevant to the validity of

6,760,737. "

If you posted a paper on website today, tomorrow I used your paper contents to file a patent. Then 102(b) says that your paper is probably not relevant to the validity of mine???

Weng