What Happens in a Patent Lawsuit?

One of the presentations in the exhibit hall, at the Chiphead Theater, was What Happens in a Patent Lawsuit? which was presented by John Strawn and Tom Millikan. I have written about patents a couple of times on Breakfast Bytes, in Patents and Standards, Managing the Challenge and in Lawyers, Guns, and Money: An Evening with EDAC on Patents. This presentation was not about the process for getting a patent granted, or whether patents for business processes like Amazon's one-click should be granted. It was purely about the process of a patent lawsuit, and what to expect if you get involved in one.

The first presenter was John Strawn, who has been an expert witness many times. The second was Tom Millikan, who is a lawyer. But he is not the usual lawyer, who got a bachelor's degree and immediately went to law school. He worked at Texas Instruments as a chip designer for years, before deciding that being a lawyer was more interesting than designing chips (hard to imagine). So he is obviously pretty technical too.

They worked together ten years ago on what was, at the time, the largest patent case ever. The jury decided to award $1.5B to the opposing side in a case that involved ripping mp3s into Windows Media Player. But their perspective on the case prevailed on appeal so they eventually won. There have been higher awards since then, especially in the smartphone market.

If you look at a patent, there is a title page, description, claims, figures (usually). If you are an engineer, you typically read the description and the figures to find out what the patent is about. But if you are a lawyer, you jump straight to the claims, since that is what the patent actually covers. In granting the patent, there is usually a negotiation with the patent office (in my experience, the patent office starts by rejecting every patent when it is first filed) and so the claims in the final patent are often a lot less than you would expect from the description and diagrams. In particular, the list of claims is a list of the things that you have to do to infringe the patent. After all, a patent is a right to stop other people doing things (even if you don't do them yourself).

The claims are also the heart of the argument in a patent trial, either to show that you don't do those things (as a defendant), or that the other guy does (as a plaintiff). Alternatively, as a defendant, you might prove that other people (not necessarily you) did all that before, and so the patent was not novel and is therefore invalid.

Pre-Suit Filing

If you are a plaintiff, you cannot file suit without doing some investigation. You can end up paying the other sides attorneys' fees if you don't. You may need to buy a competitor's products and take them apart. You can't just send a letter saying that the other guy is probably infringing something so they should pay up.

For the defendant, most of the time you are doing nothing or waiting. A few companies monitor their competitors patent filings, but mostly they do not. The one exception to doing nothing is a notice letter. Having said you can't just send a letter saying the other guy is infringing, it happens all the time. If you get a letter like that, you do not ignore it and put it in a drawer, because if eventually there is a trial and you lose, you are then deemed to be a "wilful infringer" and subject to triple damages.

By the way, earlier in my career as a VP Engineering, I was given legal advice to make sure my engineers never looked at patents. That way they could never be "wilful infringers" since that required them to know that the patent existed, and what it contained.

So if you receive a notice letter, you should respond. If the letter is very specific, then you can respond by filing your own suit and asking the court to declare that you do not infringe the patent. Often the letter doesn't address specific claims, in which case the appropriate response is to say, in effect:

We received your letter. There is not enough information to determine if we infringe. If you think we really infringe, then send more information.

Of course, if you never hear from them again, you don't need to do more. But if it does get to a lawsuit, this response is enough that you are not a wilful infringers and so not subject to triple damages.

Things have changed a bit in this area in the last couple of years. When you file a complaint, you now have to identify the claims, and within a few months proof that the conditions are met. If you get dragged to court with no basis, then you can now (as of the last two years) ask the court to award your attorneys' fees.

There are a number of places that the suit can be filed. The three main areas are California (because one of those places is where your headquarters office is), around Delaware (because many companies are incorporated in Delaware), and East Texas (since they don't like big corporations so tend to be very friendly to individual inventors). If you are a patent holder, you want to be in Texas, but the rules have recently been changed and you have to have an actual facility in Texas to be sued there. Another reason that you might want to file a declaratory judgment that you are not infringing is to control where the suit is filed.

Once the complaint is filed, litigation begins. The plaintiff may work to prove infringement, plan a licensing campaign (get more defendants), demand money, exclude a competitor, defend the patent.

Meanwhile, the defendant works to prove no infringement, prove the patent invalid, attach the patent, diminish the value, or design around it.

Another area to be aware of are indemnity agreements. Generally lawsuits target the end product, because that's where the largest revenue is, and how you get the most money. It is a lot easier to get a $1 license for a $500 smartphone, than for the $10 chip inside the phone that contains the suspect technology. However, often the end-product manufacturer has indemnity agreements with suppliers. If you are a supplier, it is very important to manage this when you negotiate indemnity obligations, or you may find yourself, worst case, responsible for a $20 license fee for a smartphone based on your $10 chip.

Discovery

Discovery is where the two-tracks of a patent lawsuit get going in earnest. On the legal side, there is a requirement to provide documents, interrogatories, admissions (is it true your product has this feature?), depostions (all-day interviews), subpoenas to third parties. As a general rule, it is a waste of time deposing someone who doesn't want to be deposed (Tom had a story about deposing the retired CFO of Bell Labs, who needed to certify that he had signed certain documents but who answered every question with "It looks like my signature, but I don't remember this specific document".)

The entire case can hinge on this phase. John said that during he discovery, he had the experience of both finding the thing needed to win, and the thing that meant that they were sure to lose. This phase can be very expensive, too, since it can be a bottomless pit, looking at work that was done twenty years ago under different legal regimes. Discovery can take a lot of time and money.

Also, if you get deposed, you do not need to help them, you just need to answer the question. If they use the wrong word, don't correct them. Don't offer up information that you were not asked. A deposition is not a conversation. However, your attorney should train you how to do this well, if you get deposed.

Contentions

The plaintiff needs to demonstrate infringement using technical means (such as source code, datasheets, product analysis). They need to put down in detail what is in the patent and why the defendant infringes it. If the clain says something specific, such as a physical button, then identify the button.

Meanwhile, the defendant is looking for things that were developed before the patent was claimed (a patent has to be novel), deficiencies in the patent, and things that don't match. If the patent says a physical button, show that the product uses an on-screen menu item, not a physical button. John had a story about claim language that says "it repeats" and "iterative procedure." By putting a breakpoint in the code he could show it didn't repeat and so was not an "iterative procedure."

There is a presumption under patent law that the patent is valid. So to prove that it is invalid, you need a high degree of proof. However, to prove infringement, there is no presumption either way.

If you are trying to challenge a patents validity, then the cheapest way is to request an inter partes review from the USPTO. This has a fixed, relatively fast, timeline. The judges have engineering degrees. However, prior art only means publications on paper, you can't produce a physical product. About 80% of reviews find at least some claims invalid on review (usually since there is prior art). This review can happen at any time, even before a suit is filed, but it has to be brought within one year after being served. The reason for this is that the patent examiners do not always have the right background. Their industry experience may be in a different industry. 

Expert Reports

The plaintiff's and defendant's experts file two expert reports, the infringement report and the validity report. They get exchanged, usually on the same day. These are often 50-200 pages long, with attachments (like schematics or Verilog). The plaintiff's expert writes the opinion that the product infringes, and the defendant's expert is going to attack the patent or give the opinion that the product does not infringe. Then rebuttal reports are issued about a month later (e.g., the plaintiff's expert failed to show where xx was in our product).

Motions

At this point, before trial, there can be a summary judgment where the judge decides based on the factual record and the lack of dispute. For example, if the source code must iterate, and the defendant shows it does not, and the plaintiff does not dispute this, then the judge may dismiss on that basis alone. Getting a summary judgment can be challenging due to the (lack of) technical knowledge of the judge.

Trial

Witnesses describe the facts and tell the story. Jury decides what are the facts. The judge only defines the law (including the meaning of words).

Witnesses are cross-examined. If you are a witness, the opposing lawyers will try to get you to say something that doesn't match what you said in deposition. This is called "impeachment." The problem is that most witnesses are not used to being grilled. One example Tom gave was that in deposition a professor had said that this was the first time something had been done, whereas during discovery there was an email saying that a graduate student had done it before.

Eventually, it is the jury decides between competing facts. So a huge judgment on a technical issue might depend on twelve non-technical jurors struggling to understand if a register counts as a "pluraiity of flipflops" or some equally esoteric point.

Ultimately, it is the people like Tom and John who have the most to gain in a patent lawsuit. The lawyers and expert witnesses get paid whoever wins. Legal fees are often so expensive than even winning can be a phyrric victory. It takes two or three years, and can consume millions of dollars in legal costs.

Sign up for Sunday Brunch, the weekly Breakfast Bytes email.