Patent litigation for dummies

July 20, 2007 6:55 AM

I'm looking for an outline/timeline of the steps/stages of patent litigation in the US. Can anyone help?

Ideally, I'd love to see a list of all the hearings, motions, whatever that are required or common to patent litigation suits, and when along the timeline of a suit they occur. Are there any online resources that would have this kind of thing along with a brief description of what's involved in each step? I'm not a lawyer (obviously) so if what I'm asking for doesn't make sense, please let me know. Much obliged...

posted by loquax to law & government (9 comments total)

Well, do you want the abstract, idealized version or the true-to-life version? The true-to-life version varies a lot depending upon the district the suit is filed in. Certain so-called 'rocket docket' districts (Virginia Eastern, Wisconsin Western) handle patent litigation very quickly, though they do so mostly by having very, very short bench trials or disposing of the matter with quickly argued summary judgment motions. The Verizon v. Vonage case, for instance, was decided in Virginia Eastern on the basis of 30 minutes of oral argument per side.

Also important to note is that only a few percent of federal civil cases actually end in a trial. The vast majority are settled, dismissed, or disposed of via summary judgment. So in the true-to-life version, the case almost always settles out of court during or shortly after discovery (i.e., the evidence-gathering stage).

So, given all that, I suppose what you probably want is the abstract, idealized version complete with jury trial?
posted by jedicus at 7:32 AM on July 20, 2007

Yes, I guess I'm looking for the details of a theoretical complete patent litigation trial, from filing to verdict. As standardized as possible I guess, if such a thing exists, excluding quirks in this district or that...thanks for the clarification...
posted by loquax at 8:00 AM on July 20, 2007

I do not assert that the following is typical at any particular firm, I merely recite some common stages based on class work, reading cases, and general observation. Relevant to U.S. practice only.

Before filing a complaint:
- plaintiff (P) contacts law firm, inquires about infringement case. Discuss fees & other matters to determine if client actually wants to continue.
- identify patents at issue, accused products/services, perform prior art searches, possibly obtain opinion letters on patent validity, sketch out damages to again re-assess whether it's worth going forward.
- engage in saber rattling by calling/emailing/writing defendant, see if settlement is possible.

P: File complaint / serve process / amendments by right

Defendant (D): Pre-answer motions
- allege court has no jurisdiction, P has not stated a claim, etc.
- typically leads to further amendments of the complaint by leave of the court.

D: Answer, affirmative defenses, counterclaims
- affirmative defenses include "your patents are not valid for the following zillion reasons," which is why they were vetted before the start of litigation: in anticipation of these defenses.

P & D: begin discovery
- zillions of requests for admission, demands for document production, interrogatories
- especially popular are "30(b)(6)" depositions, where you depose "a corporation", essentially requiring them to send an agent who can answer your questions, or tell you who can.
- the results of discovery are the raw material for trial-admissible evidence.

P & D: obtain expert witnesses
- Will typically address questions such as the state of the art in a certain industry at a certain time, the ordinary skill of a practitioner in a given industry, etc - mostly for patent validity issues.

P & D: pre-trial hearings
- Regular scheduling conference, like all federal litigation
- Unique to patent litigation, Markman claim construction hearing. As per an appeals court decision, deciding the meaning of patent claims ("claim construction") is a matter of law for the court, and not for the jury. Therefore the judge sets a Markman hearing before trial during which he will decide the meaning of the patent claims.

A cynical perspective on Markman hearings is that the judge looks at each party's asserted claim sets, picks one, and says "this is what the patent means." Some judges are more sophisticated in their analysis, but the Federal Circuit has a VERY high reverse rate on claim construction, so some courts seem to think "why bother, I'm likely to be reversed anyway."

It is difficult to overstate the importance of the Markman hearing. Depending on exactly what the judge holds the patent means, the issue of infringement frequently becomes either a shoe-in or a complete hail mary. Parties frequently settle or successfully move for summary judgment after the Markman hearing, thus avoiding trial completely.

Typically this will be followed by an appeal to the federal circuit, which will do the REAL claim construction and then it's pretty much over (The supreme court rarely takes patent cases). But:

P&D: Trial
- if any factual issues remain in dispute after discovery and Markman, the parties can move on to a trial.
- In the patent validity area, certain issues related to the obviousness of a given invention may be resolved by a jury. Typically, accused infringers hate for it to reach this stage, since it's easy to make inventors appear brilliant and their inventions non-obvious.
- In the infringement area, the ultimate issue of infringement is up to the jury.

P/D: Appeal
- If either party doesn't like the outcome of the trial, they may appeal to the Court of Appeals for the Federal Circuit, which by federal statute has exclusive jurisdiction to hear appeals of patent cases. See above.

Oh yeah, not to mention Patent reexamination proceedings, which are quasi-litigation affairs before the USPTO wherein one party tries to invalidate another's patent. Typically occur before litigation, but sometimes as a last-ditch effort during litigation by an accused infringer, since under some circumstances the decision of the PTO is considered binding on the court (the district court at least; reexams can be appealed as well). See, e.g. the Blackberry patent fiasco, where RIM finally tried to get the Blackberry patent reexamined late in the litigation game.

OK, I'm going to go study for the bar now. I wish patent law was on there.
posted by rkent at 8:10 AM on July 20, 2007 [1 favorite]

Discuss fees & other matters to determine if client actually wants to continue.

Oh yeah, I should've mentioned: patent litigation is wicked expensive; the numbers you typically hear thrown around are $2-$3M in fees if $5M or more is at issue. That's why there's so much pre-litigation assessment of whether not it's worth continuing.

This may also explain the popularity of reexams as opposed to litigation, as it is cheaper and something like 70% of patents challenged by a 3rd party are overturned during reexam.
posted by rkent at 8:13 AM on July 20, 2007

Step Zero: Obtain a patent. This kind of goes without saying.

Step One: Notice potentially infringing activity. At this point, you'll want to contact a lawyer to ensure that yes, what they're doing is at least colorably similar to what your patent covers or else there could be sanctions for you, your lawyer, or both for a frivolous suit. NB Patent infringement generally amounts to making, using, selling, offering to sell, or importing into the US anything covered by the patent. There is also liability for indirect infringement, which amounts to contributing to or inducing infringement by another party.

Step Two: Send a cease-and-desist letter. This letter will communicate to the alleged infringer that you have a patent and your concerned about what they're doing. What it (generally) shouldn't do is make specific allegations of infringement or offer a license. Doing so may open the door to a declaratory judgment suit, which would allow the alleged infringer to strike first by suing for a declaration that they do not infringe the patent (or worse, that the patent is invalid or unenforceable). Those same issues would come up in an infringement suit, but the DJ suit allows the infringer greater control over the district the suit is brought in (see above for why this is a big issue).

Step Three: Assuming the C&D letter fails and a DJ action is not brought, the patentee will start a lawsuit by filing a complaint in district court alleging infringement of the patent. The plaintiff/patentee will demand a jury trial at this point, if it wants one.

Step Four: The alleged infringer will file an answer to the complaint, typically denying all of the counts and possibly counterclaiming (common counterclaims include antitrust violations and unfair competition). The answer will also include a lot of affirmative defenses (non-infringement, license, patent invalidity, inequitable conduct, laches, estoppel, etc). If the defendant/alleged infringer wants a jury trial, it will demand one at this time.

Step Five: There will be a long, expensive discovery process. Both sides will demand all kinds of information and records and take many depositions. Expert witnesses will be called to make reports and those witnesses and reports will be challenged according to the Daubert rules (in a nutshell, the witness must use reliable methods reliably applied to come to their expert opinion).

Stept Five and a Half: (This may come before discovery, after discovery, or after or a part of summary judgment motions. Either way, it must come before the trial). The Markman decision. The judge will decide, as a matter of law, what the claims of the patent actually cover. This is called 'claim construction' or the Markman decision, after the case that made it both necessary and a matter for the judge (as opposed to the jury). Typically, once the judge has construed the claims, the writing is on the wall. It will be obvious to any competent patent attorney which side is going to win. Now, some times this is not entirely clear, and that's most often when the cases go to trial (or if the defendant/alleged infringer has a good argument regarding patent invalidity). And sometimes the patentee may 'lose' the Markman hearing but still have a case under the Doctrine of Equivalents, which generally holds that one can infringe a patent by accomplishing the same function, in the same way, to achieve the same result even if one does not literally infringe the patent (NB the DOE has another formulation dealing with the 'insubstantiality of the differences' between the patent and the allegedly infringing product or process). Some courts favor the DOE more than others.

Step Six: There will probably be some summary judgment motions with each side arguing that, even if the other side is to be believed, they still haven't proven their case.

Step Seven: If the SJ motions are denied (more common in some districts, such as the District of Delaware, than others), then the case will move to trial. If a jury trial was demanded by either side, there will be a long, long wait (longer in some districts than others). If there is to be bench trial, the wait will be less long.

Step Eight: Prior to trial, both sides will make various motions-in-limine ("motions at the threshold"). These are motions asking the judge to make decisions before the trial, typically on evidentiary issues. Some times these will be additional Daubert motions, sometimes they will be more mundane issues of whether evidence is inadmissible as hearsay/without foundation/etc.

Step Nine: Trial. (dun, dun dunnn!) Both sides will present evidence. Typically the patentee will tell an 'invention story,' especially if the inventor is personable and especially if there is a jury. This helps establish in the mind of the jury that the inventor is a good guy/gal who had a really neat idea and, gosh darn it, should be rewarded for his/her efforts. The alleged infringer will counter by claiming that someone else thought up the idea first (i.e., that it was anticipated), that the idea was obvious, that the patentee pulled one over on the patent office (inequitable conduct), etc, etc. Both sides will present lots of expert testimony and both sides will attack each other's experts. (There's always a way to attack an expert. For instance, if the expert is being paid, then he/she's a prostitute; if the expert is not being paid, then they're on a crusade).

Step Ten: The verdict. The patent may be found valid and infringed (a win for the plaintiff), valid and not infringed (the plaintiff lives to fight another day), or invalid (the plaintiff loses completely. The patent may also be found 'unenforceable,' which is the result of inequitable conduct or patent misuse.

Step Eleven: Remedies. Assuming that the patentee wins, he or she will receive money damages. This may take the form of lost profits or a reasonable royalty. Which one the patentee wants depends on a lot of things but most often on whether the patentee practices the invention or not. If the patentee does not practice the invention (e.g., is a university or a patent holding company/troll) then it will want a reasonable royalty on the defendant's sales because the patentee had no profits to lose.

Damages may be lessened if the defendant did not have notice of the patent's existence. If the patentee fails to mark its products as patented, then the defendant will not have to pay damages on any infringement up until the time that it received actual notice of the patent's existence. Typically this is either when it got a C&D letter or when the lawsuit started.

Damages may be increased (up to 3x actual damages) if the infringement was willful. Infringement is willful if the defendant has actual notice of the patent's existence and infringes anyway. Willfulness may be mitigated or disproved via a variety of ways but the most common defense is to have an attorney look at the patent and the allegedly infringing product and prepare an "opinion of counsel" that the product does not infringe or that the patent is invalid. If such a letter exists and the defendant chooses not to produce it, the law does not allow the patentee to point and say "well, then the attorney must've thought that they did infringe!" This is a relatively recent change brought about to protect attorney-client privilege. It's important to note that the so-called 'opinion counsel' and the trial attorney should almost certainly be different people and probably work for different firms. The reasons for this are a somewhat complex matter of legal ethics and evidentiary rules beyond the scope of this comment.

Also, the plaintiff may request reasonable attorney's fees and costs if the infringement is willful. This is unusual in American law, as you may know, and requests for fees (and treble damages, for that matter), are often not granted.

The patentee may also request an injunction. The primary purpose of an injunction is to allow the patentee to haul the defendant back into court in a contempt-of-court proceeding if the defendant goes back to its old infringing ways. Contempt proceedings put the initial burden on the defendant to prove that it is not infringing rather than upon the plaintiff to prove infringement. It used to be the case that the granting of an injunction was automatic in infringement cases, but the eBay v. Mercexchange case changed that. Now injunctions are granted based on the same 4 equitable factors that govern the grant of an injunction in all other cases.

Step Twelve: Appeal. An enormous number of patent infringement cases are appealed. All of them, regardless of geographic origin, go the Court of Appeals for the Federal Circuit. The reason so many are appealed (apart from the potential money at stake) is that the law of claim construction is a mess. The Fed Circuit overturns about half of all claim construction rulings brought up to it. So, all things being equal, you have a 50/50 chance of winning claim construction on appeal (which, as mentioned earlier, has a large effect on whether the patent is actually infringed (and also whether it is anticipated by a given piece of prior art under the 'that which infringes if later, anticipates if earlier' rule)).

The Supreme Court is generally reluctant to hear patent cases. There was a rash of cases recently but silence for a few years before that.
posted by jedicus at 8:25 AM on July 20, 2007 [1 favorite]

Oh, one last note about the rarity of trials in patent litigation. The main patent judge in the Eastern District of Texas (which is popular with patent litigants for various reasons), T John Ward, has overseen an astounding seven (!) patent jury trials. Apparently he is considered quite experienced in this regard. That ought to tell you something.

Okay, that number comes from an article I read about a year ago, it may be higher now but probably only a little.
posted by jedicus at 8:28 AM on July 20, 2007

Wow, thanks very much both of you, I really appreciate the effort, that's exactly what I was looking for, especially details like markman. Amazing, this really helps, thanks again.
posted by loquax at 8:47 AM on July 20, 2007

A couple more small notes. If the case goes to a jury trial, the whole bit typically takes about 3 years. With an appeal (and probably a remand back to the district court), it typically takes about 5 years. I've seen (in the observed sense, not in the had-anything-to-do-with sense) cases that went on for 10 years and are still ongoing. When a high-end patent litigation firm charges a blended rate of about $500/hour, you can see why the charges get so high so fast, especially once you toss in paying experts and court fees. You can also see why such massive litigation is almost only ever worth it to pharmaceuticals companies and a few other high-tech concerns with very valuable patents.

A very fast docket may be able to complete a case in about a year if both sides actually want it to go to trial, but it'll still get bottlenecked at the Federal Circuit if there's an appeal.
posted by jedicus at 9:10 AM on July 20, 2007

Obviously, YMMV. For example, looking at rkent's list, Susman & Godfrey, among other law firms, doesn't engage in saber-rattling until after it files a complaint; the reason is that saber-rattling might cause the target of the saber-rattling to file a declaratory judgment case first ("Please declare that we're not infringing the patent and that the patent is invalid and the patent-holder is violating the antitrust laws by threatening us"), and get to pick the forum (and thus avoid a rocket docket).

And if you're going to pick a rocket-docket, then you probably want to hire the expert before you file the complaint and have your infringement theory ducks in a row. I once saw a case where the plaintiff picked a rocket docket where matters would go to trial within a year of filing the complaint, didn't have its ducks in a row, and pissed off the judge so much in failing to produce its discovery in a timely fashion that the judge essentially ruled against it on every conceivable legal issue. A lot of people lost their jobs in the aftermath before the Federal Circuit set things right a couple of years later.
posted by commander_cool at 10:55 AM on July 20, 2007