Tonight, during Lost, I tried to decide just how “willful” Locke-the-smoke-monster was being. I didn’t reach a conclusion – and doubt I will until the end of the season. In the mean time, my anonymous lawyer friend known as Sawyer sent me a nice essay on the notion of “willful infringement in the context of patents.” He also sent me some interesting comments about how pissed off a bunch of biotech patent lawyers were today are at U.S. District Court Judge Robert W. Sweet for invalidating seven patents related to the genes BRCA1 and BRCA2, but that’s not what this essay is about. Remember, no matter how sensible Sawyer’s advice is, none of this is legal advice and, if you have a specific issue, make sure you spend lots of money on your lawyers since they also have to eat. Sawyer’s comments follow.
One of the issues that seems to most concern developers in software companies is whether they should ever look at a patent, and if they do, whether they are "tainted" by the exposure (a patent virus, perhaps?). The basic wisdom is that developers should avoid looking at or reading any patents, and I think that’s generally right, but the law isn’t as bad as it sounds.
There are two issues springing from exposure to a patent – the ticking off of the damages period for infringement of the patent, and willful infringement. The first issue is relatively easy: in order for a patentee to begin accruing damages, he must put a potential infringer on notice of his alleged infringement with some level of specificity. The requisite specificity is a thorny legal issue, but it’s not just enough to send someone a patent, a patentee must also make an allegation of infringement by something (so, the theory goes, the alleged infringer can stop infringing). Patentees can also sell products embodying the invention and mark them with the patent number, which counts as constructive notice, usually.
The second issue is the one I want to focus on. The definition of willful infringement and its legal requirements have changed a lot over time. It used to be something akin to "continuing to infringe when you knew someone was accusing you of infringement." Under the current legal regime, the alleged infringer must continue to infringe in a manner that is both objectively and subjectively reckless (the law is still muddy and some will quibble with my framing of the test, but this is roughly it). Objectivity in law means from the perspective of a hypothetical "reasonable man" (a legal fiction if there ever was one); subjectivity means from the perspective of the accused himself, i.e., that he believed or should have known that what he was doing was reckless and did it anyway. The idea is that if there are strong reasons for people to think that an accused infringer does really infringe, e.g., because there are weak litigation defenses, then his continued infringement is willful; but it isn’t enough to infringe just with mere knowledge of the patent, or even because of negligence in figuring out whether one infringes.
The upshot is that just being exposed to a patent, and then later accused of infringement, isn’t anywhere near enough to be "willful," and so developers shouldn’t be too worried if they run into something accidentally. That said, exposure to a patent is evidence that can come in at trial, and plaintiffs will use even tenuous exposure to paint a picture that could cause a jury to find willfulness even in the absence of solid, legally cognizable, evidence. So, people involved in a software business should probably still avoid patents (not the least of which because they’re often incomprehensible anyway), just to be safe.
Also, one more note: People think that a finding of willfulness means automatic treble damages, but the decision to enhance damages at all is discretionary with the trial judge, and he can choose to make no enhancement at all, or enhance to a level significantly less than treble. The real problem with willfulness is that it gets "black hat" stories about defendants in front of juries, which taint them against a defendant regardless of the merits – it’s easier to think that someone infringes if he’s just a bad guy anyway, especially if the technology is boring and the patent is hard to understand.