Following State Street Bank and now Bilski, certain Internet business methods remain “patentable subject matter” under Section 101. Given the continued rise of the Internet and its increased role in nearly everyone’s everyday lives, it would be interesting to have a better understanding of how such Internet patents fare in litigation—especially in relation to ordinary patents that bear no relation to the Internet. Recently, a group of researchers, including John Allison of the University of Texas, Emerson Tiller of Northwestern, Samantha Zyontz of George Mason, and Tristan Bligh, published a piece in the Stanford Technology Law Review that continues this worthwhile empirical inquiry. The piece analyzed well over one thousand Internet patents, as well as several thousand non-Internet patents (NIPs).
Titled Patent Litigation and the Internet, their piece raises many intriguing empirical points. The piece generated evidence that Internet patents are far more likely to be litigated than NIPs, suggesting that Internet patent-holders may perceive their patents to have more value than the owners of NIPs perceive their patents. In absolute terms, over 10% of Internet patents were litigated, while only 1.36% of NIPs were litigated.
The piece also took the inquiry a step further. Rather than focusing on Internet business method patents as a whole, the researchers broke down Internet patents into two categories based on the patents’ claim language—those that are broad, “Internet model” patents, and those that are narrower, “Internet technique” patents. Both types of patents were far more likely to be litigated than NIPs, but they also found that the Internet model patents were even more likely (by about 33%) to be litigated than the Internet technique patents. This finding, to the authors, could support the idea that broader claim language increases the universe of potential infringers and promotes litigation. At the same time, however, the authors admit that there is not a good method of measuring the breadth of claim language empirically—further research could examine how to assess the breadth of claim language more robustly.
Beyond revealing whether Internet patents were litigated more frequently than NIPs, the study also provided useful information about the outcomes of litigation for Internet patents and NIPs by examining relevant cases involving the patents. Unsurprisingly, a large proportion of cases involving both Internet patents and NIPs settled (about 80% of for Internet patents and 65.84% for NIPs). The greater settlement rate for Internet patents was a result of the high rate of settlement for the narrower Internet technique patents (87.16%), but the authors were not able to provide conclusive reasons as to why suits involving the “technique” patents settled more frequently. Other interesting results of the paper included the fact that suits involving both Internet patents and NIPs rarely went to trial (3.19% and 4.95%, respectively), and the fact that patent owners were far more likely to win their cases that reached a formal judicial disposition when they litigated NIPs than when they litigated Internet patents (16.34% win-rate compared with a 3.19% win rate).
The paper contains many more interesting empirical results that are not mentioned in this post (indeed, it contains 17 concluding points), so I would encourage you to read the paper if you are at all interested in the empirical aspects of patent litigation. Regardless of your interest in these empirics with respect to Internet patents specifically—or with respect to patents more generally—the study provides the reader with useful information on the behaviors of IP litigators and has useful insights into some of the empirical trends in IP litigation.
While one could write an entire blog post on the questions that this article raises in the IP field, I am most intrigued by this question of narrow vs. broad patents and the implications that narrower or broader claims may have on (1) the propensity to litigate; and (2) litigation outcomes. Moreover, the article raises interesting questions from a behavioral law and economics standpoint: do litigators artificially perceive Internet patents to be more valuable than they actually are, given society’s current immersion and interest in the Internet? Or are such perceptions of value warranted? Further, is the distinction between Internet “method” patents and Internet “technique” patents devised by the authors a meaningful one? If so, why are accused infringers more likely to settle in cases involving the “technique” patents? There may be an explanation for this behavior that should be further examined by the authors and other empiricists. Finally, it will be interesting to observe whether the authors’ results (specifically those distinguishing between Internet patents and NIPs) hold up over time as the holding in Bilski is applied to circumscribe the sorts of business method patents that may be issued by the PTO (and, hopefully, more certainty surrounds the issuance of these patents).