News Center

Elsewhere Online twitter Facebook SLS Blogs YouTube SLS Channel Linked In SLSNavigator SLS on Flickr

Trolling For Trouble

Publication Date: 
February 01, 2011
Intellectual Property Magazine
Maura O'Malley

Professor Mark Lemley has produced many reports on the "patent troll" phenomenon and was thus mentioned in the following article in Intellectual Property Magazine, covering the history of the term and how the concept works:

The word troll comes from the Norwegian folklore describing an ugly, dim-witted monster who lurks under a bridge threatening the goats that dare trespass the bridge. The term "patent troll" was originally attached to companies that buy patents solely for the purposes of licensing and litigating, but who never actually create, use or manufacture the patents themselves. It was coined by former Intel general counsel Peter Detkin in 2001; at the time, Intel was fighting many infringement claims from smaller, non-manufacturing companies.

The term quickly caught the imagination of the patent community and media, but many railed against the word's negative connotations and the emotional reaction elicited by labelling someone a "troll". Now people generally prefer the more broader and neutral expression non-practicing entity (NPE). But NPEs encompasses not just "trolls" that buy patents solely to license or litigate, by benign entities like independent inventors, universities and other research institutions that pump innovation back into the system, but have no ability to commercialise it.


Stanford law school professor Mark A. Lemley has co-authored a number of influential papers on the patent troll phenomenon. He says that following the decision "we have seen some increased willingness to fight cases to trial. But I think the primary effect has been to reduce the amount paid in settlement, since the patent no longer threatens to shut down the defendant's whole product."


In one of Lemley's studies he and others looked at 106 patents that had been the subject of eight or more lawsuits since 2000. They showed that the patents in this group that were most often asserted were actually the most likely to lose if the case went to trial. The authors found that patents litigated only once win in court almost 50% of the time, while the most frequently litigated and supposedly most valuable patents won in court only 10.7% of the time.

The worst court performers were software patents and patents owned by NPEs; the study explored some possible explanations for why a group of apparently weak patents nonetheless had so much influence over the patent system.

He says: "One possibility is that the patentees filing those suits either have or think they have really broad patents; that's why they have sued so many people for infringement. The broader a patent claim is, the more likely that claim is to be invalid, because it is more likely to cover something in the prior art. So the most-litigated patents may be involved in so much litigation because they are over-claimed. Alternatively, it might be that because most cases settle, patentees can make money by suing hundreds of defendants even on very weak patents, knowing that most of those defendants will settle rather than invalidate the patent."