The article is one of the few that talks about the GAO report on patent litigation. You'd think it would be all over the media. Somehow it isn't. An abstract o

The article is one of the few that talks about the GAO report on patent litigation. You'd think it would be all over the media. Somehow it isn't. An abstract of the report from the GAO website itself, worth reading as a TL;DR: http://www.gao.gov/products/GAO-13-465 http://www.gao.gov/products/GAO-13-465 The really interesting news here is that the report found that "troll" related lawsuits are actually a minority of patent litigation, somewhere around 20%, and that is despite a sharp rise in the recent years. The EFF [1] makes a good point that this data does not look at non-litigation trolling preying on small entities and unrelated businesses, a la Lodsys, that extract payment simply through the threat of litigation. (Note that the EFF headline is also not the best representation of the report.) However, such operations have typically attracted media attention. There have been many, but the first I recall reading was of Fotomedia, who sued anyone that had photo sharing on their website. That was a miniscule operation compared to Lodsys, yet it made news. (Ars Technica tracks a few such trolls. It has not yet covered this report, even though the OP has been a regular Ars contributor.) For now, I find it hard to believe that there are enough of these operations to change the GAOs conclusions. The other interesting observations are: 1. the rise in litigation, troll-related or otherwise, is attributed in part to the recent changes in patent law, specifically, the AIA. What this report, and many other studies in the area, miss is that even older decisions [2, 3] have had similar effects, something I have previously pointed out using another throwaway account. 2. companies are realizing patents are something that can be monetized rather than just being used to protect ideas. The OP glosses over all this and talks about vagueness in software patents, which is something I partially agree with. (Specifically, I believe patents being issued after the ~2004 - 2005 have much narrower claims.) However, it's not something that is endemic to software patents. The Doctrine of Equivalents [4] means that any element in any claim of any patent can be interpreted to cover other ways to implement it as long as it performs the same function. For instance, if the claim mentions using email, but somebody does it using SMS, it may still infringe if the main function is sending a message or notifying a user. However this, in turn works against the patentee when prosecuting patents at the USPTO, because examiners then apply the "broadest reasonable" interpretation of claims to find more and more prior art. 1. https://www.eff.org/deeplinks/2013/08/gao-study-confirms-obvious-bad-patents-lead-trolls https://www.eff.org/deeplinks/2013/08/gao-study-confirms-obv... 2. Michael Donovan, "The Impact of MedImmune, Inc. v. Genentech, Inc.and Its Progeny on Technology Licensing," http://digitalcommons.pepperdine.edu/cgi/viewcontent.cgi?article=1038&context=jbel http://digitalcommons.pepperdine.edu/cgi/viewcontent.cgi?art... 3. Sara Jeruss, Robin Feldman, Joshua H. Walker, "The America Invents Act 500: Effects of Patent Monetization Entities on US Litigation," http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2158455 http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2158455 4. http://en.wikipedia.org/wiki/Doctrine_of_equivalents http://en.wikipedia.org/wiki/Doctrine_of_equivalents