Patents support a fairly narrow model of what innovation is "supposed to be"-- the one man in a garage who can scale his product from zero to infinity with litt

Patents support a fairly narrow model of what innovation is "supposed to be"-- the one man in a garage who can scale his product from zero to infinity with little to no outside support. The thing is, for a lot of industries, that's a dated pipedream. It's not 1804 and you're not Richard Trevithick. You're not going to challenge TSMC with a semiconductor fab built in the backyard shed. Even if you had huge amounts of startup capital and a laser-focused market vision, there's also a lot of innovation that's time-sensitive: for example, if you're building a better petrol engine or coal-burning generator, will you be able to scale before the product is obsolete? Beyond that, I suspect there are also a fair number of patent holders who just take potshots at the market-- wildly mis-guessing their target market or (attempting to get orders or raise funds) based on promising undeliverables, because they don't have the full knowledge of getting a product to market. After their business fails, they'll just spend the rest of the patent term as legal unexploded ordnance-- useless by themselves but waiting to blow the leg off of some poor sod who happened to step in the wrong place. Reducing the term from 20 years to 10 or even 5 would tend to prioritize partnership development rather than the "one man in a garage" mentality: with the clock ticking that fast, your best chance is to license broadly, hoping that someone will get product to market before the patent expires.