Alright, Luke and I are resolved to update our stuff more frequently (yes, Luke, I resolved for you. Get to it.)
For my birthday, I got some wonderful news: Bill Gates is joining me
in calling for some improvements to the US patent system. Awesome! He also joins others
who -- perish the thought -- might be more public-spirited than the Vole?
How about a tit-for-tat?
MS: executives "recommended that the U.S. Congress end patent filing fees for small companies, nonprofit groups, universities and individual inventors" and "pushed Congress to end the diversion of patent fees from the USPTO to the U.S. government general budget, saying that the office needs more funding"
NA: "The patent office needs additional resources to hire and train more examiners, and to implement information technology that would boost its processing capabilities. "
I like it.
MS: "It is too easy for a litigant to manipulate the U.S. system and look to a patent lawsuit as the ultimate lottery ticket, hoping to confuse jurors with technical jargon that will yield the payment of a lifetime"
NA: urges "the USPTO to strenuously observe the statutory requirement known as the "nonobviousness standard," which says that in order to qualify for a patent, an invention cannot be obvious to a person of ordinary skill in a given area."
All good points, so far as they go. But the critical difference between NA and MS recommendations for reform is in the latter's silence on this issue:
NA: "Access to patented technologies should be available for research purposes and in the development of cumulative technologies, where one advance builds upon previous advances."
From the article: "In 2002, the federal circuit court ruled that scientific research is part of the "business" of universities and is not protected from liability. Prior to that decision, many in the research community presumed the existence of a research exemption and acted accordingly"
That bit encapsulates the IP crisis for me. All its implications might take some unraveling, so bear with me.
Software is not a product, nor is it a service
. It is a body of knowledge -- specifically, it is Very Abstract mathematics. Feed these bytes into a computer, and these other bytes will come out. Every time. Source code and GUIs and all the other stuff is simply a layer on top of that. Double-click this, or type that, and the interpreter/desktop manager will feed bytes into the computer. And these other bytes will come out.
Fedora Core 3 is only the most recent stopping-point in a long line of incremental advances over MULTICS. A microprocessor is an invention; a 64-bit microprocessor is an advance.
The NA does not find that the patent system requires a major overhaul, but I think the research exemption implies exactly that, if the fundamentals of software are understood.
Software ~= research != business
(That's the approximately-equal sign there)
There's a great deal more to say on this, but I'll wait for comments.