Timothy B. Lee is a little off in his analysis. He refers to the Supreme Court not allowing algorithms to be patented. This is true. He is right here. But his next line about the Federal Circuits is off.
"Unfortunately, the Federal Circuit has ruled that merely loading an algorithm onto a general-purpose computer and executing it transforms it into a patentable invention."
That is also true, and furthermore, which Lee doesn't point out, is that the Supreme Court has affirmed this. See Diamond v. Diehr, 450 U.S. 175 (1981).
The often cited landmark case that affirmed this is State Street Bank. See State Street Bank & Trust Co. v. Signature Financial Group, 149 F.3d 1368 (Mass. 1998).
That decision ruled that an algorithm that is calculated by a computer meets the statutory subject matter of patentable inventions. The Federal Circuit in Massachusetts decided that case. The Supreme Court declined to review the decision because at that time the SCOTUS thought the Federal Court got it right.
I am still reviewing Lee's other analysis. There is lots of stuff there. It will take some more time.
In any case, if the principle stands that embodying an algorithm in a general-purpose computer does not meet the standard for a patentable device, so many bogus patents will be invalidated that it will be like Christmas. We can end this ridiculous cold war and get on with the work of creating great software without worrying about trolls and bullies.
We're not there yet but there's more hope now than I've felt in ages.
