Is obvious now defined as "predictable"

Life is full of subjective measures - Am I a "good" person? Depends. Is this a "long" piece of string? Depends. Is Playboy "pornographic"? Depends. Is that figure skating routine worthy of 9.8 or 9.9? Depends. Is my invention "obvious"? Depends.

By now, you probably know that the Supreme Court, in the "KSR" ruling, made getting a patent significantly harder by throwing out a test that had been [mis]used for many years. This week, according to Greg Aharonian of the Internet Patent News Service, the PTO issued revised guidelines for obviousness determination during patent examination in response to the KSR decision. According to Greg, who has time to read these things, the new guidelines effectively define obvious as "predictable".

Like trying to prove the non-existence of a non-observable phenomenon, being absolute about what's good, pornographic, or obvious is clearly impossible. If we agree with the premise that patents should only accrue to inventors who make a meaningful improvement (by which I "obviously" mean to make an improvement that was not "obvious" to the bulk of society) to their fields of endeavor, then we buy into the conclusion that some subset of society will have to sit in judgment of these improvements and try to apply a fair (= equally applied) standard to all inventions. That the current situation is a poor implementation of this goal does not invalidate the goal.

The standard to which inventors are held, however, will always be defined by a sense of what society believes or understands. The KSR ruling was to obviousness what Potter Stewart's words were to pornography ("I shall not today attempt further to define [pornographic] material...but I know it when I see it.") Or, to paraphrase Cole Porter, "In olden days a glimpse of prior art was looked on as something shocking but now, God knows, anything goes" (sorry Cole).

Anyway, I think we'd just better suck it up and work our arguments at the PTO or CAFC the best we can. I personally don't like moving from "obvious" toward "predictable" because I don't want to limit inventions to items developed by trial and error (like Edison's light bulb filament); I want to reward the inventor who uses whatever skills he or she has to predict a result that the rest of us didn't have the insight to consider.