"Anticipation" has a narrow definition, as used in the examination process. It is not the dictionary definition that we use every day. The narrow definition is that the invention must be described in a single reference.
The combination of PTO "anticipation" and "obviousness" make up our everyday notion of an invention being "anticipated" by other, preceding inventions.
The combination of nuclear reactors, Closed cycle steam turbines and submarines -- is it "non-obvious?". . . . an inventive Navy guy named Hyman Rickover (maybe some fiction here, but it personalizes the scenario) put the existing elements together along with a few ideas of his own to come up with the nuclear submarine.
At first blush, I would tend to argue that the use of nuclear reactor to power the steam cycle on a submarine is obvious. But you raise a critical point, without emphasis. The "along with some ideas of his own" is often the inventive step that justifies the grant of a patent. In this case, perhaps it is technology that permits the shrinking of the reactor.
One test for obviousness is whether the "new" invention serves a long felt need. If it IS obvious, then why wasn't it done already?
Just theoretically speaking, is a patent examiner somehow precluded from using hindsight to determine obviousness as well as anticipation in judging the merit of an application?
Yes. The benefit of hindsight is not permitted. One must argue from the incentives to combine, use or otherwise innovate. There are many many interesting examples. I'll draw one from the PDR, where a certain dose of a certain medication was indicated for the treatment of ulcers. Then, somebody figured out that in massively larger doses, this medication stifled weight gain. Old medicine, new use - PATENTABLE! The argument was that the existing art taught away from the new use. The PDR said DO NOT exceed dose, etc.
Seems like there was a Commissioner of Patents in the 19th century who said something like "Everything useful has already been invented."
And patents were mighty hard to come by in that era. The pendulum swings both ways. As history is our guide, today's environment of "easy" patent grants will not endure.
I pity the poor examiners! Software and electronic inventions can be exceedingly complicated, and producing some of them even with the benefit of all the drawings, diagrams, and other related documentation can still require an immense amount of skill with the needed equipment and fundamentals involved. How difficult it must be for a patent examiner to even understand what the invention's about (except in the most cursory terms) much less be able to "make" one as a person of "ordinary skill in the art." Mousetraps and things you can take apart and see aren't too difficult to visualize, but nobody's ever seen an electron do anything, and software is a whole different world from that!
Projecting increased complexity forward, I'm afraid the courts are going to be more and more involved because any individual patent examiner would need years of full time attention to a single complex application to untangle the complexities in order to make an informed decision on novelty and usefulness. Maybe some of the "easy" patent grants are already an indication of an "let the courts settle it" tendency -- sort of a Pontius Pilate examination system.
It could be: prosecution of the application in the USPTO followed by rejection followed by an appeal wherein the government would have to prove the invention was not worthy of a patent. While a patent under the examination system has a presumption of validity, overturned rejections would soon destroy any presumption of invalidity due to rejections by examiners who are simply not humanly capable of handling extremely complex inventions.
That would be something like an expensive, round-a-bout registration system, I guess. One thing is for sure, things won't stay the same as they are now! (I can say that with confidence because they never have.)
Great discussion, and very informative! Thanks!