Doesn't it depend on what the math involved is? If you have an invention claiming A+B+C, you don't need each of A, B, and C to be independently patentable, novel, and nonobvious. In fact, if you had them, you'd then just claim A, B, and C separately.
More concretely, if you claim [an automobile chassis] (known)+[4 wheels] (known)+[a micro sized cold fusion-based electric power plant] (unknown), why is that not "worthy of a patent"? After all, it can be ANY wheels. Goodyear, Michelin, BF Goodrich, or other? Any chassis can be used, nearly any hardware configuration for those other two elements. So therefore, cold fusion power plants shouldn't be patentable, because they're in the same ballpark as "on the internet" types of patents?

Of course not. And yet, for some reason, this gets thrown out the window when we talk about software.

What's tough about that? Had someone else come up with those instructions or that cold fusion based engine before? Then what does it matter if the wheels of the car or the processor of the computer already existed? You're not relying on those elements for patentability, but the new stuff that you invented.

Unlikely. Forget your desire to see them abolished and think pragmatically about the amount of money involved. They're not going anywhere, and arguments that they should be abolished are going to fall on deaf ears in Congress. It would be more fruitful to focus arguments on how to reform the patent system to address trolls, poor quality of examination, etc.

Not a good recent example, though - the UDRP decision did indicate that it was a possible trademark violation because visitors could confuse it with the Glenn Beck mark. Eiland-Hall won on the other two factors - lack of bad faith, and that he had legitimate interests in the name.

Ah, but fortunately, we live in a representative democracy, and not in a "my situation is different than everybody else's, but nonetheless, the laws should be written specifically for me" dictatorship. I have no plans for kids either, but am not so selfish that I believe that people who do want kids should be impaired in their ability to pass on an inheritance. After all, I realize that I too had parents and inherited from them. Perhaps this does not apply in your case?

No, his objection points to the term needing to exist after the inventor or author's death, to protect older inventors and authors. That the term is too long overall is a different argument. You could have a 5 year term, but that doesn't help the author who dies the day after publishing if it doesn't extend past his or her death.

Not so. Patents are absolute monopolies that can be used either for injunctive relief, which allows no royalties by definition because the infringer is no longer allowed to continue infringing; or for monetary damages, which allows a "reasonable royalty". In other words, you can either get 0 royalties and have an absolute monopoly, or you can get a reasonable royalty and no monopoly. You simply cannot have "any and all royalty rates," like 100% of revenue or infinite dollars per unit or whatnot. Check out 35 USC 284.

Method patents haven't been "practically negated by Bilski." In fact, the Supreme Court explicitly said in Bilski that software and business methods were patentable, provided they were not merely abstract ideas. For example, a claim that recites software executed by hardware is not "a mathematical operation" any more than a machine with a mathematical gear ratio is "a mathematical operation" - or rather, both are executing mathematical operations, but are not merely mathematical operations.

Yes, all software is math and alone, without being tied to machine or otherwise non-abstract, is unpatentable. However, math executed by a machine is patentable because, no matter how much you hate software patents, you can't simply ignore half of the patent claim in order to declare it invalid.

Agreed, but that's not what we're talking about here. You specified "known" algorithms. H.264 encoding wasn't a "known" algorithm prior to 2001, even without a computer.

(i) Yes, but only those claims that invoke means-plus-function limitations under 35 U.S.C. 112(6). System claims, method claims, and Beauregard claims would still be valid;

(ii) And no, most "with a computer/on the internet" patents* do include a "full listing of the algorithm". They aren't required to include source code, or even pseudocode, but rather, as the judge notes:

And honestly, it's pretty rare that a software related patent doesn't include a flow chart these days. Consider the beloved Amazon one-click patent - it has 5 flow charts, starting at figure 3. The fact that these three patents (really one patent and two continuations with the same specification and figures) didn't include a flow chart is an anomaly, not the norm.

*there are no "with a computer/on the internet" patents. People on Slashdot may characterize them that way, but there are no patents that include a claim of "A method, comprising: [known process], on the internet." The mere addition of the limitation "on the internet" can't be the key to patentability, because the internet is known. As a combination of two known pieces of prior art, it would be invalid under KSR. Instead, you'll always find that there's some additional element in the claim that is not involved in the off-line operation.

I believe this FUD got started because someone saw a dependent claim that said, "The method of claim 1, wherein the network is the internet," and suddenly thought that that was the key to patentability of the entire patent. Not true. It's really just a doctrine called claim differentiation. Patent claims are like Venn Diagrams, where dependent claims are bubbles within the large bubble of the independent claim from which they depend. So, if claim 1 says a network, and claim 2 says the network is the internet, then claim 1 must include other networks including LANs and VPNs as well as the internet. But that's it - claim 1, which never explicitly mentions the internet, must still be patentable on its own.

The Patent Office doesn't make law or decide legal issues - they decide factual ones. The patents in this case were not invalidated as obvious or anticipated by some prior art, which would be a matter of fact. Instead, the judge determined that 35 U.S.C. 112(2) requires disclosure of an algorithm when claims utilize the means-plus-function format of 35 U.S.C. 112(6). That's a matter of law. In other words, the "science of these things" is the science of jurisprudence, not the science of video encoding.

Or, short version: the Patent Office applied the law as it was properly understood at the time. The judge has now said, "no, that's not the proper law, it's this instead."

Uh, no, it doesn't grant what the summary said, and the fact that you ignored half of the limitations in the claims makes me wonder if those people in be examination corps were ever actually your colleagues.

Because, contrary to what the above posters have ranted about, this patent does not claim an entirely known process "... ON A COMPUTER!" In fact, as you note, the Supreme Court said they weren't patentable in KSR (and they weren't patentable before it either) and there are NO totally known process "... ON A COMPUTER!" patents. Not one. People might try to paraphrase a patent that way, but their description is not the patent - the patent is the claims, and there are no issued patents with a claim like that.

Ah, but no one ever said that Robertson screws wouldn't work or violated fundamental laws of physics without explanation. Robertson screws are easily measurable... microwave drives, not so much. Bad analogy on you.

GP is right. Actions for infringement of a registered mark are unde sec. 32 of the Lanham Act. Actions for infringement of an *un*registered mark - I,e., a "common law" mark - are under sec. 43(a) of the Lanham Act. There's a reason why wiki is useful as a guide to find primary sources, like the statute, but is not a replacement for those sources,

And yes, I am an IP lawyer.

That's a huge presumption that your argument relies on that you haven't shown to be true.

Many (but not all) conspiracy theories are actually rational lines of logic, stemming from different presumptions - such as whether "all the evidence" is actually available or not. To dismiss them out of hand simply because you have a different, unproven assumption is to make the same logic error you're accusing them of.

Uh, yeah? It says it right there. What do you think "exclusive rights" are used for except excluding others and/or selling licenses?

Ah, so you want the USPTO to qualitatively classify each patent as being "good for society"? So, for example, if a teetotaler is in charge of the office, no patents on new distillery technology. Or if a luddite is in charge, no patents on anything related to computers. Or if it's one of those anti-gun, anti-video game folks, then no patents on anything related to firearms or gaming? What about a Ralph Nader "Unsafe at any speed" person who bans patents on automobiles? What if it's a hardcore Christian who believes that patents should only be granted for things that lead one away from sin?

How does society benefit by having each inventor's protection be allowed or not allowed based on the capricious whim of some political appointee?

It's certainly one of the useful arts. The terms in the Constitution are archaic uses - "science" refers actually to things like literature, art, paintings, etc. that are covered by copyright, and "useful arts" is in contradistinction to aesthetic arts and refers to things that do something, like machines, industrial processes, or useful articles manufactured by industrial processes. In this case, the patent claims a machine and an article of manufacture that are both useful to industry. Why shouldn't that be patentable? Simply because you think it's unfair that a corporation makes money? If that's a legitimate complaint, then shouldn't your solution have something to do with taxing corporations or limiting profits, rather than creating a blanket rule that will harm sole inventors?