I'm not a shilling for Zuckerberg but you are being too hard on him. First, the patent isn't for "summary of privacy settings", it is for a systems/methods to generate the said summary. Second, a patent isn't supposed to be original or clever. The criteria are 1) novel, which can be a non-original, i.e. a derivative work. 2) non-obvious (to one skilled in the arts) or in other words non-obvious to a patent examiner who doesn't know much except for some prior patents in the field.

Any software patent is 'non-obvious' to a non-programmer. That's the problem with them.

them = patents, or non-programmers?

Patents, of course.

Well the real problem is the patent examiners probably aren't programmers, thus won't have a good idea of what should be patentable.

For most CS students/programmers, I'd wager that being a patent examiner isn't very alluring. This is even more evident at career fairs where representatives from the USPTO have to get in your face to grab your attention. You just don't see too many students lining up to talk to them, rather than say, Google or Microsoft.

I can't decide if he was joking or not.

So why not give your opinion, and add to the debate? Do you think most software patents are non-obvious? How can a software patent be non-obvious? All programs are pretty obvious. Algorithms, not so much, but then you're talking more mathematics, which should not be patentable anyway.

>"All programs are pretty obvious. Algorithms, not so much."

How do you expect anyone to respond to something like that? Are you suggesting programs are not algorithms? Can you point to an authoritative reference explaining the difference?

I gave my informed opinion on patents, unfortunately this site is turning into reddit just like reddit turned into digg.

There's a big difference between "System and apparatus to display 'hello world'", and say an encryption algorithm.

Yes it's a blurred line, but I'd say most programs are todo lists, rather than complex algorithms of scientific merit.

For example, say the software that runs facebook - that's pretty much going to be 'grunt work' - lots of building blocks, shoved together, to make something work.

But a compression algorithm, has more of a basis in mathematics entropy etc.

My sincere apologies if this once again disappoints your view of what comments on hacker news should be.

(I have several software patents, but I firmly believe they are a terrible idea).

Sorry, but Windows has had the ability to generate summaries of privacy settings ever since NT and the creation of NTFS w/ ACL configuration. It's called "Effective Permissions" and tells you - bottom line, all settings and configurations taken into account - what each user can access and what they can't.

Obviously MS hasn't contested this patent yet, but I wonder what they'll do when they read it up....

Again, the patent isn't for "ability to generate summaries of privacy settings". The patent is for systems/methods to do so. Consider a patent for systems/methods of flying. Such a patent doesn't infringe on attempts of flying, just on systems/methods described by the patent.

Oh, yeah. Sarcasm detector. That's a real useful invention.

Note to readers: this is a patent application, not an issued patent.