Proof from either of you?

Whether someone is anti-science or not, pointing out corruption in the field of science is a good thing. Corruption wastes time, money, and can hurt people.

The same thing can be said of gross errors that drastically change results.

It is a reasonable example of science working. I say reasonable because it wasn't initially a lack of duplication of results that sparked concern, it was alleged plagiarism and image manipulation.

It's a pity all research results weren't required to be duplicated by an independent team as a prerequisite to being published. And then peer reviewed in light of the secondary results.

"In the biomedical research field, everybody fabricates results."

A nice anecdote. Do you have any data to back that up?

It does seem very similar to the shape a badly twisted slinky can make.

"That's a result of the Paris Convention treaty, decades ago."

Is that this one: http://www.wipo.int/treaties/e... ?

Thanks. I do like your line of reasoning.

Thanks. That does make things clearer in that regard for US law.

The wiki well may be valid as a generalised international version though - would you think so? (a lot of us on /. are not from the US).

Is the Federal Circuit jurisdiction nationwide in the US?

"it would be good if you'd inform me first."

Noted, but it worked after the fact as well.

"But trivial and obvious have little to do with each other. E=MC^2 is trivial, but not obvious."

They often have a lot to do with each other - just not always - I'm pretty sure I expressed that. (note: e=mc^2 is by itself simple but part of a fairly complex system of equations).

"You can make an FPGA as complicated as you want and its reconfiguration can be a "higher level program". In fact, forget even field programmable arrays - from a purely theoretical basis, you can make a fixed (albeit huge) circuit that will execute a specified program, even a "higher level one". You could hard code Diablo III, if you had enough transistors and an infinite amount of patience. Are circuits patent eligible?"

FPGAs are not a common CPU and only make a tiny proportion of the CPU market. It's reconfiguration could be executed by a higher level program, but it will still happen at a low level. Further to what I wrote in regards to the FPGA - I would only allow code to be patented as part of the CPU - i.e it would have to form part of the circuit itself in some special way. The code sent to it from a higher level would not be protected. I don't know the exact mechanism for reconfiguring FPGAs though - so this is a little guess work. Basically - if it's just software (which is abstracted maths) it wouldn't be protected.

Yes you could implement in hardware a lot of software (think ASICs). I wouldn't accept a patent on that - except if it was from an amazing technique in actually making the circuit - i.e. some novel way of implementing it in hardware. Further abstracting a piece of software won't make it anything but what it is - still just that - software.

My boundary was expressed as anything other than something that might alter a FPGAs logic. Basically all software. It doesn't matter how you store it or what language you code it in - no software patents. I don't know of Bilski's hedging algorithm so can't comment on it (but if it's a piece of maths - which all software is - then I don't care for it).

"typo" - I should have guessed as much.

You're a patent lawyer, where would your boundary be?

My apologies I haven't expressed this well. I'm talking about how you have expressed something and the name you have called it. I'll try again.

Wikipedia says if an invention has been described by "prior art" it is invalid, while you say that everything in the patent must be taught and it is called "anticipatory prior art". You took someone up on their use of the phrase "prior art" and said he had to refer to "anticipatory prior art". I think that wikipedia shows otherwise. I don't see anywhere on the wiki as it being described specifically as "anticipatory prior art". I don't see anywhere on the wiki that every claim must be described - it simply says it must sufficiently described.

"Not legally, no."

I'm not referring to a legal definition so it's a moot point bringing it up. And it was trivial and it's relationship to obviousness I was referring to. Not obviousness in and of itself. Your analogy is not suitable.

"But a machine patent is fine? What about a machine modeled in a virtual environment? What about an FPGA executing software? Where do you draw the boundary?"

In order:

Maybe. Maybe. Maybe - as long as the software you refer to is for reconfiguring the FPGA logic and not a higher level program - there is a pretty clear distinction. In regards to software that would be the boundary. So 99.9% of software would not be protected by patents.

"Well, technically, the US patent is valid anywhere outside the US."

You better tell Europe that, because they seem have firmly invalidated it.

I say it's valid only when other countries let it be valid - the US has no intrinsic jurisdiction overseas.

I'm simply pointing out the subtle differences in definitions. You're welcome to edit wikipedia to reflect a more exact or correct view in their opening sentence.

When writing on Slashdot - where we talk informally - I don't care about the process of disproving a patent - although I believe that due process is necessary for those that want to formally disprove it in court.

Yes, I can find it in prior art. Handheld devices (handheld device itself is a vague term which I don't like) have existed for thousands of years. I believe the process of going from large to small (and small may be handheld - but not always) is obvious across every field of human endeavour. So the element existed, and the process of transferring software to that smaller device was in full swing across the board. To me, it (a handheld device) does not pass go and it should never be allowed as part of a patent in the first place.

When I referred to trivial it is to bring attention to the fact that there is a curve that you can plot that shows the relationship between complexity and obviousness. Complex systems are in general much less obvious and can be a factor when proving non-obviousness (or the reverse - triviality can show obviousness).

I don't like that bogus patents are being granted that hinder innovation (and not just for Apple - I'm not an Apple hater). I don't like that patent examiners don't have the technical expertise to be able to trash bullshit patents or individual claims within patents. I really don't like that software patents are allowed at all.

This patent isn't valid in half of the world - where they recognise that Apple did not invent anything new.

"If an invention has been described in the prior art, a patent on that invention is not valid." vs "That doesn't mean they're anticipatory prior art, which is art that teaches everything in a patent claim."

To a lot of people here, Apple is just rehashing an already existing technology in it's entirety.

"because everything can be considered obvious in hindsight." But that doesn't mean it wasn't actually obvious in that previous point in time.

"So why should it matter whether it was on a handheld device? Because "handheld device" is in the claim. It's an element. That means you can't just ignore it, even if you think it's trivial, just like you can't ignore the inputs to a function. Go find that trivial element in the prior art. If it's truly trivial, then finding it should be easy, right?"

How does that defeat the non-obvious clause?

It is both trivial (to implement) and obvious (anything you can put on a large computer you can put on a small computer).