[snlacks]

You make a valid point, it is for a different purpose. I think many people aren't arguing the actual law, on Hacker News, I feel as if the commenters are expressing their expert opinions on what it should be. I'm not an expert, but I sure do have opinions. That purpose basically reduces to "on a computer, for computer reasons." If you put a bathroom slide lock on a chicken coop to keep the chickens from getting out, that's a different purpose, but it shouldn't be a new patent.

That being said, if you had to use a newly invented screw or slide to get that chicken coop mod to work, then that probably should be patentable. Similarly, on a phone, if you invented a new means of translating that motion into a slide, or other actual invention, than that probably should be valid. Using existing touch drivers to recognize an obvious real world motion, probably not.

[axman6]

You're looking at the process backwards. Just because latches exist does not mean that, at the time the patent was filed, it was obvious to use a latch analog to unlock a digital device. A person skilled in the art would NOT be lead directly to that solution, but one which is already know: a pin, password, fingerprint, keycard, heck even facial or voice recognition. The test is not "did latches already exist" but "is it obvious to use them in the method claimed, as the time of application", and I would strongly argue that it is not, without some evidence to support the contrary.

I say this as someone who has worked as a patent examiner and understand the law (not the US law exactly, but they're quite similar world wide).