[nl]

I posit that android would not be on the market in the touch-screen form it is now, if Apple hadn't made its inventions public due to the patent process.

Seriously?

I can see you might say Android would not have shipped a touch-screen version if the iPhone hadn't first, but I can't see the argument that public patents helped with that much at all.

Once you've seen a multi-touch touchscreen work, it isn't hard to work out how to make a capacitive touchscreen do it.

[gjm11]

> Once you've seen a multi-touch touchscreen work, it isn't hard to work out how to make a capacitive touchscreen do it.

I can't comment on whatever patents Apple may have, because I haven't read them, but the above is surely an oversimplification.

Let me describe how a typical capacitive touch sensor works. It's basically a grid of wires, one set running horizontally and the other vertically. (They aren't really wires, and the difference matters a lot for multiple reasons, but we can ignore it here.) For each wire, you can measure (kinda) whether a user's finger is near to it, and how near if so. (Because the finger will couple to it capacitively, which changes how the wire responds when you wiggle its voltage up and down.) So you use the horizontal wires to measure where the finger is vertically, and the vertical wires to measure where it is horizontally.

So far, so good. Now you want to make it work with multiple fingers. Problem 1: if you have a finger at position (3,3) and another at (10,10), what you see is (crudely) lots of coupling to wires 3 and 10 on each axis. But that is also what you'd see from a finger at (3,10) and another at (10,3). How do you tell the difference? Problem 2: if you have a finger at (3,3) and another at (3.5,10) then you can see the y-coordinates clearly enough but the signals on the x-measuring wires are going to be hard to disentangle.

These are not trivial problems to solve. (I know of a few approaches. I don't know what Apple actually do. I have worked for a company that makes capacitive position sensors, but not for Apple. Nothing I've said here is anyone's trade secret.) The more-obvious things you might do to try to solve them all have substantial difficulties. And of course these aren't the only difficulties in making a multi-touch capacitive touchscreen. (Making a decent capacitive touchscreen at all isn't trivial, though it's pretty much a solved problem nowadays.)

So this is exactly the sort of situation in which, in an ideal world with a sensible patent system, patents might be the Right Thing: there's a tricky technical problem, it's solvable but you can't just Do The Obvious Thing and have it work, so having someone find a good solution, publish it, but still be able to get commercial advantage from it seems like a pretty good outcome for everyone. (Note: this does not constitute an endorsement of how patents currently are in the real world.)