Hacker News Reader

Comment by jasonkester | Hacker News Reader

Mr. Forstall said he invented a patent for double-tapping on Web pages because as he had been using a prototype of the iPhone to surf the web, he realized he was spending a lot of time pinching and zooming the page to fit text perfectly on the screen.

"I realized I have this incredibly powerful device, why can't it figure out the right size for me?" he said. So, he challenged his team to make the software automatically size the text into the center of the screen when he double-tapped around a webpage."

This is why we're in such a sad state with regard to software patents. This guy genuinely believes that he "invented" something. And that it should be patentable.

Of course, anybody who has actually built anything knows that what he actually did was "decide how something should work". You do this dozens of times when putting out a new product, and it's not in any way a big deal. Certainly not something you should call "inventing", and absolutely not something that you should consider patenting.

It's just one of thousands of design decisions you make. It's just sad to watch people who don't understand that making things worse for everybody.

View on HN

I think double-tap-to-zoom is a perfect example of something that should be patentable. It's a huge improvement over existing systems that often used a zoom box you would position with cursor keys. To work properly it has to take into account the major design elements on a web page and make intelligent choices based on the user's tap target. Web browsing before double-tap-to-zoom and multi-touch zoom was a nightmare. These two features reinvented web browsing on small displays. This was clearly one of the killer features of the original iPhone.

> To work properly it has to take into account the major design elements on a web page and make intelligent choices based on the user's tap target.

And that is what should be patentable, if anything at all should be patentable. The algorithm to account for design elements and make intelligent choices.

Not "tap twice to zoom".

In theory, patents exist to cover ideas which are hard to think up, but obvious once you've seen them. Double-tap fits this description perfectly.

I am against patents in general. But this is perfectly in line with past hardware patents. For example compare to 1950s radios where the volume control was also the on/off switch. Pull it out, and the radio was on, push it in, it was off, and it remembered the volume that you wanted. That button design was patented.

I don't see a difference in kind on that patent and the double-click patent.

I'd wager anybody with a little work, when told "Don't add another interaction point to this radio" or "Try to find some way of cutting down on distinct inputs" would stumble upon the radio knob solution given a little time.

This isn't art--even a cursory exploration of the solution space with vague constraints should get you there.

I've been trying to think up a practical way to use that very test as the barometer for patentability. Announce the problem that has been solved, and let the public submit solutions. If any of them are close to what is in the prospective patent, the patent shouldn't be granted.

It's tough though, because often the inventive thing is framing the problem correctly, or even identifying that there is a problem to be solved here. By assuming that, you're in danger of a hindsight bias.

Patents were never intended to grant monopolies over a given problem. If you're solving a problem that no one else has solved before, people will try to solve it once they use your product. I don't think granting monopolies over a problem space helps society. Quite the contrary.

> Patents were never intended to grant monopolies over a given problem.

Well I didn't say that they were... but I'm not sure I agree with your statement anyway. Do you have any evidence that that isn't what they were intended for?

I don't have any evidence. It just doesn't make sense to grant monopolies on a problem. It would harm innovation rather than help it. We'd have one medicine to treat each ailment, for instance.

Like the article that was recently posted about smart headlights letting drivers see through rain. Once you read the headline, you instantly know "camera(s), projectors, and motion-tracking algorithms" without having to click through to the article. Maybe the specific combination of motion tracking algorithms could be patentable for a short time, but not in such a way that an independent implementation necessarily infringes the patent.

That may be. But my point is that the standard that is being applied here is exactly what has been applied in hardware for decades, and what has been held up before legislators as exemplars of what patents are supposed to be for.

Incidentally I think that there was some ingenuity there. When Jacob Rabinow created the pull-button, he had been asked to improve a push-button tuner. (A review of the source informs me that when you pulled it had the station, not the volume. My bad. I have not seen one of these radios in decades...) His reason for switching is that he knew the problems with existing push-button tuners, and could design a better pull-button tuner. The manufacturer he invented it for did not want to accept it because nobody else was building them.

Incidentally the inventor of that was Jacob Rabinow. Take a glance at http://museum.nist.gov/exhibits/rabinow/patents.html to see all of his patents. Given how many inventions he had, he clearly did not spend long on each. But his claim would e that without the patent system he would have had no incentive to invent them, and would have worked on other stuff.

I'm not defending this either way. Merely stating the official view of what the patent system is supposed to have been for. (And my personal feelings are the other way.)

The patent system is supposed to encourage the progress of science and useful arts, not to reward inventors. While that is a good thing, it's not something to sacrifice progress for.

We're heading towards a world where all the money is getting funneled towards patent lawsuits and away from building products. This is not a good thing.

The patent system is supposed to encourage the progress of science and useful arts THROUGH rewarding inventors.

It says "may" not "shall" though, so to the extent it's not doing the job, it can (and should) be reworked.

That's well and good too--excepting the fact that the free market already does this, and at the rate the patent system/legal system move, the deck is utterly stacked against the pioneers unless they've got gigantic money behind them.

And since software has such low barriers to entry, this really does not help progress.

Prior art: double clicking the application bar with _ O X on it for a web browser. Double tap is as old as the GUI.

Actually, double-click was invented at Apple in 1980 or 1981, some ten years after the WIMP paradigm was invented at Xerox, and 18 years after Ivan Sutherland invented a sort of graphical user interface for Sketchpad, which Alan Kay credits as "the invention of the GUI".

See http://www.folklore.org/StoryView.py?project=Macintosh&s..., in the comments: "I remember when Larry was doing that testing. One of the things he tested was the three-button mouse, which is what Xerox had used. He found out that inexperienced computer users (the market that Lisa was aimed at, believe it or not) got too confused -- they couldn't remember what each of the buttons did. Same problem with a two-button mouse. So the one-button mouse was born. Of course, that meant that we needed double-click and triple-click, but that's another story..."

http://www.folklore.org/StoryView.py?project=Macintosh&s... shows screenshots of the UI that had the first double-click interface, in 1980 or 1981.

That doesn't mean that Apple deserves a patent for every double-finger-twitch action in perpetuity, or that double-click doesn't logically follow from the reduction in mouse button count.

> That doesn't mean that Apple deserves a patent for every double-finger-twitch action in perpetuity,

And they don't, so why bring it up.

> or that double-click doesn't logically follow from the reduction in mouse button count.

Lots of things "logically follow", but only a few make sense. Figuring that out seems to actually be fairly difficult as the state of the art shows.

Of course.

Double-tap fits this description perfectly.

Excuse me, what on earth is non-obvious about that?

When enumerating the alternatives to the pinch-gesture (the awkwardness of which becomes obvious within the first few days of use) then you end up with a rather short list of 10-12 items - and that's counting rather esoteric ones like a physical jog-dial.

Apple chose double-tap, just like anyone else in their right mind (other than Sony) would.

If that's not textbook example for a trivial "invention" then I don't know what is.

The parent was pointing out the conflation of action (double-tap to fit, overloading volume control with on/off features) and mechanism (fitting algorithm, physical design of said button). The former is not patentable, or rather should not be, while the latter might be.

It's the difference between going to the patent office with an idea for "an engine that runs on steam" and actually designing a steam engine. See how much broader in scope the first one is?

I have serious trouble figuring out any possible difficulties in implementing this. Text is already in a bounding box. Zoom to the width of that box, and give it a little margin to make it look pretty.

It's not like zoom hadn't been implemented, or they had to reflow anything.

Patents aren't about doing things that are hard. They're about doing things that are new.

Patents aren't about doing something new. They are non obvious ideas - read the definition please. Amazon 1 click patent was an example of idiodicy of "something new". This blasé attitude to patents is extremely dangerous. It stifles innovation and hurts society - the exact opposite of why patents were created. Imagine if i patented RGB 128, 179, 253 because I couldn't find that value hadn't been used before. Should this be patentable? What about patenting a piece of timber 3.914 ft x 1.234 ft? This is where Apple deserve a really hard kick in the balls. They patented something so generic and obvious that it is almost impossible not to breach it. You've heard the saying "standing on the shoulders of giants" - what it means is that technology has enabled a bunch of potential inventions. Patterns of these combinations emerge naturally - sine pick up on it quicker than others... But such incremental steps should not be patentable because someone else would have easily come up with the same thing. Now if you tell me that someone invented a new DSP, or a new compression algorithm that was genuinely different, then that is a good example. Ask the question: would someone have easily come up with it, roughly at the same time ... And... Does it help of harm society to grant this patent... Then and only then should a patent bd granted. - posted from my iPhone

Actually, part of the criteria for getting a patent (at least in the states) is that the invention is non-obvious to those in the trade. So patents kind of are about doing things that are hard.

And you could argue that zooming in on an HTML bounding box would be obvious to an engineer developing web browsers.

As Rob Pike has said about his patents, everything is obvious in retrospect.

In any case, if it's so obvious, where's the prior art? It's not as if web browsers haven't been around for nearly 20 years.

It seems to me that Apple has justification to defend wholesale copying of the myriad of little design decisions that they made that are synergistic. It's clear that they spent a huge amount of effort and cost doing this, and were able to succeed in a way that many had tried previously and failed.

The legal ways that one can defend such synergistic design work is limited. One common way, for better or worse, is to patent some of the individual elements to throw gum in the works of anyone trying to make derivatives of your work.

Web browsers on tiny devices that have the processing power to do a scaled zoom in a reasonable period of time have most certainly not been around for 20 years.

Not that I was making any comment on patentability.

But if the process isn't difficult, and we can assume that people individually have used zoom to zoom in on text, then the extent of the innovation is having it as a preset.

In theory, you're wrong (and you certainly should be).

In practice, I'd take it a step further: Patents aren't about doing things that are hard. They're about doing things that are new... to the patent office.

My old Symbian phone in 2006 had an intelligent zoom, either in the built-in browser or in Opera Mobile. May have just been the latter. You didn't tap to activate it, you moved a little box, but it smart-sized the column you selected to fill the screen.

So is changing the mechanism to a double-tap when implementing it on a device with a touchscreen worth patenting?

And that's the thing about software. It's so easy to make, compared to designing/manufacturing something physical, that there are thousands of programs out there that you've probably never used, doing who knows what, and yet people are expected to know if something's novel and patentable?

I'm sure you also believe the UX in any given video game should be patentable, and video games need to license each other for the effect of tapping buttons here or yanking joysticks there in order to even build their own.

I couldn't possibly disagree more.

Technology has and continues to advance at ridiculous pace. Many software developments are dependent on hardware. Until the hardware exists no one can even try to solve the problem. I don't want a company which has first access to hardware to automatically get all the software patents that derive from it.

Google is hard at work on their glasses. Lots of companies are. The hardware isn't there yet. We're getting there though. Eventually someone is going to have hardware that enables all kinds of cool things to be done that have never been done before. Our culture / society / economy doesn't need to compound their reward of being first with a 20 year exclusive.

"it has to take into account the major design elements on a web page and make intelligent choices based on the user's tap target"

That's fine, so patent the _specific algorithm_ used to make that decision. Implementations are fine to patent, IMHO. Broad concepts, such as "it should zoom in when I tap twice" are not.

There are two kinds of inventions.

(1) Providing a solution to a known problem.

(2) Figuring out that something is a problem and providing a solution.

I'd argue that (2) is more valuable than (1) as there are likely to be other solutions in the case of (1) so the payoff is relatively small.

Figuring out that a problem exists and solving it is far more valuable.

Yes, I know that most of us work on (1), but surely our self-interest is irrlevant.

You've made (2) a superset of (1) here.

The fact is, he identified a problem and then handed it off to others to solve. Any of us could've identified that problem, and most of us could've implemented the solution after a bit of work.

Especially in a group environment (dev listserv, public forums, etc.) we expect to have users that can identify problems and others that can solve them. Adding patents and claiming "Invention!" for incremental improvements is harming progress.

> Any of us could've identified that problem,

Not so fast. None of you did, despite the fact that mobile devices aren't new.

When I first saw tetris, I said "wow, I could have invented that, I was just looking at asymmetric block figures". However, I didn't.

Almost everything is trivial in hindsight.

>Especially in a group environment (dev listserv, public forums, etc.) we expect to have users that can identify problems and others that can solve them.

Are you claiming that Apple execs can't invent anything because they have too many indirect subordinates? Or that groups only produce improvements that are too trivial to warrant protection?

First, about me not solving that problem.

I wasn't in industry when that issue came up, and I wasn't employed by Apple. So, it really isn't fair to hold me responsible for a problem I didn't have. I've only recently begun carrying a mobile device of any sort, furthermore.

There are infinitely many problems that I haven't solved nor, I reckon, have you. I will not argue, though, that we automatically are not capable of solving them, as we have not encountered them and have no reason to solve them.

Given the constraints, given the hardware, given the opportunity and incentive, we certainly could come up with something similar. Just because we did not does not mean we could not.

This logic puts us in the unfortunate opportunity of saying "Well, the first person to run into the problem, to them goes the spoils!"

That's rubbish, and heavily encourages either wasteful flailing about in the problem space wasting resources, or worse, explicitly avoiding exploring even the same areas as others because you'd hate to be a few minutes late.

We need to be free to attempt solutions to problems regardless of who else has worked on them and to what success.

Second, the point I was making was not that Apple execs can't invent anything (indeed, I believe I set the bar fairly low for novelty!) or that groups only find trivial progress.

The point was that in a community we expect some users to find problems and others to find solutions--and these may be different folks, spread far in time and space. You made some statement seeming to imply that one must both discover and solve a problem, and that is what I took issue with.

> I wasn't in industry when that issue came up,

Not relevant.

> and I wasn't employed by Apple.

Huh? Lots of people not employed by apple invent things.

> So, it really isn't fair to hold me responsible for a problem I didn't have.

"You" wasn't directed at just you. It was directed at the thousands of people who walked right by that problem.

Like I said, hindsight, just like I had with tetris.

> Given the constraints, given the hardware, given the opportunity and incentive,

Why do you think that being given something is relevant?

Some folks have "time machines" (Bill Joy reference). They get to invent in that context.

> Well, the first person to run into the problem, to them goes the spoils!"

That's pretty much how it works. Folks with domain knowledge have an advantage. So do people who think creatively. And so on.

> heavily encourages either wasteful flailing about in the problem space wasting resources

No, it doesn't.

> We need to be free to attempt solutions to problems regardless of who else has worked on them and to what success.

You are "free to attempt solutions". You just have to succeed first.

> You made some statement seeming to imply that one must both discover and solve a problem

No I didn't. I said that some inventions address known problems while others both discover the problem and a solution.

I pointed out that lots of us work on the first kind of invention, but the second kind is just as valid.

Maybe we should define "invention" in terms of "design decision". An invention should be something that requires no fewer than three hundred design decisions to create.

They're already unable to handle other subjective requirements like "non trivial" and still struggle with the slightly objective "prior art". I don't think adding new subjective definitions would help.

In my opinion the biggest problem with patents is that it's impossible to put the theory in practice no matter how great the theory is. On paper, patenting only non-trivial ideas sounds good. On practice that rule simply doesn't exist because it's impossible to define objectively.

> You do this dozens of times when putting out a new product, and it's not in any way a big deal.

The sum of these design decisions make the product, they are the most important aspect the project.

A Patent for "tapping on something to make it do something".

Who owns the patents for tool use and fire brandishing?

Completely agree, the implementation should be protected by copyright but not the idea itself.

So if you had the idea to create an open roof "sunroof" in a car, the way the roof opens mechanically should be protected by copyright, not the fact that you are putting a rectangular hole above drivers' heads for more light and air? Could the shade slide be protected too?

Good point except that I'm strictly speaking about Software. I didn't explicitly state it because I thought it was implied.

>Of course, anybody who has actually built anything knows that what he actually did was "decide how something should work". You do this dozens of times when putting out a new product, and it's not in any way a big deal.

Sounds like a very useful feature to me, and something that competitors or previous products did not had.

And "deciding how something should work" sounds like a perfectly good description of an _invention_.

Anyone that has implemented something like that knows implementation and execution is the really hard part.

Implementation has the chance of being the "hard part" only _after_ the idea has been thought.

What are you smoking? If the implementation is too hard, the idea was worthless.

True but there are multiple ways of doing something. So after thinking about an idea, finding the easiest way to implement it is also a task in itself.

I'm not smoking anything, thank you very much for your polite inquiry.

How you came to the conclusion that "If the implementation is too hard, the idea was worthless", I can't even begin to think.

Difficulty of implementation and worthiness of idea are totally orthogonal.

There are tons of very important ideas that had a very difficult implementation. E.g the idea behind the atomic bomb was very valuable despite the fact that the implementation took some of the top minds, several years and millions of dollars.

Is "the progress of science and useful arts" promoted by granting monopolies over such things? That is, would Apple have chosen not to create the iPhone without patents like this? I think not.

>That is, would Apple have chosen not to create the iPhone without patents like this? I think not.

If you take only one patent as an example, maybe not. But if Apple couldn't secure ANY patent on the iPhone, and any competitor was free to copy it as he liked, well, not only the might have not produced it, but the very industry might not even exist and have the scope it has today.

Aren't the iPhone and the Samsung Galaxy S3 functionally equivalent? For the sake of argument, haven't competitors succeeded in "copying" the iPhone to their hearts' content? Note that trade dress is a class of IP on its own that definitely has helped Apple, since competitors can't create products and try to pass them off as the iPhone itself.

The iPhone is a very successful product, and it owes very little of that success to patents.

There were few-to-no software patents until the 80s. In that period there was rapid progress, and many new industries; some say much faster progress than since.

It's true that when some areas offer artificial monopolies and others don't, investors like the former. But extending the domain of monopolies isn't the only way to debias that.

What are you talking about?

Apple makes the Mac Pro, the iMac, Macbook Pro, Macbook Air, which are all generic computers, which anyone can (and does) copy. Do you think they would not do it since they can't patent a laptop computer, that's rectangular? Or an all-in-one computer? Or a tower computer?

Reply on news.ycombinator.com