Sunday, April 6, 2014

'Teaching away' is why Apple's rubber-banding patent has far more merit than slide-to-unlock

Numerous patents are litigated the world over without the public taking notice, and the vast majority of them would be too esoteric for a large audience. But when Apple's patents on touchscreen features everyone knows go to trial in the U.S., where jury trials on patents are in no small part a show business, there is a great deal of interest. And there is public debate. With this post I want to contribute to the quality of the debate by sharing information on what a couple of particularly famous Apple touchscreen patents are about and, even more so, what kinds of ideas patent law does or does not reward with 20-year monopoly rights.

Apple's four most famous touchscreen software patents are rubber-banding (the "overscroll bounce", asserted in the first Apple v. Samsung case in California), slide-to-unlock (at issue in the ongoing Apple v. Samsung II trial in California, pinch-to-zoom (at issue in the first California case), and the "Steve Jobs patent" on touchscreen heuristics. Two of those patents have been asserted (and maybe even granted) only in the U.S.: pinch-to-zoom and the "Steve Jobs patent". That is one reason to focus on rubber-banding and slide-to-unlock now. There are even bigger reasons. Apple is already enforcing, through a U.S. import ban, the "Steve Jobs patent" against Samsung, but in vain: the ITC, a U.S. government agency with quasijudicial competencies, cleared Samsung's workaround and consumers don't even notice a thing. That makes the "Steve Jobs patent" a blunt sword at best. And pinch-to-zoom, which was deemed valid by the first California jury (because the jury didn't even care to evaluate all of the prior art), was rejected in its entirety by the Central Reexamination Division of the USPTO (Apple appealed to the Patent Trial and Appeal Board).

Of the two remaining patents, rubber-banding is the much more defensible and meritorious one in terms of representing an inventive step -- even if not rocket science, it is inventive per se -- relative to the start of the art. It's a useful exercise, or case study, to compare the two patents and, at a high level, their validity challenges. Since Apple has asserted both patents internationally, I will make a clear distinction between general principles of patent law that are not specific to any jurisdiction, U.S.-only parameters, and special European requirements. These patents are easy enough to explain that all of this is doable without making things too complicated.

Noise vs. signal: U.S. jury trials involve evidence that is unrelated to the actual merits of a patent

I won't go into detail here on why it's a fundamentally flawed idea to put the technical aspects of patent cases before juries. I'll just point to what a judge and a software company CEO said. Judge Posner suggested that these could be handled better by USPTO officials for the most part. Red Hat's CEO said: "I have a degree in computer science, and these things are so far over my head, and it takes weeks of Ph.D.s to figure out exactly what this is, and then you're going to adjudicate it in front of a jury that really is not technically savvy. Do you think this patent is valid or not? It's both expensive and just a bizarre way to adjudicate it." But that bizarre way works better for Apple -- and your average Eastern District of Texas troll -- than more rational ways.

In the rest of the world, patent cases are typically adjudicated by judges who hear such cases all the time and acquire some understanding of technical issues. In some cases there are even technically-trained judges. For example, each five-member panel of a nullity (invalidation) senate of the Federal Patent Court of Germany has three technical judges (a majority). Those judges have degrees in fields like physics and engineering.

The way the process works in the U.S. is, however, separate from the question of whether a patent is valid and infringed under U.S. law. Once a case goes to an appeals court, there won't be any jury. To the extent that trial judges overrule juries, they also have to focus strictly on the actual merits.

Don't be confused by what you see at, or hear about, a trial. Large parts of what juries and reporters get to see at a trial like Apple v. Samsung are absolutely irrelevant to the questions of validity and infringement. The primary purpose is to tell a certain story to the jury. The legal justification for this is not that certain types of evidence, testimony and lawyer argument have any bearing on validity and infringement. Instead, the justification -- or basically, the excuse -- is that a jury also has to determine willfulness and damages.

Apple Insider's Daniel Eran Dilger yesterday published "Copy Cat docs [shown by Apple at the trial] as evidence Samsung pilfered [the] iPhone unlock". But all that those docs show is the perfectly lawful process of benchmarking: Samsung was looking at the way the iPhone did it (you don't need a license from Apple to look at its products), compared them to its own, conducted an analysis of what could enhance customer satisfaction, and then improved its products. This is done countless times each day by some company somewhere on the world. Ford will certainly look at a new Toyota (and vice versa), and if it finds that a competitor has an advantage in one area, it will look for ways to improve its own technology. This has nothing to do with whether Apple met the requirements for a 20-year monopoly in terms of an inventive step relative to the start of the art, nothing to do with how to interpret the asserted claim, and nothing to do with whether Apple's slide-to-unlock patent claims are broad enough to capture Samsung's mechanisms (at least for Samsung's newer mechanisms, such as the circular unlock in the Galaxy Nexus, I believe the patent is not broad enough).

Apple's counsel also asked Greg Christie, named as an inventor on slide-to-unlock, whether he was still married to the same woman. He answered yes. The transparent purpose was to portray him as a reliable, faithful, trusthworthy kind of person and to make female jurors like him. The marital status of an inventor is no less related to the issues of validity and infringement of a patent than documents that merely prove legitimate benchmarking. All of this is meant to influence the jury, hoping that it will base its verdict on likes and dislikes instead of the actually relevant parts. I can perfectly understand that Samsung is not prepared to enter into a global settlement with Apple, which has yet to score any meaningful win outside the U.S. (besides rubber-banding), based on sometimes highly questionable jury verdicts handed down in a jurisdiction in which a patent validity or infringement decision may hinge on someone's marital status...

Unfortunately, the appeal of certain evidence and testimony to juries, reporters (several of whom said on Friday that they were bored by expert testimony on technical issues) and the general public is, as a rule of thumb, reversely proportional to its actual relevance. But that need not stop you and me from focusing on what really matters.

In a U.S. copyright case, actual copying doesn't have to be proven if it's clear that something was not independently created just based on its nature and stature. But if copying is proven, then there's a lower similarity standard. However, Apple alleges copying without asserting copyright. It's asserting patents, and independent invention is not a defense to patent infringement anyway (unfortunately).

Where I attend patent trials in person, the focus is (unless licensing issues, such as related to industry standards, come into play) exclusively on technical validity and infringement questions and no one ever asked anyone about his marital status. That experience helps to filter out meaningless noise from what is presented at a U.S. jury trial.

An invention must be (among other things) novel and non-obvious over the prior art

In this field of technology, innovation is incremental. Everything builds on something else and combines previously-known elements in some way. The first question (apart from whether an invention is made in a patent-eligible field) is then whether a given combination of previously-known elements is novel, i.e., whether this particular combination has not been published or patented before. If it's not, then the patent is invalid for "anticipation"; end of story in that case. If it is, then the next question to answer is whether this combination was obvious at the relevant time (in which case the patent is invalid) or non-obvious, in which case there is an inventive step relative to the prior art (in Europe this is sometimes referred to as an "inventive contribution to the state of the art").

There's a key difference between whether someone technically could combine certain elements and whether someone looking for a solution to a problem likely would. Is there a teaching, suggestion, or motivation that leads to a combination? If someone combines two patented inventions that share more keywords than two patent documents usually do, then we're probably talking about an obvious combination. If, however, someone combines elements from two distinct fields of technology, then someone else who is an expert in only one of those two fields would probably not have done so.

The best case: to come up with something the prior art teaches away from -- such as rubber-banding

What helps a great deal is if an inventor came up with something that, at the time the invention was made, was flying in the face of conventional wisdom. If basically all medical literature, including patents, counsels against giving substance A to influenza patients but someone finds out that a certain limited dose of substance A, if combined with a certain dose of substance B, accelerates recovery, then the prior art "teaches away" from the new invention.

Think of Columbus (a discoverer, not an inventor) who went in a direction in which most people believed he was doomed to fall off the earth.

Conversely, one could also say that the new patent teaches away from the prior art. It works either way. But in patent law you usually hear the former, not the latter.

And now, finally, we get to the specifics of rubber-banding and slide-to-unlock. Rubber-banding is a revolutionary teach-away case while slide-to-unlock is merely evolutionary.

About three years ago I described rubber-banding in detail, with videos, and explained why I considered it a strong patent, even though it's easy for a software developer to implement. You can find a pseudocode example in this article by a non-governmental organization opposing software patents in Europe.

Rubber-banding solves the problem that an abrupt end to a scrolling operation instinctively makes a user press harder because it appears that a device is not responding to a gesture. We'll get to the question of whether this is a technical problem (key in Europe, where the patent is subject to an ongoing opposition proceeding) later. But from an end user's perspective, rubber-banding is a solution to a problem. Having a "glow" at one edge of the screen is better than getting no visual feedback at all, but it's too subtle. (I know that some people find rubber-banding excessive on iOS, but I liked it when Android provided it and I'd like to see it come back as a result of license deals.)

The existence of the problem was obvious. I already identified the problem on some Nokia Symbian phones with traditional scrolling mechanisms. It's not a problem on a large monitor because you have a large slider bar and see easily when you hit the end of the list, and the connection between you and the screen content is less direct if you use a mouse. Even on larger touchscreens it's not a problem if you move the scroll bar with your fingers rather than the content of a window. But on small touchscreens it's a real issue.

The purpose of patent law -- and this is something very important that some of Apple's supporters get wrong in the debate that is raging, not with respect to rubber-banding but in connection with countless other Apple patents -- is not to grant anyone a monopoly on the right to solve a problem but on a particular (novel and inventive) solution.

Apple's overscroll bounce, while trivial at the implementation level, is inventive in my view because the prior art -- all other scrolling techniques -- used to teach the opposite. It was always clear that scrolling ends when you hit the top or bottom of a window. This was abrupt, and everyone considered it logical and appropriate. In fact, just like older medical literature might have counseled against a certain substance in connection with a particular illness, it was considered a bug to overscroll. Apple turned a bug into a great feature. Like Columbus, it went in a direction in which others thought you'd fall off the earth.

It was absolutely brilliant for Apple to depart -- to teach away -- from conventional wisdom. They figured out that you actually do want to overscroll on a touchscreen because a finger movement that goes significantly (can't be too much anyway on a small screen) beyond the end of a document is much less frustrating than instinctively pressing harder because the device appears to be unresponsive. You then need a bounce-back, which is fun to look at. So, from a user experience point of view, it's a superior solution on the bottom line.

I've drawn up a simple graphic to show the fundamental departure from conventional window scrolling techniques that rubber-banding constituted (click on the image to enlarge; this post continues below the image):

It's another question whether it was necessary for the world to grant Apple a 20-year monopoly on rubber-banding. If Apple had not come up with rubber-banding, it would probably have taken some more time before someone else would have. Traditional phone companies like Nokia were simply not unconventional enough. Their approach to product development appeared very bureaucratic and conservative at the time. But even though there is always the risk of underestimating the merits of an invention with the benefit of 20/20 hindsight, I guess that there's at least a significant probability that rubber-banding would have been created by someone else in the meantime -- and almost certainly ahead of the end of the 20-year term.

Even for rubber-banding there is prior art that comes relatively close in terms of navigation through items that involves a bounceback. Obviously Apple didn't invent the bounceback. It's all about the application of the bounceback effect to document scrolling. And other document scrolling that has a bounceback (which was discussed at some hearings and trials I attended as well as in U.S. court and USPTO documents) does not specifically teach the post-bounceback alignment of the edge of the document with the edge of the window that is key to rubber-banding. Other mechanisms have different purposes, such as centering around a given document, than an alignment of edges. Even if other bounceback mechanisms may in some scenarios have a similar visual effect as rubberbanding, they aren't really identical.

There is still some risk to Apple that rubberbanding might be invalidated in Europe. But so far it's Apple's only truly successful patent in more than four years of litigation against Android device makers. It's the only patent so far that can't be worked around without losing the entire benefit of the invention. In that post from 2011 I linked to further above I described it as Apple's "favorite make-Android-awkward patent". And now, almost three years and hundreds of millions of dollars in legal fees later, it has turned out to be the only touchscreen-related feature Apple has proven it "owns".

Slide-to-unlock: there's a difference between a breakthrough invention and a particularly great execution

For slide-to-unlock it would be impossible to draw up a graphic like the one further above that shows how rubber-banding was a total departure from conventional wisdom. That's because sliders have always (even prior to touchscreen phones) been the usual way to unlock mobile devices. Another traditional way is to press a button, but this entails the risk of "butt dialing" (unintended operations as a result of inadvertently pressing a button). Avoiding "butt dialing" was key, and it again shows how companies like Nokia lacked inspiration for a long time (they were destined, and deserved, to fail). If someone had asked me back then as part of a patent damages survey how much I'd have been willing to pay to solve that problem, I'd have have given an answer in the $10 range. But Apple-style slide-to-unlock is just one of various ways to avoid butt dialing. I've seen a diversity of unlocking mechanisms on my six different Samsung Android phones over the years and none of them ever had a "butt dialing" problem, regardless of its degree of similarity (which was never really high) to Apple's slide-to-unlock mechanism.

If you use a slider to lock or unlock something, you have to move the movable part of the slider along the slider bar from one end to the other. It's always been like that. Even on relatively old phones it was possible to avoid butt dialing with a (physical) slider.

The idea of interpreting a sliding gesture on a touchscreen as an "unlock" command by the user was very obvious. The Neonode N1m offered this before the iPhone:

In Europe, ten (of ten) judges deemed Apple's patent invalid because it was obvious over the Neonode. The eligibility of a Swedish phone as prior art in the U.S. under applicable rules (it apparently wouldn't be a problem with respect to the most recently-filed patents) is in dispute. But regardless of the purely legal question of eligibility, this shows that it was, unlike rubber-banding, not a stroke of genius to implement on a touchscreen something similar to a physical slider.

Apple's '721 patent covers certain ways of unlocking a device by moving an unlock image across the screen. It's a slide-to-unlock image, not general slide-to-unlock, patent. That, of course, limits its scope and its commercial value. But it didn't help Apple defend the validity of this patent wherever it came to judgment because it's just too simple an idea to move an image on a touchscreen.

There's another piece of prior art worth showing here, and it was also discussed in Europe but the Neonode made the invalidity findings here so easy that Plaisant didn't have to be reached. This is a 1991 video and paper on "Touchscreen Toggle design" by (then Microsoft-employed) researchers Catherine Plaisant and Daniel Wallace. The video is available on YouTube, and starting at around 3:00 (thanks to Twitter user Greg Forrest, @gg4rest) you can see touchscreen-based on/off switches:

At one or more of the German trials, Apple's counsel argued that Plaisant is different because an on/off switch is not identical to the process of unlocking. There is, but it says something that Apple needs to resort to arguments of that kind to defend the patent. If the prior art contains a solution for an on/off switch, it's not really inventive to apply it to an unlocking process.

A patent holder doesn't own an end user's mind, eyes, or fingers

In my post on the very limited impact that Apple's patent assertions against Android have had so far, I explained that "[i]f you want to obtain a patent on an aspect of a user interface, you are limited by the fact that you can't monopolize end users' body parts". The same applies to a user's mind, of course. What I explained there is that in the user interface context you can patent techniques that display something and ways to interpret or respond to a user action, but you must show an inventive step in one or more of those areas and can't claim a two-finger pinching gesture per se, much less the mental processes of a user.

The fact that a user instinctively presses harder if a touchscreen scrolling operation ends abruptly also relates to the user's mind. That's why it's not in the patent claims. But as a simple graphic showed further above, rubber-banding was at least a fundamental departure from decades of conventional window scrolling operations.

At the Federal Patent Court of Germany, where a trial resulting in the invalidation of the European slide-to-unlock patent took place exactly one year before the testimony by Greg Christie in San Jose related to the U.S. '721 slide-to-unlock patent, Apple's counsel (from an absolutely top-notch firm, Bardehle Pagenberg, which I mentioned in a recent post) faced an uphill battle that he just couldn't win. The court made clear at the outset of the nullity trial that it considered the patent invalid, as it had previously communicated to the parties in the form of a preliminary ruling. After a few hours the court said that it thought the patent failed to meet both the European technicity requirement and the non-obviousness requirement. It was clear that the position on technicity wouldn't change, but since the appeals court might disagree, the court and the parties (Apple, Samsung, Google/Motorola) still discussed the non-obviousness part in order to have a complete record.

In the technicity context, the court said that user interfaces can have patent-eligible elements, but if it all just comes down to saying that it's easier for a user to figure out how a slider works than how other unlocking mechanisms do, that doesn't make it a technical invention. In that regard, Apple's counsel argued, for example, that a slider is so intuitive that it involves only a user's cerebellum, not their cerebrum. That one drew laughter from Google's counsel, not because it was wrong or stupid (it wasn't) but because you usually don't hear this kind of mental argument in a discussion of whether a patent covers a technical invention.

Another problem with the cerebrum-cerebellum argument is that Apple actually uses its slide-to-unlock patent (in the U.S. as well as in Europe) against all sorts of Android unlocking mechanisms that don't sport an iOS-style slider bar but require the user to think a little more.

This contrasts nicely with rubber-banding. For rubber-banding, one can also make an argument relating to subconscious user benefits and it's valid -- but at the level at which patentability must be determined, one can argue that everyone did scrolling differently, even on small touchscreens, until Apple came up with a teach-away idea.

Samsung's internal benchmarking documents still don't render slide-to-unlock inventive over the prior art, while rubber-banding was a brilliant idea in and of itself. Again, at the policy level one can debate whether a 20-year monopoly is justified even for rubber-banding (and whether Apple and its rivals should have had to spend hundreds of millions of dollars in legal fees when the only touchscreen feature Apple has proven to own is this one), but for slide-to-unlock, ten European judges have already found that Apple did not deserve a monopoly on it for even one day. In the U.S., an inventor's marital status may tip the scales, however. And that would not be the right outcome.

If you'd like to be updated on the smartphone patent disputes and other intellectual property matters I cover, please subscribe to my RSS feed (in the right-hand column) and/or follow me on Twitter @FOSSpatents and Google+.

Share with other professionals via LinkedIn: