Saturday, April 26, 2014

Copying is all that Samsung needs to do in order to work around four of Apple's five patents-in-suit

Apple's "copying" accusations against Samsung have over the years degenerated into a red herring. In 2011 I truly admired the coherent and compelling fashion in which Apple's lawyers told a complex copycat story in their original complaint against Samsung over a diversity of intellectual property rights. And I thought Samsung responded to it in a very smart way, countering Apple's copying mantra with a competition mantra:

"The Samsung Defendants admit that they have not ceased competing with Apple notwithstanding Apple's efforts to avoid such competition."

The original lawsuit and certain related lawsuits in other jurisdictions undoubtedly had a positive effect on product differentiation at a visual, physical level: Samsung's Galaxy products are much more distinguishable now from the iPhone and iPad than they used to be. I believe Samsung still owes Apple money for past design patent infringement, as I also pointed out earlier this month when I outlined the terms on which I believe they should settle.

But more than three years after the original complaint (which went to trial in 2012), Apple's enforcement efforts have more to do with avoiding competition than with discouraging copying. As counterintuitive as it may be, the patent claims Apple selected for the ongoing California lawsuit actually encourage copying because copying would result in workarounds -- non-infringing alternatives -- for at least four of the five patents-in-suit (and the fifth one is just slide-to-unlock, which can be worked around in numerous other ways; Apple is not even accusing such devices as the Note 2 of infringement of that patent). In one case (the '647 "quick links" patent), Samsung and Google would have to copy certain Android code into applications, and in three other cases, Samsung and Google would be in the clear with products resembling more closely the way Apple implements the related features.

That fact contrasts with Apple's continuing smear campaign against Samsung's alleged copying, including a questionable Earth Day ad that some people on the Internet speculated might have been run purposely during the ongoing trial. Things like that don't get funnier or more accurate with the passage of time. Quite the contrary.

Apple has achieved visual differentiation with its enforcement efforts, and that's positive. But it has been unable to ensure functional differentiation apart from rubberbanding. As a result, its market share continues to erode. I was impressed by the iPhone sales figures Apple reported this week (though the iPad did not do too wel). However, even at that rate, Apple continues to lose market share. Tomi Ahonen (with whom I don't always agree) pointed out that "[y]es, iPhone sales are up from this same period one year ago but that is up only 20% while the industry grew neary twice as much at over 35% in the same period" and he rightly asks where "THAT story" is. I also tend to agree with BGC analyst Colin Gibbs who sees "Apple as a provider of premium priced electronics, a lucrative market but one that may not sustain its current market valuation [...] in the years ahead".

Maybe Apple believes it can slow down the erosion of its market share by making Samsung out to be a copyist, just to make its own customers feel better for buying the original thing. This may work with a certain, impressionable audience. Those who look more closely at what kinds of intellectual property rights Apple is actually asserting can easily see that Apple v. Samsung II is just about an attempt to win some infringement findings regardless of whether there is actual copying involved -- and even if a "win" would actually encourage copying.

There are two kinds of copying here that can result in non-infringing alternatives (should Samsung's current technologies be found to infringe, despite reasonably good non-infringement and invalidity theories): copying Apple and copying Google.

As for the first category, Apple was not off to a great start in this new trial when the first major controversy related not to whether Samsung infringed Apple's patents but to whether Apple itself even practiced certain of the asserted patent claims. Apple would have liked to argue (which I thought was a reasonable plan) with respect to the three of its patents that its current products practice different claims than the asserted ones, but still certain claims of those patents. But as far as the actually-asserted claims of those three patents are concerned, Apple just argued that it practiced them in the past (partly not even in products that were ever sold, and Samsung disputed everything). This means that Samsung would avoid infringing those three actually-asserted claims by implementing the related features just the way Apple's more recent products do. Copying as a recipe for non-infringement.

If Samsung did that, Apple could assert other claims of the same patents, but it would have to bring a whole new lawsuit for that purpose, which would take time. In that case, Apple would have to assert broader claims, and broader claims are more susceptible to invalidation challenges, which is why in at least one case (the '414 synchronization patent) Judge Koh had indicated to Apple that a broader originally-asserted claim would have been invalidated at the summary judgment stage.

A lot of people out there may have misconceptions about the significance of an infringement finding. If an infringement finding is a thing of the past and the defendant can simply work around the asserted claim going forward, then the only remedy is damages for past infringement -- but there is no strategic leverage to gain from past-infringement damages in a dispute between the global market leader and the original category creator, as opposed to a lawsuit brought by a patent assertion entity against an operating company where the only objective is to extract a settlement.

There's a difference between symbolic and strategic wins, and Apple appears to have prioritized the probability of symbolic wins over the possibility of having serious business impact.

Software patent infringement allegations are rarely due to copying (in which case one can assert copyright, which is what Oracle, unlike Apple, is doing against Google) but to independent authorship that may or may not be inspired by seeing a certain feature somewhere. And in connection with operating system patents, a workaround often consists in simply copying over certain program code from the operating system level to the application level.

Yesterday, when everyone thought Apple and Samsung were done presenting evidence at this trial, the United States Court of Appeals for the Federal Circuit handed down a ruling that contains some really good stuff for Apple in its dispute with Motorola -- and for other defendants against FRAND-pledged standard-essential patents, a fact that I'm very happy about -- but also defanged and devalued the '647 "quick links" patent. The appeals court affirmed Judge Posner's claim construction of the '647 patent (almost the only thing, other than his "no injunction over FRAND SEPs" determination, that was affirmed). Based on that narrower interpretation of the patent, it's possible to work around it by simply copying and compiling the structure detection and link generation code from the operating system (the Android Linkify library) to the applications that make use of it. This is the recipe for a workaround that Judge Posner himself provided two years ago (based on his claim construction, as opposed to the broader one that Judge Koh allowed Apple to base its infringement argument on):

"[...] inventing around the '647 patent by reprogramming Motorola's smartphones to avoid at least one claim limitation, for example by simply creating copies of the code that performs structure detection and linking for each particular program rather than using a common code module for all programs, because if there is no common code there is no 'analyzer server,' as required by the patent claim."

(first emphasis added, second in original)

Against that background, Apple only asked for 60 cents per device from Motorola (which still seemed too much for Judge Posner, though not for the appeals court), less than one-twentieth of its per-unit damages demand from Samsung in the current case, in which Apple pursued a broader claim construction that enabled it to claim ownership of the whole feature as opposed to a particular internal architecture that can be avoided.

The Federal Circuit stresses that, as Judge Posner recognized, "the plain meaning of 'server,' when viewed from the perspective of a person of ordinary skill in the art, entails a client-server relationship", and points to Fig. 1 of the patent specification, which shows an analyzer server as a separate module (item 165, "program of the present invention") from the "application" (item 167). The Federal Circuit declined to adopt Apple's proposal to render the term "server" meaningless.

Defanged: the patent can be worked around now, so any finding of past infringement is not going to hurt Samsung and Google in the future. Devalued: on this basis, Apple expert Dr. Hauser's feature-focused conjoint analysis was not the way to arrive at an accurate valuation of the '647 patent, as Apple's own 60-cent claim in the Motorola case shows.

Prior to the Federal Circuit claim construction ruling, I thought it might take a long time to find out how strong that patent is. Now its limits are pretty clear. In this case, copying (Google's own code) is the solution. A solution that Judge Posner outlined but didn't (and couldn't) patent.

There will be some more expert testimony on the '647 patent on Monday, and it's not impossible that Apple will be able to present a good infringement theory even based on the narrower interpretation of the patent. But past infringement findings are not the name of the strategy game.

The whole Apple v. Samsung II story is not just a disappointing sequel. It increasingly looks like a modern-day version of The Emperor's New Clothes. A lot of people may think Samsung is doomed should the jury (despite pretty good defenses) identify past infringements because they don't understand that the name of the game is to assert patent claims that can't be worked around without serious commercial implications. But reasonably informed people won't be impressed.

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: