Friday, November 13, 2009

What if Microsoft actually did patent Sudo? Is open source safe from patent lawsuits? Should it be?

The big controversy in the Unix/open source community this week arose from a post on Groklaw opining that Microsoft's "rights elevator" patent was a patent on the well-known and much beloved Unix "Sudo" command, which gives a user temporary "godlike" powers.

The horror in that tight-knit group, who make it their business to distrust anything that somes out of Redmond, was palpable. By the end of the week, some of the concern had faded, with many experts stating that there was nothing to fear -- that Sudo was still safe.

Notably, this this not the first time Microsoft has been accused of trying to monopolize something the open source community believes belongs to all -- See the post "Did Microsoft Just Patent Sudo? -- Holy crap I think they did" from 2007.

However, the controversy itself raises three important questions. (1) What would be the impact on the software industry in general and on the open source software industry in particular if Microsoft was able to successfully patent one of the core functions of Unix? (2) Could the open source community "work around" Microsoft in a way that would prevent a Microsoft lawsuit from gaining any traction? (3) What kind of monetary recovery could Microsoft expect if the patent held up and it could prove infringement?

And what effect might a SCOTUS Bilski decision that weakens software patents have on all this?

Frankly, I think if Microsoft is trying to strangle Linux by these tactics, they need better tactics. The open source community has proven that it can adapt very quickly and easily to virtually any attack. If Microsoft were to truly be able to patent something that was core to Unix, I think that there would be a lot of sturm und drang at the beginning, but that the industry would adapt in fairly short order -- and Microsoft would have just bought itself a bunch of new enemies.

If Microsoft were to start bringing lawsuits against open source vendors based on a patent like this, I think it would have pretty tough sledding. The open source industry -- led by companies like Red Hat and organizations like OIN -- has proven itself very adept at finding devastating prior art and making sure it is published. I believe that Microsoft would regret the day it tried to enforce such a patent, as there would be hundreds devoted to invalidating it.

In terms of any financial recovery -- assuming that the patent survived an invalidity attack and was found infringed -- Microsoft's ability to recover damages for any patent attacking open source software would be severely hampered by the very adaptability of open source. If there was an open source alternative to something like Sudo, even if it was not quite as good, Microsoft's recovery would be very limited since, if an infringer could get around the patent easily by switching to another open source alternative, the "reasonable royalty" for the Microsoft patent would be low.

And if the Bilski decision substantially weakens software patents, something like this "Sudo patent" might be thrown right out the window anyway.

Tuesday, November 10, 2009

Could Bilski Eliminate Damages for Method Claims Covered by Open Source Software? Maybe!

If you've found this article, I don't have to give you a primer on Bilski. By the time the SCOTUS is through, we should have a much better idea of the metes and bounds of what kinds of intangible intellectual property is patentable.

Given the tart comments by the various justices at the hearing on the viability of busines method claims -- from Justice Scalia's speculation about the availability of "horse whisperer" patents in the 1840's to Justice Breyer's wishing he could get a patent on his technique of teaching antitrust law to sleepy law students -- it appears that there will be a drastic narrowing of the scope of allowable business method patents that do not involve computers.

Now, in the courts, you don't really see the kind of business method patents you see in Bilski -- hedging risk in commodities trading. So, if the Bilski ruling does nothing more than strike down those types of patents, it may have little effect on big time patent litigation.

However, if some, like Red Hat, have their way and software patents are substantially weakened -- or even eliminated -- this could have a huge effect on the damages a plaintiff might be able to get for infringement of a method claim. Especially if that method is normally performed by software. Especially if that method can be performed by open source software.

For the most part, damages are awarded in patent cases based on what the jury finds to be a "reasonable royalty." That royalty is determined by looking at how much the infringer would have paid to license the plaintiff's patent if the parties had met and negotiated when the infringer started using the plaintiff's technology.

The more valuable the patented technology was to the infringer, the more he would have paid for the right to use it -- and the higher the "reasonable royalty."

If, however, the infringer didn't really need the plaintiff's technology or it didn't give him much economic benefit, the infringer would not have paid very much to license the patent and the "reasonable royalty" would be quite low.

One big reason that the infringer might not have put much value on the patented technology, is if he could have acquired that same -- or similar -- technology (which did not infringe) from some other source for less, or even for free; thsi is known as a "non-infringing alternative." If -- as in the Red Bend v. Google case recently filed -- the defendant could convincingly argue that he could replace the patented techology with free open source software, the "reasonable royalty" may well be minimal -- or even zero.

Now what does this have to do with Bilski?

If the Supreme Court weakens or elminates software patent protection, this will dramatically increase the viability of free open source software, as companies like Red Hat will largely be able to operate without the constant danger of being sued by a software patent plaintiff. The open source industry will no doubt take full advantage to increase their "market share." Tux the Linux Penguin will have a party!

So what happens to your damages case if you own a patent and are suing on a non-software method claim that can be implemented in software? And what if a similar function is performed in a non-infringing way in some piece of open source software that came out of the post-Bilski flood of open source? And what if the infringer could just pop that open source module into its product and perform the same function as claimed in your patent without losing a sale?

As Tux would say -- You're fragged and you go home with nothing.

Think this can't happen? Watch the skies for the Bilski decision and wait.

Tuesday, November 03, 2009

TSMC v. SMIC: When Chip Foundries Go to the Dark Side

Chinese chip foundry SMIC, who paid $175M a few years ago to settle a claim of trade secret theft from its Taiwanese rival TSMC still couldn't keep its hands out of the cookie jar and is now looking at over $1B in damages for continuing to steal its trade secrets.

SMIC evidently hung its hat on trying to prove that the technology it stole wasn't really a trade secret. For 63 out of the 65 claimed trade secrets, the jury didn't buy it.

More to come on how damages will be calculated in this case -- the jury will consider this phase of the case starting Thursday -- but the usual method is to look at how the defendant was unjustly enriched.

Big win for Keker. Big big loss for Wilson, Sonsini (for those of you keeping score on the law firm side), who really needs to find better clients.

More to come as we get more information.

Sunday, November 01, 2009

Nokia v. Apple -- FRAND and the Standards Wars

Erica Ogg over at CNET explained what is really going on on the business side in the big Nokia/Apple patent blowout -- The biggest handset maker, fallen on hard times, competes in the courtroom with the up-and-comer, seeking to add $6-12 per phone to Apple's costs

But, when the parties actually get to court, what is this case really going to be about and how will it get resolved? For the answer, you have to look at the history of how the imposition of technical standards throws a big monkey wrench into both patent litigation and antitrust law.

After Teddy Roosevelt got through with the "trusts" which strangled the economy in the late 19th Century, it became illegal for competitors to get together to fix the prices of the goods they sold -- in fact you can go to jail for it. In fact, the Justice Department doesn't like competitors getting together to do much of anything.

This applies equally to licensing of patents -- competitors can't get together to set the rates at which they will license their patents.

However, this posed a problem for some industries in which the products of all companies have to work in the same way -- all electronic appliances have to use the same shape plug, for example.

In the present high-tech world, where many different types of computers have to equally work over the Internet, having a set of common standards has been critical.

And in the communications field, where every phone has to be able to talk to every other phone, without standardization, you could only talk to phones made by the same company that made your phone -- totally unacceptable.

So, there are various US and international bodies who come up with standards (like 3G) that will enable everyone to talk to everyone else. These standards are constructed from the patents submitted by the various industry leaders.

Sometimes, these patents are licensed in one portfolio with the blessing of the antitrust authorities (like the well-known MPEG-LA group).

But in many cases, like this one, the various contributors to the standard claim that their patents are "essential" to the standard -- i.e., that if your product practices the standard, you must be infringing their patent.

The determination that patents are "essential," however, is not made by the standards group -- but is simply declared unilaterally, by each company. The deal the companies make with the standards group (to get the group to include their patents in the standard) is that they will license the patents to all comers at Fair, Reasonable And Non-Discriminatory rates (known as FRAND.

As with the decision as to whether a patent is "essential," the determination that the rate offered by the patentholder is "fair" and "reasonable," is made unilaterally and usually secretly.

So, what is really going on between Apple and Nokia in the lawsuit and how might it be resolved?

Personally, I disagree with Jason Schultz, quoted in Ogg's article, that Apple will try to show that Nokia does not own the patents or that the patents are invalid -- proving either is too hard and there are simply too many of them.

I believe that Apple's strategy will be two-fold: (1) To show that the Nokia patents asserted in the lawsuit are not really "essential" to the GSM, 3G or WiFi standards; and (2) to show that the rates demanded by Nokia are neither "fair" nor "reasonable" and that the patents are relatively insignificant to the standards. Under a recent Federal appeals court decision (Lucent v. Gateway), the lower courts are instructed to limit a plaintiff's recovery of damages to the value the patent actually has.

It's obvious that Apple doesn't have as high an opinion of these patents as Nokia does.

Showing that Nokia may not have real confidence in either the value of its patents or its ability to compete with Apple in the marketplace, it is also trying to recover damages for Apple's appropriation of Nokia's market share.