« Biocide | Main | Appeal of Appeal »

June 11, 2009

Top Dogs

The most valuable patents are naturally the ones covering the most profitable technologies. It would be a facile conclusion that the most valuable patents might also be the most litigated. Take patented drugs as an example. Because only drug companies provoke litigation over drug patents, the number of litigations for such are low. Yet blockbuster patented drugs are worth hundreds of billions in sales, above and beyond what generics later prove to be. So, while the stakes over drug patents are quite high, the number of litigations for any single drug patent is quite low. But take a look at the roster of litigations filed over time, and see that a plurality involve drugs or biotechnology. So, with drugs, lots of litigations, owing to the value of patents generally, but no particular patent family is multiplicatively litigated.

Then there's the wild west of invention: software. Becoming a most valuable software patent almost always means widespread adoption. Software innovations spread like wildfire. Unlike drugs, valuable software patents are likely to be heavily litigated, in terms of number of assertions, because many companies infringe. Nature of the business.

Number bums John R. Allison, Mark A. Lemley, and Joshua Walker, with more number-crunching than analytic firepower, have published a biased and misleading survey, that unsurprisingly comes to a preordained conclusion, as the article title pronounces: "Trolls on Top?" This has nothing to do with trolls favoring the missionary position, at least literally. But figuratively, it's the entire voyeuristic spectacle. Which brings one to wonder - what is one to think of nincompoops, like these three, that are, basically, troll fuckers? One may wonder if they are filled with envy. After all, them trolls, with their patent booty, are making out like bandits. Pardon the pun (knowing some won't). After all, being in academia, with its publish or perish ethos, is only lucrative if spilling sensational beans. This "study" being a case in point.

The conclusion of the study is that software patents owned by trolls are most heavily litigated. Bring out the violins. And the shotguns.

We identify the patents litigated most frequently between 2000 and 2007, and compare those patents to a control set of patents that have been litigated only once in that period. The results are startling. The most litigated patents are far more likely to be software and telecommunications patents, not mechanical or other types of patents. They are significantly different from once-litigated patents in ways that signal their value up front. And they are disproportionately owned by non-practicing entities (aka trolls).

"The results are startling." Mock surprise.

The flaws in analyses of Trolls on Top? overwhelm whatever residual value remains. The survey is of patents, not patent families. The charts are of a meaningless comparison: once litigated patents versus most litigated patents. Any decent statistician can tell you that looking at the tails of a bell curve tells you nothing. And there is no assessment of value by either damage award or of profits protected by patenting.

As one might suspect, the whole point of the exercise to give the impression that trolls are a clear and present danger. There are at least a dozen different varieties of trolls, according to the three.

As noted in Part I, we divide the patent plaintiffs in our data sets into twelve different categories.

One of the biggest policy debates in patent reform has been over the nature and extent of "patent trolls. "

Actually, only anti-patent troll fuckers consider trolls worth more than passing mention, as the ownership of intellectual property is at best tangential to informing what patent public policy should be.

The first thing that stands out is the powerful evidence that the most-litigated patents have different, clearly identifiable characteristics that distinguish them from once-litigated patents (and distinguish them even more dramatically from ordinary, unlitigated patents). Notably, the characteristics that distinguish the most-litigated patents from other patents are also the ones that researchers have long used to identify the most valuable patents: more claims, more prior art citations, more forward citations, more assignments between issue and litigation, and larger numbers of continuing applications. A reasonable conclusion, therefore, is that the most litigated patents are also the most valuable patents.

Not so, at all. There is no established correlation between most valuable patents are most litigated patents. Practically tautologically, the most valuable patents are those covering the most profitable technologies. Given the profits of patented drugs, the patents in that area surely qualify as highly valuable, even as their litigation assertion numbers are quite low.

Posted by Patent Hawk at June 11, 2009 5:19 PM | Patents In Business


Hey Hawk,

Wuz up with the colorful language? This used to be a family channel.

Your anger is showing.

Extreme emotion is exactly what the Lemleys of this world want. They push your emotional buttons and then jump for joy when you react as they had wanted.

Time to "step back" and cool it. Time to realize that a corporation (or "company") is a legal fiction. Companies don't invent things or make and distribute things. People do.

Essentially all inventors in essentially all "companies" are "trolls" because they, the human inventors, don't immediately manufacture and distribute to the public that which they had invented.

These non-practicing humans (a.k.a. inventors who, by the way, are demoted by psycho-linguistic trickery into being called "entities") sell their legal rights in the invention (often by operation of an employment contract) to the legal fiction known as the corporate "entity" or "company". Sale of the invention from the non-practicing first "entities" (inventors) to the non-existing but legally-hypothecated second "entities" (companies) cannot logically occur until the moment after the invention was made (conceived and reduced constructively or actually into practice). Before that magical moment there was no invention to sell.

Even at the magical moment of inter-entity transfer of ownership of "the invention", the acquiring corporate entity is not "practicing" the "innovation" by immediately mass producing it and disseminating it to the starved-for-technology masses. So it too, for at least that magical moment, is a non-practicing entity (NPE) involved in the evil practice of buying and selling mere rights in not-yet-practiced inventions. Oh the horror of it all. we suddenly realize that ALL companies are "trolls". The only ones who are not trolls are the word-weaving Lemleys of this world. They spin words into pure gold. (You can free yourself of their magic by repeating this secret magical word ... but don't tell anyone ... Rumpleskillstink.)

Posted by: step back at June 12, 2009 2:46 AM


not all "software patents" (if there is such beast) are the same

There are patents like yours for GUI interface, for example, where infringement is clear and obvious to anybody looking at the computer screenshot

There are other patents on some key algorithms which can be hidden inside a huge pile of compiled code and will require expensive reverse-engineering to detect infringement
Even worse, they can be inmplemented in hardware and the cost of reverse-engineering can be prohibitive for individuals or even small company's patent-holders

For each highly litigated and profitable patent there are at least 10 other patents vanishing in obscurity (while the industry is making a big buck from the unlicensed use of invention)

Posted by: angry dude at June 12, 2009 6:36 AM

From the article's "classification" of patents:

"(2) Pure software: An invention consisting only of data processing—all claim elements in the patent consist of data processing. However, we do include within this definition a patent on data processing in which there is a trivial non-data processing element such as a generic input, output, or storage element clearly not intended to represent any novel technical advance. This category is a subset of Software.
(7) Imaging: A process, machine, or product in which the invention or a critical part thereof involves the creation and/or processing of images for various purposes. The imaging may be analog or digital. The majority of imaging patents have medical uses, but some serve other purposes such as security or meteorology."

Obviously, Lemkey and the rest of academic punks know jack shit about the subject

Aren't all images digitized and processed using algorithms implemented in software ?

What is Adobe Photoshop ?

Posted by: angry dude at June 12, 2009 7:02 AM

I'd like to know the source of funding for Mr. Lemley and Co.

Who is sponsoring this shit ?

Posted by: angry dude at June 12, 2009 7:12 AM

step back,

"Wuz up with the colorful language? This used to be a family channel." Your anger is showing.

Your projection of my emotional state is anti-empathic. You confuse anger with a choice of colorful nomenclature, to emphasize the thrust of the three bozos’ argument. As Elvis Costello once wrote/sang: “I used to be disgusted, but now I’m just amused.”

You’re right about this entry being not suitable for reading before 10 pm. But the only thing really naughty about words is when they are used to manipulate by distortion, so the real cursing belongs to Lemley et al.

I liked your logic train that ended with “ALL companies are ‘trolls,’” even if the whole subject of non-practicing entities is blather by would-be IP thieves and their syncophants, with an ax to grind, for which Lemley et al are shilling.

Posted by: Patent Hawk at June 13, 2009 11:08 AM

The link you give up to the paper by the non-practicing trio (NPT) didn't work (well at least not for my browser). However, if one goes to one of the NTP professors' web site and searches their list of publications for the word "troll", a downloadable version of the paper is found.

Posted by: step back at June 14, 2009 3:32 AM

meant to say, "up top"

Posted by: step back at June 14, 2009 3:34 AM

Them NPT professors is real good at high falluten math and percentages. For example, in their Fig. 2 graph of percentages, the sum in the "computer" industry sector of "once-litigated" versus "most-litigated" patents exceeds 100%. Only the most brilliant and loyal subjects of the emperor can make percentages like these add up to more than 100%. So these NPT professors is obviously a using mathematics so sophisticated that only PhD's in the math arts can understand that their high falluten stuff. Wow.

Also, their definitions of "electronics" and of "pure software" are quite fascinating. For example, they define the "semiconductor" industry sector as including "inventions of any kind intended to advance the state of the art in ... designing ... semiconductor chips." So that means that all software inventions (i.e. CAD) "intended" to be used for designing chips are in the semiconductor industry category. Interesting. You don't get scholarship of this caliber too often.

Posted by: step back at June 14, 2009 4:09 AM

Anyone who continues to use the word troll in a serious context (and I'm talking about guys like Lemley, who, incidentally, needs to cut back on the Twinkies) should be roundly shunned by the patent bar.

Posted by: Just sayin' at June 14, 2009 5:38 PM