« Here Boy | Main | Nortel Phones It In »

March 17, 2008

Drama

Professor Dennis Crouch: "Over the past two decades, the number of patents being litigated has risen dramatically." Crouch then shows a graph that's clear as mud: one axis labeled "Patent Count;" the other "Year Complaint Was Filed." One has no way of knowing, either by graph or explanation, whether "patent count" represents number of complaints; or number of litigated patents total, irrespective of number of complaints. Regardless what "patent count" means, most damning is failure to normalize: to take into account the number of patents granted.

Most meaningful would be the percentage of patents granted litigated each year, if there were any meaning to be had. Keep in mind that patents aren't necessarily litigated out of the chute; it may be years from grant before a patent is infringed, and litigation initiated.

Regardless of past numbers, in the litigation department, expect drama on drama, if you care to call it that. Since the courts in 2006 made declaratory judgment actions easy, litigation is now the only avenue for opening licensing negotiations. Expect litigation numbers to now dance to that new tune.

An interesting statistic would be patent success percentage: what percent of patents were successfully litigated, either to settlement or judgment. That would convey a sense of the real junk patent rate.

Ignorance of proper statistical application is widespread and lamentable. The most important aspects of appropriate statistical application are: 1) using normalized numbers; that is, in the proper context; and 2) knowing when the right statistics tell a meaningful story, as contrasted to presenting graphical drama when the reasons behind the numbers are the real revelation.

Example - The government and the press report GDP (gross domestic product) changes to demonstrate the country's economic growth. But that GDP, according to the Economist, equals a Grossly Distorted Picture, because it fails to normalize for population changes. GDP per person would be a more proper output figure. The difference -

GDP growth figures flatter America's relative performance, because its population is rising much faster, by 1% a year, thanks to immigration and a higher birth rate. In contrast, the number of Japanese citizens has been shrinking since 2005. Once you take account of this, Japan's GDP per head increased at an annual rate of 2.1% in the five years to 2007, slightly faster than America's 1.9% and much better than Germany's 1.4%. In other words, contrary to the popular pessimism about Japan's economy, it has actually enjoyed the biggest gain in average income among the big three rich economies.

Posted by Patent Hawk at March 17, 2008 4:22 PM | Litigation

Comments

Gary, Thanks for the link. However, I do believe that my graph is fairly easy to understand -- The number of patents associated with complaints filed in a given year.

I have a problem with normalization without also providing the underlying data. In your example, the graph is designed to impart the feeling that Britian's economy is the strongest in the world. It is not.

Posted by: Dennis Crouch at March 18, 2008 1:31 AM

Dennis, I would say that is brash academia at its worst - a patent professor making allegations about the relative strengths of world economies. On what underlying basis? [This conduct depicts one of the main problems in academia - professors who cannot grasp how little they may know, and whose so-called "knowledge" is not seasoned with humbling salt of practice.]

Contrary to your assertion, the graph is in no way "designed to impart the feeling that Britian's economy is the strongest in the world", and no thinking human could look at it and ever say it was. Reread the graph.

David Testardi

Posted by: anonymous at March 18, 2008 2:22 AM

Gary & David,

Thanks for noting this oft overlooked detail about the population bomb.

"Growth" looks oh so self complimentary when it's on academic paper rather than the bathroom scale.

Posted by: step back at March 18, 2008 3:07 AM

Dennis,

Thanks for the clarification. Your graph shows: number of patents asserted in litigation by year.

My point regarding normalization stands. More important is the point of storytelling. "Over the past two decades, the number of patents being litigated has risen dramatically." That could indicate a problem, but the story is misleadingly incomplete. Are the number of patents asserted, accounting for number of patents issued, rising?; in other words, patents asserted per issue rising? Maybe assertions, normalized by issuances, dropped over those years. What would that tell you?

Just as change in GDP per person is more a statement of economic growth from a human perspective; a big picture view of productivity; so percent of patents asserted would convey some statement, albeit exactly what is debatable; whereas number of patents litigated says practically nothing. It’s like stating cost or revenue figures over the years without accounting for inflation.

As to the Economist graph, for 2003-2007, Britain had the fastest growing economy per person. To say, “the graph is designed to impart the feeling that Britain’s economy is the strongest in the world” conveys a misunderstanding of what GDP growth is. “Economic strength” is a woolly term: what is that supposed to mean?

Posted by: Patent Hawk at March 18, 2008 12:06 PM

"Just as change in GDP per person is more a statement of economic growth from a human perspective; a big picture view of productivity; so percent of patents asserted would convey some statement, albeit exactly what is debatable; whereas number of patents litigated says practically nothing. It’s like stating cost or revenue figures over the years without accounting for inflation."

Patenthawk, please peruse your paragraphs for proper punctation prior to posting (i.e., please learn how to use semicolons).

Posted by: Grammar Guy at March 18, 2008 11:04 PM

Grammer Guy,

Thanks for the hitting me upside the head about semicolons. You are right, in that semicolons are not to be used between sentence parts of unequal grammatical rank, such as a main clause and a subordinate clause.

Posted by: Patent Hawk at March 18, 2008 11:35 PM

"punctation"

aS USUAL, THOSE WHO JUDGE WILL FALL BY THEIR OWN CONDEMNATION... ;-) (pROPER USE OF SEMICOLON IN EMOTICON, IMPROPER USE OF sHIFT kEY.)

lIGHTEN UP, gRAMM*R GUY

Posted by: nipra ANONYMOUS at March 19, 2008 5:13 AM