Changes
Jump to navigation
Jump to search
Strandburg (2006) - Law And The Science Of Networks (view source)
Revision as of 12:38, 3 April 2013
, 12:38, 3 April 2013no edit summary
*This page is referenced in the [[Patent Thicket Literature Review]]
*This page is listed on the [[PTLR Core Papers]] page
==Abstract==
The network may be the metaphor of the present era. A network, consisting of “nodes” and “links,” may be a group of individuals linked by friendship; a group of computers linked by network cables; a system of roads or airline flights -- or another of a virtually limitless variety of systems of connected “things.” The past few years have seen an explosion of interest in “network science,” which seeks to move beyond metaphor to analysis in fields from physics to sociology. Network science highlights the role of relationship patterns in determining collective behavior. It underscores and begins to address the difficulty of predicting collective behavior from individual interactions. This Article seeks first to describe how network science can provide new conceptual and empirical approaches to legal questions because of its focus on analyzing the effects of patterns of relationships on collective behavior.
The Article then illustrates the network approach by describing a study of the network created by patents and the citations between them. Burgeoning patenting has raised concerns about patent quality, reflected in proposed legislation and in renewed Supreme Court attention to patent law. The network approach allows us to get behind the increasing numbers and investigate the patterns of relationships between patented technologies. We can then distinguish between faster technological progress, increasing breadth of patented technologies, and a lower patentability standard as possible explanations for increased patenting. We find that, since the late 1980s, the disparity in likelihood of citation between the most “citable” and least “citable” patents has grown, suggesting that the least citable patents may represent increasingly trivial inventions. One possible explanation of this increasing stratification is increasing reliance by the Federal Circuit Court of Appeals on the widely criticized “motivation or suggestion to combine” test for nonobviousness, which is at issue in the case of KSR v. Teleflex, currently pending at the Supreme Court. We also discuss how network science may be employed to address other issues of patent law.
==Review==
===Definition of Patent Thicket===
:''"On the one hand, there is the fear of a patent "thicket" in which the transaction costs associated with obtaining the necessary patent licenses to do something of practical usefulness become so high as to undermine the social value of the patents."''
===Sample===
*16 million citations by over 2 million patents issued by the USPTO from 1975-1999 from the USPTO (and application counts since 1800).
**Dataset constructed by Hall, Jaffe, and Trajtenberg;
**LEXIS searches of Federal Circuit for cases involving patent obviousness claims referring to the "suggestion, teaching, or motivation to combine" test, 1977-2006.
===Results===
*Authors define a "citability function" Ak(k)~kα, where k is the number of times a patent has been cited, and A(k) is the probability that the patent will be cited again.
**Alpha is a measure of the extent to which highly cited patents are preferred.
**Alpha is estimated to be 1.19 from 1975 onward.
**Variations in alpha by industry were not found.
:''"However, the fact that α>1 in the patent network still suggests that citations are highly concentrated -- there is a highly unequal distribution of patent 'citability.' " ''
:''"Since citability is likely related to a patent’s technical importance, the increasing stratification suggests that the patentability may have decreased, resulting in the issuance of a larger fraction of more trivial patents."''
*Al(l)~l-β, where l is the patent’s age in patent numbers, and A(l) is the probability that the patent will be cited again.
**Beta is a measure of the extent to which highly cited patents are preferred;
**Beta is estimated to be 1.6 from 1975 onward implying a long, slow decay. Citability peaks at relatively young age
** Al(l) was found to be relatively independent of k.
===Social Welfare Consequences===
*It is argued that the new standard of nonobviousness is not rigorous enough, and thus, it leads to too many patents being granted.
:''"The proliferation of patents results in a situation in which many commercial products require the use of more than one -- and in some cases hundreds of – patented technologies...there is the fear of a patent 'thicket' in which the transaction costs associated with obtaining the necessary patent licenses to do something of practical usefulness become so high as to undermine the social value of the patents."''
===Policy Advocated===
*Network science should be further developed to evaluate and determine the best practices for the USPTO.
