Traces of prior art An analysis of non-patent refe

TracesofPriorArtAnanalysisofnon-patentReferencesfoundinpatentdocumentsJulieCallaert1,BartVanLooy1,2,ArnoldVerbeek2,KoenraadDebackere1,2andBartThijs21K.U.Leuven,FaculteitETEW,ResearchDivisionINCENTIM,Naamsestraat69,3000Leuven2K.U.Leuven,SteunpuntO&OStatistieken,Dekenstraat2,3000LeuvenAbstractResearchintheareaofinnovationhaspointedouttherelevanceofconceivinginnovationasprocessesinwhichamultitudeofactorsandavarietyofinteractionsplayarole.Integrativenotionslike(national)innovationssystems,(techno-scientific)networks,orthetriplehelixmetaphor,havebeenwidelyacceptedasrelevanttograspthecomplexitiesentailed.Thisdevelopmenthighlightstheneedforindicatorsthatmirrorthedynamicsinvolved.Thiscontributionpresentsanin-depthexaminationoftheroleof“Non-Patentreferences”,foundinpatents.AfterexaminingtheoccurrenceofthesereferencesintheUSPTOandEPOpatentsystems,theprecisenatureofthesereferencesisdelineatedbymeansofasystematiccontentanalysisoftwosamplesofnon-patentreferences(n=10.000).Ourobservationsrevealtherelevanceof‘non-patentreferences’fordevelopingindicatorstodepicttheproximityoftechnologicalandscientificdevelopments.Applicationareas,limitationsanddirectionsforfutureresearcharediscussed.21.IntroductionOverthelastdecades,modelsoninnovationandinnovationpoliciesbecameincreasinglysystemic;innovationprocessesarenowlookeduponascontinuousflowsofinteractionsbetweencompanies,knowledgegeneratinginstituteslikeuniversitiesandresearchorganizationsandgovernmentalagenciesandinstitutions.Notionscapturingtheseinnovationdynamicsincludetheconceptof(national)innovationsystem(Freeman,1987,1994;Lundvall,1992;Nelson,1993;NelsonandRosenberg,1993;MoweryandNelson,1999;Dosi,2000),theTripleHelixmetaphor(LeydesdorffandEtzkowitz,1996,1998;EtzkowitzandLeydesdorff,1997,1998)andthescientificnetworkconcept(Pavitt,1997;Steinmuller,1994;Davidetal.,1997).Centraltothesemodelsisthemultiple-actorperspectiveandthenon-linearnatureofknowledgeflows.Forpractitionersandscienceandtechnology(S&T)policymakers,addressinginnovation-relatedconcernsrequiresindicatorsthatprovideinsightintothestructure,effectivenessanddynamicsofinnovationsystems.Forthesepurposes,scientificindicators,basedonpublicationoutput,havebeenpioneeredbydeSollaPrice(1963),Pritchard(1969),vanRaan(1988),Schubert,GlänzelandBraun(1989)amongstothers.Similardevelopmentsattheleveloftechnologytookplacewherebypatentinformationplaysacentralrole(deSollaPrice,1963,1965;Schmookler,1966;Callonetal.,1986;Griliches,1990;NarinandNoma,1985).Bibliometricindicatorspertainingtobothscientificandtechnologicalactivityarenowadaysgatheredonaregularbase,providingreferencepointsformappingandanalyzingactivitiesandtheiroutcomes(e.g.theEuropeanScienceandTechnologyIndicatorReports;theUSNSFReportsonScienceandEngineeringIndicators).Atthesametime,themajorityofindicatorsfocusexclusivelyoneitherscientificortechnologicalactivity.Toquantitativelygraspthefullbodyofwhatisreferredtoas“innovationsystem”,indicatorsreflectingrelationshipsbetweenthedifferentactivityrealmsseemhighlyrelevant.Inthiscontribution,weexaminetheusefulnessofinformationthatisprovidedinpatentdocumentsformappingandassessingconnectionsbetweentechnologicalactivitydevelopmentandscientificactivities.Patentsaredocumentsissuedandpublishedbyanauthorizedgovernmentalagency,grantingtheexclusiverighttotheapplicanttoproduceoruseaspecificnewdevice,apparatusorprocessforalimitedtimeperiod.Thispublishedinformationprovidesdifferentpossibilitiesforanalyzingthespecificenvironmentinwhichdevelopmenteffortstakeplace(Trippe,2003)andhencetoarriveatindicationsofscience-technologyexchanges.Firstly,interpersonalinventor-authorlinkagescanbeassessed:towhatextentareinventorsalsoactivelyengagedinscientificactivities,whichmakesthemappearasauthorswithinscientificpublications?Examplesofthisapproach,wherebyindividuals-crossingtheboundariesofscienceandtechnologyspheres-actastheunitofanalysis,canforinstancebefoundintheworkofNoyonsetal.(1994),PackerandWebster(1996),Balconietal.(2004)andMurray(2004).Asecondapproachconsistsoflookingatorganisationalentitiesthatappearbothasassigneeinpatentsandasaffiliationinpublications,shiftingthelevelofanalysistowardstheorganizationallevel.Thirdly,andsituatedatthesamelevelofanalysis,mappingandassessingco-patentingbehaviour–involvingcompaniesandresearchorganizationsincludinguniversities–canbeenvisaged,beitthatthelowlevelsofoccurrenceofco-patentingmightlimitwideapplicability(HicksandNarin,2004).Afinalwayof3revealingscience-technologyassociations,basedonpatentdocuments,consistsofexaminingthenatureoftheso-called‘priorart’foundinpublishedpatentdocuments;thefocusofthisarticle.Here,oneengagesinmappingandanalyzingthenatureandthesourcesofthecitationsprovidedbythepatentexaminerand/ortheinventorwhens/heconsidersthemrelevantforassessingtheinventionandtheclaimsitentails.Intermsofcontent,severaltypesofpriorartcanbedistinguished.Adistinctionisgenerallymadebetweenpatentreferencesandother–mostlyscientific-references1.WhereastheworkofJaffeetal.(2002)andothers(FlemingandSorenson,2001,Harhoffetal.,2003;Reit