用户行为，交互，用户战略

Formalization,UserStrategyandInteractionDesign:Users’BehaviourwithDiscourseTaggingSemanticsBertrandSereno,1SimonBuckinghamShum2andEnricoMotta2CentreforAdvancedLearningTechnologies,INSEADBoulevarddeConstance,[e-mail]te,TheOpenUniversityWaltonHall,[e-mail]/[e-mail]or/owner(s).’susabilityandthetagstructurescreated,contributingtoourunderstandingoftheimplicationsofaddingstructuretocollaborativetagging.CategoriesandSubjectDescriptorsH.1.2[InformationSystems]:User/MachineSystems—Humaninformationprocessing;H.4.m[InformationSystems]:MiscellaneousKeywordsSocialtagging,Sensemaking,DiscourseRelations,Semantics,Argumentation,Usability,PragmaticWeb1INTRODUCTIONOurcommunities,local,nationalandinternational,areconfrontedbyproblemsthatarecomplexduetothechangingenvironment,incompleteorambiguousinformation,andstakeholderswithdifferentperspectives.Suchdomainsincludestrategicplanninginbusiness,governmentpolicyformulation,time-pressuredmissionoperations,andalmostallscholarlyresearch.Thesensemakingactivitythatthesecontextsdemand[25],requiresanalyststoconstructplausiblenarrativesthatframetheproblem,accountfortheavailableevidence,andmotivateaction,enabling“anopenlyreflexiveforuminwhichcommunitiesofknowingexplicitlytalkabouttheirunderstandings”[2].Progressismadebymakingmovesthatexpressandcontestinterpretationsoftheworld,althoughthesedifferentcontextsclearlyhaveverydifferentgenresofdiscourseandcriteriaforacceptance.Thefocusofthispaperisontheworkofacademicresearchers,butwearguethatrelatedworkshowsthatanapproachgroundedindiscourserelationsisapplicabletoabroaderrangeofapplications.Itisestablishedfromcorpusanalysesthatwhenresearcherspublishtheirinterpretationsoftheideas,opinions,claimsorrebuttalsintheliterature,theyaredrawingonarepertoireofwellunderstoodmoves,contributingtoanextendeddiscourse[21].AlthoughtheInternetisacceleratingthepaceofexchanges,scholarlyandscientificdiscoursestillproceedsintheshadowoftheprintingpress,withexchangesnowdisseminatedasdigitalprose.Whileinformationretrievalandtextanalysistechnologieshelptoinfercertainkindsofstructurewithinandbetweenpapers,ourresearchiscomplementary,exploringa‘network-native’paradigminwhichthekeyclaimsmadebyanauthor(andtheinterpretationsmadebytheirreaders)arepublishedasexplicitnewconnectionstotheliterature.Theresearchquestiondrivingourresearchis:canwemodelthediscoursestructureswefindinresearchcommunitiesasexplicitstructures,andifso,whatsupporttoolscanweprovidetoconstruct,navigate,andinterrogatesuchstructures?SuchapproachestoknowledgepublishingandnegotiationontheWebwillideallybebothquicklylearnable,yetsufficientlyexpressivetopermitresearcherstomakeimportantscholarlymoves,andassisttheminmakingsenseoftheemergentstructuresatscale.Whatisespeciallyinterestingaboutscholarlydiscourseisthefactthat“thetruth”or“thesignificance”ofanyclaimisopentocontest.Whilethismaybeextremeinthecaseofphilosophyandthehumanities,itisself-evidentlyalsothecaseevenincomputingandthehardsciences.Thereisnosinglereadingofapaper;interpretationsmaydiffersignificantlybetweenreadersandauthors(hencetheneedforpeerreview);readersbringtheirownuniqueperspectivetoapaper,seeingnewconnectionsthattheauthormayneverhaveintended.Seeingoldthingsinnewwaysistheessenceofcreativity.Thisistheorientationwebringtoharnessingthepowerofsocialtagging,butaugmentedwithdiscoursesemantics,aswestrivetocreateeffectiveinfrastructureforscholarstoexpress—andcontest—claimstoknowledge.2SCHOLARLYTAGGING2.1QuestionsnosearchenginecananswerConsiderthefollowingquestionsthatintereststudentsandresearchers,butwhichneitherInternetsearchenginesnordomain-specificdigitallibrariescanassistinanswering:Whatdatarefutesthishypothesis?Aretheredifferentschoolsofthoughtinthisfield?IsthereananalogybetweenthisprocessinfieldsXandY?Whydo