Tải bản đầy đủ

Digital libraries and crowdsourcing


TableofContents
Cover
Title
Copyright
Preface
Introduction
1AConceptualIntroductiontotheConceptofCrowdsourcinginLibraries:ANew
Paradigm?
1.1.Arapidlygrowingeconomicmodel
1.2.Origin,definitionandscopeofcrowdsourcing
1.3.Historicalchronologyofcrowdsourcing
1.4.Philosophicalandpoliticalcontroversies
1.5.Economic,sociologicalandlegalconsequences
1.6.Managerial,libraryscienceandtechnologicalconsequences
2OverviewofSeveralCrowdsourcingProjectsAppliedtotheDigitizationofLibraries
2.1.Puttingcontentonlineandparticipativecuration:theOxford’sGreatWarArchive
andEuropeana1914–1918
2.2.Digitizationondemandintheformofcrowdfundingappliedtodigitallibraries:
theEuropeaneBooksonDemandnetwork
2.3.Printingondemand(POD):theEspressoBookMachine

2.4.ParticipativeOCRcorrectionandparticipativetranscriptionofmanuscripts
2.5.Folksonomy,cataloguingandparticipativeindexing
3OverviewandKeystoSuccess
3.1.Typologiesandtaxonomiesofprojects
3.2.Communicationandmarketingforrecruitingvolunteers
3.3.Thequestionofmotivations
3.4.Sociologyofthecontributorsandcommunitymanagement
3.5.Thequestionofthequalityofthecontributions
3.6.Theevaluationofcrowdsourcingprojects
3.7.Changemanagement
Conclusion
Bibliography
Index


EndUserLicenseAgreement

ListofTables
1AConceptualIntroductiontotheConceptofCrowdsourcinginLibraries:ANew
Paradigm?
Table1.1.Multicriteriadefinitionsofcrowdsourcing
2OverviewofSeveralCrowdsourcingProjectsAppliedtotheDigitizationofLibraries
Table2.1.StatisticsofEODordersfromtheBibliothèqueinteruniversitairede
Santé,from[KLO14],translatedbyus
Table2.2.Ratesofferedbyvariousinstitutionsofferingdigitizationandprintingon
demand
Table2.3.ExamplesofOCRization
Table2.4.StatisticsofthenumberofInternetusersnecessarytocorrectaword,
after[VON08b]
Table2.5.StatisticscollectedintheliteratureregardingthereCAPTCHAproject
Table2.6.ComparativecostsbetweenOCRcorrectionviatheAMTandviaaservice
provider
Table2.7.EstimateofthecostsnotpaidforOCRcorrectionservicesbecauseofthe
useofcrowdsourcing
3OverviewandKeystoSuccess
Table3.1.Modelofpublicparticipationinspiredby[BON09]
Table3.2.Activitiesofadigitizationprojectcrossedwiththetypesof
crowdsourcing.Foracolorversionofthetable,see
www.iste.co.uk/andro/libraries.zip
Table3.3.Existingtypesofcrowdsourcingappliedtodigitization
Table3.4.Typesofcrowdsourcingappliedtodigitizationthatremaintobeinvented
Table3.5.Taxonomyofcrowdsourcingappliedtodigitization
Table3.6.Datacollectedintheliteratureaboutthesociologyofthecontributorsto
differentprojects
Table3.7.Distributionoftheworkingtimeofcrowdsourcingstaffaccordingto
activitiesandmissions,from[SMI11]
Table3.8.Useofsocialmetadatamadebyculturalinstitutions,accordingtothe
OCLCstudy[SMI11]


Table3.9.StatisticsbeforeandaftercrowdsourcingfortheCaliforniaDigital
NewspaperCollection,from[GEI12]
Table3.10.IndicatorsofquantitativeanalysisofOCRcorrectionortranscription
projects
Table3.11.Indicatorsofquantitativeanalysisofcontentindexingprojects
Table3.12.Indicatorsofquantitativeanalysisofdigitizationondemandprojects
Table3.13.Otherindicatorsofevolution
Table3.14.CalculationofwhatOCRcorrectionwouldhavecostwithoutuseof
crowdsourcingforseveralrepresentativeprojects,from[AND15]

ListofIllustrations
1AConceptualIntroductiontotheConceptofCrowdsourcinginLibraries:ANew
Paradigm?
Figure1.1.TheartworkTenThousandCents3.Foracolorversionofthefigure,see
www.iste.co.uk/andro/libraries.zip
Figure1.2.Anartworkjuxtaposingsheep4
Figure1.3.13thCenturyswordwhosephotographwaspublishedbytheBritish
Library5.Foracolorversionofthefigure,seewww.iste.co.uk/andro/libraries.zip
Figure1.4.Changeinthenumberorsearchesfortheword“crowdsourcing”on
Googleforeachcountry,accordingtoGoogleTrends.Foracolorversionofthe
figure,seewww.iste.co.uk/andro/libraries.zip
Figure1.5.CountriesrepresentedinthesurveyconductedbyOCLCaboutsocial
metadata,from[SMI11].Foracolorversionofthefigure,see
www.iste.co.uk/andro/libraries.zip
Figure1.6.Changeinthenumberofpublicationsoncrowdsourcingindexedby
GoogleScholarappliedtothedigitizationoflibraries.Foracolorversionofthe
figure,seewww.iste.co.uk/andro/libraries.zip
Figure1.7.Relationshipsbetweenhumancomputation,collectiveintelligenceand
crowdsourcing,accordingto[HAR13].Foracolorversionofthefigure,see
www.iste.co.uk/andro/libraries.zip
Figure1.8.Positionofcrowdsourcingamongneighboringareas,accordingto[SCH
10].Foracolorversionofthefigure,seewww.iste.co.uk/andro/libraries.zip
Figure1.9.Thefirstformofcrowdfunding.From
http://gallica.bnf.fr/ark:/12148/btv1b8509563b(consultedJune23,2016).Fora
colorversionofthefigure,seewww.iste.co.uk/andro/libraries.zip


Figure1.10.PercentageofWikipediansbybirthdate,accordingtoWikipedia
2OverviewofSeveralCrowdsourcingProjectsAppliedtotheDigitizationofLibraries
Figure2.1.LocationofthemembersofeBooksonDemandnetworkonJuly8,2014,
fromhttps://www.facebook.com/eod.ebooks/app_402463363098062(consultedJune
23,2016).Foracolorversionofthefigure,seewww.iste.co.uk/andro/libraries.zip
Figure2.2.ExtractfromanEODactivityreport,from[KLO14]
Figure2.3.Ordersperpriceclassduringthe2009–2011periodattheNational
LibraryofSlovenia,from[BRU12]
Figure2.4.Theforminwhichusersprefertoconsultdocuments,accordingtothe
surveyrelatedby[MUH09]
Figure2.5.Positive/negativeperceptionaccordingtopricesanddeliverytimes,
accordingtothesurveyrelatedby[MUH09]
Figure2.6.Areasofinterestforusers,from[GST11]
Figure2.7.Reasonswhyusersplacedorders,from[GST11]
Figure2.8.PhotographofanEspressoBookMachine,fromondemandbooks.com.
Foracolorversionofthefigure,seewww.iste.co.uk/andro/libraries.zip
Figure2.9.DistributionofEBMthroughouttheworld,accordingto
http://www.ondemandbooks.com/ebm_locations.php(consultedonJuly9,2014).For
acolorversionofthefigure,seewww.iste.co.uk/andro/libraries.zip
Figure2.10.ScreencaptureofarawOCRtext.Foracolorversionofthefigure,see
www.iste.co.uk/andro/libraries.zip
Figure2.11.ScreencaptureofadigitizednewspaperanditsOCR.Foracolor
versionofthefigure,seewww.iste.co.uk/andro/libraries.zip
Figure2.12.ChangeinthenumberofcorrectionsonlinesonTROVEaccordingto
statisticsobtainedfromthesiteitself(source:http://trove.nla.gov.au/system/stats?
env=prod)
Figure2.13.ScreencaptureofTROVE3.Foracolorversionofthefigure,see
www.iste.co.uk/andro/libraries.zip
Figure2.14.BudgetoftheTranscribeBenthamproject,accordingto[CAU12b]
Figure2.15.Evolutionofthenumberofaccounts,manuscriptstranscribedand
completedbetweenSeptember8,2010andMarch8,2011,accordingto[CAU12b].
Foracolorversionofthefigure,seewww.iste.co.uk/andro/libraries.zip
Figure2.16.ButtonusedbyTranscribeBentham.Foracolorversionofthefigure,
seewww.iste.co.uk/andro/libraries.zip
Figure2.17.ThetranscriptioninterfaceofTranscribeBentham,from[BRO12].For


acolorversionofthefigure,seewww.iste.co.uk/andro/libraries.zip
Figure2.18.DiagramrepresentinghowInternetusersdiscoveredtheTranscribe
Benthamproject,accordingto[CAU12a].Foracolorversionofthefigure,see
www.iste.co.uk/andro/libraries.zip
Figure2.19.DiagramrepresentingthedistributionofcontributorstoTranscribe
Benthamaccordingtoage,accordingto[CAU12a].Foracolorversionofthe
figure,seewww.iste.co.uk/andro/libraries.zip
Figure2.20.MotivationsofthevolunteersoftheTranscribeBenthamproject,from
[CAU12a]
Figure2.21.ScreencaptureofthegameMoleHunt.Foracolorversionofthefigure,
seewww.iste.co.uk/andro/libraries.zip
Figure2.22.ScreencaptureofthegameMoleBridge.Foracolorversionofthe
figure,seewww.iste.co.uk/andro/libraries.zip
Figure2.23.Proportionofworkcarriedoutby1,10and25%,ofthebest
contributors,from[CHR11]
Figure2.24.DiagramexplaininghowreCAPTCHAworks,accordingtothesite
Google.com.Foracolorversionofthefigure,seewww.iste.co.uk/andro/libraries.zip
Figure2.25.AnotherdiagramexplaininghowreCAPTCHAworks,from[IPE11].For
acolorversionofthefigure,seewww.iste.co.uk/andro/libraries.zip
Figure2.26.TheTurkishchessplayer,TuerkischerschachspielerwindischbyKarl
GottliebvonWindisch,1783,publicdomainviaWikimediaCommons
Figure2.27.NumberofHITsinNovember2013,accordingtotheMechanicalTurk
tracker
Figure2.28.DistributionofIndianworkersandAmericanworkersonAMTbysex,
accordingto[IPE10b].Foracolorversionofthefigure,see
www.iste.co.uk/andro/libraries.zip
Figure2.29.BirthyearofworkersontheAMT,accordingto[IPE10b]
Figure2.30.EducationallevelofworkersontheAMT,accordingto[IPE10b].Fora
colorversionofthefigure,seewww.iste.co.uk/andro/libraries.zip
Figure2.31.AveragetimededicatedtotheAMT,accordingto[IPE10b]
Figure2.32.AverageincomemadefromtheAMT,accordingto[IPE10b]
Figure2.33.NumberofworkersstatingthatAMTistheirprimarysourceofincome,
accordingto[IPE10b].Foracolorversionofthefigure,see
www.iste.co.uk/andro/libraries.zip
Figure2.34.Typesofmotivationaccordingtothegreaterorlesserdedicationof
workersontheAMTplatform,accordingto[KAU11]


Figure2.35.NumberofcorrectionsonTROVEbetween2008and2012,accordingto
[HAG13]
Figure2.36.Changeintheamountofcontentcomparedtothatofthenumberof
correctionsonTROVE,accordingto[HAG13].Foracolorversionofthefigure,see
www.iste.co.uk/andro/libraries.zip
Figure2.37.Proportionofgenealogistsamongthecontributors,accordingtoa
CDNC/CambridgePublicLibrarysurvey.Foracolorversionofthefigure,see
www.iste.co.uk/andro/libraries.zip
Figure2.38.Distributionofvolunteersbyagegroup,accordingtoa
CDNC/CambridgePublicLibrarysurvey.Foracolorversionofthefigure,see
www.iste.co.uk/andro/libraries.zip
Figure2.39.ThetypesofdocumentsdistributedonTROVEcomparedtothetypesof
documentsthatarecorrectedthere,accordingto[HAG13].Foracolorversionof
thefigure,seewww.iste.co.uk/andro/libraries.zip
Figure2.40.MostcorrectedtypesofdocumentsonTROVE,accordingto[HAG13].
Foracolorversionofthefigure,seewww.iste.co.uk/andro/libraries.zip
Figure2.41.Classificationofcontributorsaccordingtothenumberoflines
correctedfortheTROVEandCDNCprojects,accordingto[ZAR14].Foracolor
versionofthefigure,seewww.iste.co.uk/andro/libraries.zip
Figure2.42.PortionoftheworkaccomplishedbyeachcontributortotheOld
Weatherprojectofferingtotranscribemeteorologicalobservations,fromBrumfield,
manuscripttranscription.blogspot.fr,2013.Foracolorversionofthefigure,see
www.iste.co.uk/andro/libraries.zip
Figure2.43.ScreencaptureofthegameArtCollector,firstround,from[PAR13].
Foracolorversionofthefigure,seewww.iste.co.uk/andro/libraries.zip
Figure2.44.ScreencaptureofthegameArtCollector,round2,choiceofapiece,
from[PAR13].Foracolorversionofthefigure,see
www.iste.co.uk/andro/libraries.zip
Figure2.45.ScreencaptureofthegameArtCollector,round2,tryingtowinawork,
from[PAR13].Foracolorversionofthefigure,see
www.iste.co.uk/andro/libraries.zip
Figure2.46.GenderandageoftheplayersofArtCollector,accordingto[PAR13].
Foracolorversionofthefigure,seewww.iste.co.uk/andro/libraries.zip
3OverviewandKeystoSuccess
Figure3.1.Taxonomyofcrowdsourcing,from[HAR13]
Figure3.2.Taxonomyofthe4Csofcrowdsourcing,from[REN14b]


Figure3.3.Timeevolutionsince2011andforecastofthefuturegamificationmarket,
from[OLL13]
Figure3.4.Seriousgamesandgamification,from[DET11a].Foracolorversionof
thefigure,seewww.iste.co.uk/andro/libraries.zip
Figure3.5.ScreencaptureoftheWhat’sonthemenu?pressrelease:“HelptheNew
YorkPublicLibraryimproveauniquecollection.Weneedyou!Helptranscribe.It’s
easy!Noregistrationrequired!”from[VER13].Foracolorversionofthefigure,
seewww.iste.co.uk/andro/libraries.zip
Figure3.6.Taxonomyofthemotivationsofvolunteersinacrowdsourcingproject.
Foracolorversionofthefigure,seewww.iste.co.uk/andro/libraries.zip
Figure3.7.Maslow’sHierarchyofneeds,Byuser:Factoryjoe(Mazlow'sHierarchy
ofNeeds.svg)[CCBY-SA3.0(https://creativecommons.org/licenses/by-sa/3.0)],via
WikimediaCommons(consultedOctober4,2017).Foracolorversionofthefigure,
seewww.iste.co.uk/andro/libraries.zip
Figure3.8.DiagramshowingthatahandfulofInternetusersarethesourceofthe
majorityofcontributions,fromBrumfield,manuscripttranscription.blogspot.fr,
20134.Foracolorversionofthefigure,seewww.iste.co.uk/andro/libraries.zip
Figure3.9.Distributionofstaffactivitiesinmanagementofcrowdsourcingprojects,
from[SMI11]
Figure3.10.Theworkingtimeofcrowdsourcingprojectstaff,from[SMI11].Fora
colorversionofthefigure,seewww.iste.co.uk/andro/libraries.zip
Figure3.11.Frequencywithwhichsitesputnewcontentonline,from[SMI11b].For
acolorversionofthefigure,seewww.iste.co.uk/andro/libraries.zip
Figure3.12.Thecriteriaforsuccess,from[SMI11]
Figure3.13.Numberofuniquevisitorspermonthforcrowdsourcingprojects,from
[SMI11].Foracolorversionofthefigure,seewww.iste.co.uk/andro/libraries.zip
Figure3.14.Numberofcontributorspermonthforculturalinstitutions,from[SMI
11].Foracolorversionofthefigure,seewww.iste.co.uk/andro/libraries.zip


DigitalToolsandUsesSet
coordinatedby
ImadSaleh
Volume5

DigitalLibrariesandCrowdsourcing
MathieuAndro


Firstpublished2018inGreatBritainandtheUnitedStatesbyISTELtdandJohnWiley&Sons,Inc.
Apartfromanyfairdealingforthepurposesofresearchorprivatestudy,orcriticismorreview,aspermittedunderthe
Copyright,DesignsandPatentsAct1988,thispublicationmayonlybereproduced,storedortransmitted,inanyformorbyany
means,withthepriorpermissioninwritingofthepublishers,orinthecaseofreprographicreproductioninaccordancewiththe
termsandlicensesissuedbytheCLA.Enquiriesconcerningreproductionoutsidethesetermsshouldbesenttothepublishersat
theundermentionedaddress:
ISTELtd
27-37StGeorge’sRoad
LondonSW194EU
UK
www.iste.co.uk
JohnWiley&Sons,Inc.
111RiverStreet
Hoboken,NJ07030
USA
www.wiley.com
©ISTELtd2018
TherightsofMathieuAndrotobeidentifiedastheauthorofthisworkhavebeenassertedbyhiminaccordancewiththe
Copyright,DesignsandPatentsAct1988.
LibraryofCongressControlNumber:2017958934
BritishLibraryCataloguing-in-PublicationData
ACIPrecordforthisbookisavailablefromtheBritishLibrary
ISBN978-1-78630-161-1


Preface
Inlieuofoutsourcingcertaintaskstoserviceproviderswithaccesstocountrieswherelaboris
cheap,librariesthroughouttheworldarerelyingmoreandmoreongroupsofinternetusers,
turningtheirrelationshipwithusersintoonethatismorecollaborative.Afteraconceptual
chapterabouttheconsequencesofthisneweconomicmodelonsocietyandonlibraries,an
overviewofprojectsintheareasofon-demanddigitization,participativecorrectionofOCR
especiallyintheformofgames(gamification)andfolksonomywillbepresented.This
panoramaleadstoanoverviewofcrowdsourcingappliedtodigitizationanddigitallibraries
andanalysesintheareaofinformationandcommunicationsciences.

Acknowledgments
IwouldliketothankImadSaleh,ProfessorattheParagrapheLaboratoryofParis8University,
forhavingagreedtosupervisemythesisproject,forhiskindnessandforhisadvicethroughout
theentireproject;SamuelSzoniecky,SeniorLecturerattheParagrapheLaboratoryofParis8
University,forhavingagreedtobetheco-directorofmythesisandforhavinginvitedmeto
speakwithhisstudents;GhislaineChartron(ProfessorattheNationalConservatoryofArts
andCrafts);StộphaneChaudiron(ProfessoratCharles-de-GaulleUniversity,Lille3);Cộline
Paganelli,SeniorLecturerHDR(accreditationtosuperviseresearch)atPaulValộry
University,Montpellier;AlainGarnier,CEOofJamespotandcrowdsourcingadvisoratthe
GroupementFranỗaisdesIndustriesdelInformation(GFII)forhavingagreedtobean
examinerofmythesis;FranỗoisHoullier,InstitutUniversity,NationaldelaRecherche
Agronomique(INRA),forlettingmeparticipatealongsidehiminataskforceoncitizen
scienceinordertosubmitareportonthesubjectattherequestoftheappropriateministers;
OdileHologne,fromthedepartmentofpromotingscientificandtechnicalinformationofthe
InstitutNationaldelaRechercheAgronomique(INRA),forhavingencouraged
experimentationaroundINRAsNumalireprojectwithintheframeworkofmywork;Filippo
GropalloandDenisMaingreaud,fromthecompanyOrangeandthecompanyYabộ,fortheir
projectNumalireinwhichtheyallowedmetoparticipate,andfortheircollaboration
throughoutthisresearchproject;MarcMaisonneuveandEmmanuelleAsselin,fromthe
consultingfirmTOSCA,fortheircollaborationonthebookthatwepublishedtogetheron
softwareandplatformsfordevelopingdigitallibraries;GaởtanTrửger,EcoledesPonts
ParisTech,forhiscollaborationinthestudythatwecarriedoutonthevisibilityandstatistics
oftheconsultationofdigitallibraries;PaulineRiviốre,Sainte-GeneviốveLibrary,andAnaùs
Dupuy-Olivier,AcadộmiedeMộdecine,fortheircollaborationinthefeedbackonthe
Numalireexperimentthatwewrotetogether;RobertMiller,InternetArchive,forthe
collaborationthatwehadatSainte-GeneviốveLibrary,whichbecamethefirstlibraryin
FrancetoparticipateintheInternetArchive;StộphaneIpert,CentredeConservationduLivre,
forthecollaborationsandinterestingdiscussionsthatwehad;PierreBeaudoinandRộmi
Mathis,previousandcurrentpresidentsofWikimediaFrance,anassociationwithwhich


collaborationswithWikisourcewereachieved(NationalVeterinarySchoolofToulousein
2008)oronlyenvisioned(Sainte-GenevièveLibrary);ValérieChansigaud,sciencehistorian
andWikipediacontributor,withwhomfirstcontactwasestablishedatthemuseumfollowedby
apilotexperimentindigitizationandparticipativecorrectionofOCR,whichwasconductedin
2008attheNationalVeterinarySchoolofToulouse;Gilonned’Origny,fromthecompany
ondemandbooks.com,withwhomacollaborationonthefirstinstallationofanEspressoBook
MachineinFrancewasunsuccessful;DanielTeeter,fromthecompanyAmazon,forthe
interestingopportunityforpartnershipthatwasnearlyestablished;JuanPirlotdeCorbion,
founderofchapitre.comandYouScribe,forthepassionatediscussionsthatwehadoverthe
courseofourmeetings;DanielBenoilid,founderofthepaidcrowdsourcingcompanyFoule
Factory,forthediscussionsthatwehad;Jean-PierreGerault,CEOofthecompanyI2S,leader
intheareaofmanufacturingscannersforthedigitizationofheritage,presidentoftheComité
RichelieuandCEOofPublishroom,fortheinterestingdiscussionsthatwehad;Arnaud
Beaufort,NationalLibraryofFrance,whomImetduringtheWikimediadaysattheNational
AssemblyandwithwhomIthenhadaninterestingconversation;SilviaGstreinandVeronika
Gründhammer,UniversityofInnsbruck,forhavinginvitedmetospeakattheEbookson
Demand2014conference;YvesDesrichardandArmelledeBoisse,EcoleNationale
SupérieuredesSciencesdel’InformationetdesBibliothèques,forhavingallowedmetospeak
duringthe“Quoideneufenbibliothèques?”daystheselast5years;ThierryClaerr,Ministry
ofCultureandCommunication,whoallowedmetospeakregularlyattheENSSIBandsought
meouttowriteacollaborativework,andwithwhomIhadsomeveryenrichingdiscussions;
Jean-MarieFeurtet,AgenceBibliographiquedel’EnseignementSupérieur,forour
collaborationonamutualizationprojectofadigitallibraryandforhavinginvitedmetospeak
atthe2011ABES;NicolasTurenne,InstitutNationaldelaRechercheAgronomique(INRA),
forhavinginvitedmetoshowthepreliminaryresultsofthisworkattheseminarentitled
“DigitalTraces”(Cortextgroup,InstituteforResearchandInnovationinSociety);PierreBenoîtJoly,directoroftheInstituteforResearchandInnovationinSociety(IFRIS),forhaving
invitedmetogiveamaster’slevelcourseinDigitalStudiesandInnovation(NUMI);SNCFfor
thecomfortofthetraintripsItookwhilewritingthisthesis;GooglefortheGoogleDrive
service,whichwasusedtowritethethesiswhileprovidingreal-timeaccesstoitforthe
director,mycollaboratorsandmycontactswhothenhadtheopportunitytoaddcomments;my
wifeVéroniqueandmythreechildrenTerence,OréganeandEloïse.
Ialsowanttothankthefollowingpeoplefortheconstructivecommentsthattheyaddedtothe
textofthethesismadeavailableinitsfirstdraftonGoogleDrive:ChristineYoung
(proofreadingthearticleinEnglish),WilfridNiobet(oneidea,eightcomments,six
corrections),CélyaGruson-Daniel(threecomments,fourcorrections),OliviaDejean(nine
corrections),MichaëlJeulin(sevencorrections),CatherineThiolon(tencomments),Caroline
Dandurand(fivecomments),DianeLeHénaff(threecomments),SophieAubin(two
comments),NicolasRicci(onecomment),PaulineRivière(onecomment),Frédérique
Bordignon(onecomment),SylvieCocaud(onecomment),MarjolaineHamelin(onecomment),
SilvèreHanguehard(onecomment),ChristineSireyjol(onecomment),OdileViseux(one
comment),VéroniqueDecognet(onecomment),DominiqueFournier(twocorrections)andall
ofthe“unknownsoldiers”whoremainedanonymousintheircomments(82corrections).


MathieuANDRO
November2017


Introduction
Librariesalreadyresorttooutsourcingcertaintasksinvolvedinenteringbibliographic
records,cataloguing,indexationorOCRcorrection,toserviceprovidersincountrieswhere
laborisinexpensive.Thisoutsourcinghasremainedwithinacontractualandlimited
frameworkandhasnotprofoundlyoverturnedtheunderlyingwaysinwhichlibrarieswork.
However,withthedevelopmentofcrowdsourcing,itispossibletoimagineexternalizing
(outsourcing)someofthesetasksnottoserviceprovidersbutto“crowds”ofInternetusersand
thereforehavingamateurscarryoutsomeoftheprofessionals’work.Crowdsourcingthus
changestheparadigmuponwhichlibrariesarebased,whichnowlargelycentersaroundthe
creationandconservationofcollections.Italsochangedtherelationshipbetweentheservice
providers,namelythelibrarians,andtheirconsumers,namelytheusers.Thelatterarealso
becomingactiveproducersofservices.Crowdsourcingcouldalsointerrogatethecollection
managementpoliciesoflibraries,whichanticipateneedbasedonasupplythatisnotdirectly
orimmediatelydeterminedbydemand.Thisisespeciallythecasewiththeon-demand
digitizationbycrowdfunding,aformofcrowdsourcingthatcallsnotontheworkofcrowds,
butontheirfinancialresources,orwiththeprintingondemandwhichisinseparablefromit.
Withtheseon-demandeconomicmodels,thecollectionmanagementpolicyisfinallyshared
withuserswhodecidewhatwillbedigitizedand/orprinted.Inthisway,thecollections
becometheworkoftheusers.
Thisbookhasthegoalofprovidingresponsestothequestionofrelyingoncrowdsourcingfor
libraryprofessionals,aswellasforstudents,researchersininformationandcommunication
sciencesand,moregenerally,peopleinterestedincollectiveintelligenceprojects.Itisthe
resultofathesisoninformationandcommunicationsciencesthatsimultaneouslyincludes
actionresearch,anexperimentandananalysisoftheliterature[AND16].Thisthesisitselfhas
previouslybeenthesubjectofanarticleusingthemaincontributions[AND17].
Beyondthequestionsofcosts/benefitsandadvantages/disadvantages,thequestionofan
evolutionofthelibrarian’sprofessionrefocusedontheirsingularskillswillbeaddressed.
Thisworkalsohasthescientificgoalofprovidingacontributiontoknowledgeof
crowdsourcingonthetheoreticalandconceptuallevelaroundeconomicmodels.
Thisworkislimitedtotheapplicationofcrowdsourcingintheareaofdigitizationanddigital
libraries.Sincethe1990s,thedigitizationofdocumentshasbeenwidespreadinlibraries.
Today,withmassdigitizationandthedevelopmentofgiganticdigitallibrariessuchasGoogle
Books,whichhascrossedthethresholdof30millionbooks,orInternetArchive,HathiTrust,
Europeana,the“harvester”ofEuropeandigitallibraries,itisbecomingmoreandmore
difficulttoidentifyprintedmatterthathasnotbeendigitizedandstilldeservestobe,amongthe
130million1existingtitlesprintedsincetheinventionofprinting.
Asignificantpartofwhathasbeendigitizedbylibrarieshasneverbeenputonline.Itgenerates
duplicatedigitizationandis“sleeping”onCD-ROMs,DVDsorexternalharddriveswhose


lifetimeislimited.Thedevelopmentofadigitallibrarycan,infact,beexpensiveintermsof
softwareadministrationandservers,andtheresultcanbedisappointingintermsof
functionalities,durability,costsandvisibility.In2012,wepublishedastudydedicatedtothe
softwareprogramsYooLib(Polinum),Invenio(CERN),ORI-OAI(universities),DSpace
(DuraSpace),DigiTool(ExLibris),Mnesys(Naoned),ContentDM(OCLC),Eprint(University
ofSouthampton),Greenstone(UniversityofWaikato)andOmeka(GeorgeMasonUniversity)
[AND12].Inthisstudy,wefoundthatitwasmoreadvantageousforlibrariestoparticipatein
ashareddigitallibrarysuchasInternetArchiveasmuchfromthepointofviewofcosts(free),
functions(opticalchapterrecognitionandconversionintoEPUBandMOBIfore-readers
directlyimplementedonarchive.org)andpermanentarchiving(multiplemirrorserversaround
theworld)asfromthatofvisibility.Indeed,thepositionofawebsiteinthelistofGoogle
searchresultsdependsonitsPageRank.Thisdependslargelyonthenumberoflinksthatpoint
toitsdomainname.Undertheseconditions,adigitallibrarywithalargeamountofcontent
willautomaticallyhaveabetterPageRankandbettervisibilityonthewebandwilltherefore
generatemuchmorewebtrafficthanasmalldigitallibrarywithverylittlecontent.
AsWaibel[WAI08]maintains,twoschoolsofthoughtexist:anoldschoolthatbelievesthat
eachlibraryneedstocreateitsowndigitallibraryandattempttoattractInternetuserstoit,and
anewschoolthatinsteadbelievesthatingoingbeyondinstitutionalcommunicationandbetter
satisfyingtheneedsofInternetusers,librarieswouldbebetteroffparticipatinginthedigital
librariescollectivesalreadyvisitedbyInternetusers,suchasInternetArchiveorevenFlickr.
Thisisalsoourpointofview.Withenoughwebtraffic,librariesmayprompttheparticipation
ofInternetusers.
Theintroductorypartofthebookattemptstoarticulateitscontextandthemethodologythat
wasused.
Chapter1addressesthephilosophical,politicalandeconomicrepresentationsof
crowdsourcinganditsconsequencesregardingthewayinwhichlibrariesfunction.This
conceptualchaptercontains,inparticular:
–acriticaldiscussionregardingthedefinitionofcrowdsourcing;
–anoriginalchronologyofitshistoricalorigins;
–ananalysisonthesubjectofitsconceptualoriginsinphilosophicalcurrentsthatare
sometimesdiametricallyopposedand,inparticular,aconceptualcontributionaroundthe
lawofvalue;
–areflectionontheconceptofthewisdomofcrowds;
–ananalysisofthediversecritiquesofcrowdsourcingappliedtodigitallibrariesthat
somepeoplecouldtodaydescribeasthe“uberization”ofdigitallibraries.
Chapter2containsaselectionofprojectsthroughtypesoftasksincluding:
–puttingcontentonlineandparticipativecuration;
–digitizationandprintingondemandintheformofcrowdfunding;


–participativecorrectionofOCRandparticipativetranscriptionofmanuscripts;
–folksonomy.
Thischaptercontainsdataandinformationcollectedfromtheliteratureforeachproject.
OriginalanalysesforeachmajortypeofprojectaregivenintheconclusionofChapter2.
InChapter3,analysesfromthepointofviewofinformationandcommunicationsciencesanda
stateoftheartareofferedwith,notably:
–anoriginaltaxonomyofcrowdsourcingindigitallibrariesdistinguishingexplicit(or
conscious),voluntaryandpaidcrowdsourcingandimplicit(orunconscious)
crowdsourcing,gamificationandcrowdfunding;
–ananalysisofthemotivationsoflibrariesandtheconditionsnecessaryforthe
developmentofcrowdsourcingprojects;
–ataxonomyofthemotivationsofInternetuserswhocontributetotheirprojects;
–analysesofthepossiblerewardsandremuneration;
–clarificationregardingthecommunicationnecessaryforrecruitment;
–developmentsinthespecificcommunitymanagementofthistypeofproject;
–analysesofthequestionofthequalityandreintegrationofthedataproduced;
–areflectionontheevaluationofcrowdsourcingprojects.
1ThenumberofbooksthathavebeenprintedsinceGutenberg’sinventionoftheprintingpress
isestimatedat129,864,880byLeonidTaycher,anengineeratGoogle,accordingtoan
articlepublishedonhisblogonAugust5,2010.


1
AConceptualIntroductiontotheConceptof
CrowdsourcinginLibraries:ANewParadigm?
1.1.Arapidlygrowingeconomicmodel
1.1.1.Whatmadethisneweconomicmodelpossible
Internetusersaregrowingmoreandmorenumerousandthetimethattheyspendsurfingthe
Internetisgrowing.TheonlineencyclopediaWikipediarequired100millioncumulativehours
tobeconstructed.AsClayShirkystatedonAugust28,2008attheWiki-ConferenceNYC,if
Americans,whowatch200billionhoursoftelevisioneveryyear,usedthattimeforcreative
activitiesinstead,theycouldcreate2,000projectssuchasWikipediaeachyearinsteadof
watchingtelevision.
Duringa2011TEDconference,LuisVonAhn1claimedthatusingonly100,000people,
humanitysucceededinbuildingpyramidsanddiggingthePanamaCanal,andthatbecauseof
theInternetandsocialnetworks,itisnowpossibletoassemble750millionpeople,for
example,foraprojectcorrectingtheOpticalCharacterRecognition(OCR)suchas
reCAPTCHA.Anamazing“reservoirofgoodwill”isthereforepotentiallyavailablefor
culturalinstitutionsiftheyknowhowtobenefitfromit.
ParticipatorymodelscameaboutwiththedevelopmentoftheWeb2.0.Thetermwasinvented
byDiNucciin1999[NGU12]orbyDaleDoughertyin2004[SAR14]andpopularizedby
TimO’Reillyin2005[TRA08].CrowdsourcingnowmeansthatInternetusersnolongerhave
tobecontentwithpassivelyconsumingWebcontentwithinahierarchical,unilateralandstatic
diffusionmodel(Web1.0),butcanactivelyparticipateinitsdevelopment.Thediffusionof
informationhasbecomereciprocal,interactiveanddynamic.TheInternetuserthereforeceases
tobeaconsumer,areaderandapassivereceptorwhoiscontenttobrowse,andbecomesa
producer,anauthor,anactiveemitterofinformation,acontributorwhocanparticipateinthe
writingandmodificationofcontentontheWeb(comments,tags,wikis,socialnetworks,etc.)
andintheproductionofdataandmetadata.Theauthorityofdatahasthusbeenmovedfromthe
servertothecustomer[BAI12].AstelecommunicationsexpertBenjaminBayartemphasizes,
ifprintingtaughtpeopletoread,theInternetisnowteachingthemtowrite2.
WellbeforeWeb2.0,theinventionof“self-service”whichgrantedtheconsumerdirectaccess
tomerchandisewithouttheintermediaryofavendorandwhichwasappliedtolibrariesinthe
formofopenaccesscollections,wasanearlyformoftheintegrationoftheconsumerintothe
productionprocess.ThiseconomicmodelwasinventedbyAristideBoucicautinhis
departmentstore“LeBonMarché”whosesloganwas“self-service,freetotouch”giving
customers,asdescribedinZola’sAubonheurdesdames(translatedintoEnglishasThe


Ladies’DelightorTheLadies’Paradise),theopportunitytoaccessthemerchandiseactively
andfreely,withoutashopkeeperasanintermediary,and,infine,totakeoverpartofthe
merchants’andstoreowners’jobs.Broadlyspeaking,productionseemstohavethus
progressivelylostthecentralplacethatitoccupiedinfavorofconsumptionandtheconsumer
societythatdevelopedaftertheSecondWorldWar.
Later,the“justintime”model,developedatToyota,consistedofproducingproducts“on
demand”forthecustomerinordertoavoidunsoldstockbyproducingjust-in-timesupplyina
waythatissynchronizedwithanddrivenbydemand.Thismodelof“manufacturingwithout
waste”,“leanmanufacturing”or“fat-freemanufacturing”consistsofproducingonlywhatyou
strictlyneed,withthenecessarycorrectmeans,atthetimewhenitisneededandattheleast
possiblecosttotheproducertoexternalizethedecisiontobeginproductionwiththeconsumer.
ThismodelwasbornfromthedifficultyJapanesestoreshadinstockingmerchandisedueto
insufficientspaceandthenecessityofresupplyingonlywhenstockranout.Itwasalso
significantlyinspiredbythewayinwhichsupermarketsoperate.Inthesameway,theclothing
chainZarakeepsonlyasinglemonthworthofinventoryandthusbetteradaptsitsproductionto
trendsinthemarket,producingmodelsdependingonsales[SUR04].Advertisingitself
participatesintheintegrationoftheconsumerintotheproductionprocess.Indeed,whenwe
viewatelevisionprogramorwebsite,weproducestatisticsanddata,orwhenweview
advertisements,wealsoproducevalue.Wecanthereforetalkaboutaneconomyofattention
[CIT14].Thedecisiontovisitthisorthatsitecouldthereforebelikenedtoavote,avotethat
participatesinproductionandrevenuesoftheproducers.Thismodelhasfounditsapplication
inlibraries,inon-demanddigitizationbyparticipatoryfinancing(crowdfunding)andin
printingondemand,whichwillbeaddressedinthisbook.
Today,crowdsourcingcontinuestherelativelyoldmovementofintegratingtheconsumerinto
theproductionprocess.ItwasmadepossiblebythedevelopmentofthetechnologiesofWeb
2.0.Bornfromaculturalevolutiontowardmoreparticipativeandcollaborativeapproaches,
crowdsourcingwasmadetechnologicallypossiblebyWeb2.0,thatistosay,thepossibilityof
havingalargenumberofpeople,whohavefreetimeavailableontheWeb,workremotelyon
collectiveprojects.Itisespeciallyinspiredbythewaycommunitiesoffreewaredevelopers
work.BycallingonacrowdofInternetusers,itispossibletocarryout,inverylittletime,
tasksthatpreviouslywouldhavebeenimpossibletocompleteorevenimagine,orthatwould
haverequiredhugeamountsoftime.Inshort,crowdsourcing“isawaytofindaneedleina
haystack”,asLebratyandLobre[LEB15]state.Sagotetal.[SAG11]talkabout
“myriadizationofdividedwork”andmicroworking.Wecouldalsotalkaboutthe
“taskification”ofwork.Crowdsourcinghassomesimilaritiestotheconstructionofmedieval
cathedrals,whichrequiredthecapacityto“thinkbig”,todelegate,toorganizeeverytaskand
abovealltomobilizealargenumberofpeoplearoundacommonvisionandgoal,asLevi
[LEV14]recalls.Itisalso,totakeamorerecentexample,whatAlfredSloanofGeneral
Motorsdescribedas“groupmanagement”,whichconsistsofthesolicitationofnumerous
collaboratorstomakethemostimportantdecisions.
WeillustratethisideawithcontemporaryworksofartinFigures1.1and1.2.


Figure1.1.TheartworkTenThousandCents3.Foracolorversionofthefigure,see
www.iste.co.uk/andro/libraries.zip

Figure1.2.Anartworkjuxtaposingsheep4
Inadditiontoart,crowdsourcinghasalreadyfoundapplicationsinmanyareas.Forexample,
inthefieldofvideo,YouTubeandDailyMotioncouldnotfunctionwithoutcontentposted
onlinebyInternetusers.Crowdsourcinghasalsofoundapplicationsinmusic,politics,fashion,
banking,tourism,innovation,cartography,thesearchformissingplanes,medicine,scientific
research,publishing,translationandjournalism.Usingcrowdsourcingisalsotopicalinthe
fieldofGLAM(galleries,libraries,archivesandmuseums)anddigitallibrariesinparticular,
whichisthesubjectofthisbook.

1.1.2.Applicationtodigitallibraries
Forlibraries,digitizinganddiffusingtheircollectionsontheWebmeansthattheyfind
themselvesinthesamespaceastheirusers.Thissituationmakespossiblemultiplesynergies
andcollaborations.Amongculturalinstitutions,theamountofcontentthattheymakeavailable
ontheWebhasgrownexponentiallyandthereisnolackofpainstakingworkinindexing,
describingandcorrectingthiscontent.However,theirbudgetsandtheirworkforcehave
experiencedanoppositetrendwhichoftenleavesthemsorelylacking.Thisstateofaffairs
makesmanygoalsimpossibleandthecarryingoutofotherprojectsunimaginablewithout
externalaid.Inaddition,therealorvirtualpublicsoftheseinstitutionsarelessandless


contentwiththeroleofpassiveconsumerofculturalinformationandwouldincreasinglylike
togetinvolvedinservicetoheritageandculture.Inculturalinstitutions,theideaofbeing
receptivetointeractionwithaparticipatingpublicandvolunteerslargelyprecededthe
emergenceoftheWeb2.0.However,theRelationalWebhasfosteredtheemergenceofa
participativecultureonwhichthemodelofcrowdsourcinginlibrariesfeeds.
Indigitallibraries,crowdsourcingthusmakesitpossibletocompletetasksthatwouldbe
impossibletoundertakewithoutthehelpofvolunteerInternetusers,intheabsenceoffinancial
andhumanmeans.Thismeans,forexample,toimprovethequalityofmetadataortoenrichit
(comments,tags,analyses,etc.),tobenefitfromtheknowledgeandskillsofscholars,to
developcommunitiesaroundprojects,toincreasevisitstotheresourcesproduced,tomakethe
generalpublicmoreawareoftheconservationofcommonculturalheritage,togeneratemore
interactions,innovativeideasandcollaboration.Forexample,withintheonlinepublic,there
mightbesomeonewhowouldknowhowtoidentifyachurchinaphotograph,ascholarcould
provideinformationaboutitsconstructionanditshistory,anelderlyvillagerabletoidentifya
personinthephoto,etc.Theknowledgethatteamsoflibrarianshaveaccesstoismuchtoo
limitedtobeabletorespondtoallofthesequestions.Theknowledgepresentinthecrowdof
Internetusersislimitless.
TheBritishMuseumunderstoodthiswellwhen,onAugust3,2015,itpublishedacallto
Internetusersonbritishlibrary.typepad.co.ukwiththetitle,“HelpUsDecipherthis
Inscription”.BetweenAugust3and18,2015,theposthadbeensharedalmost32,000times
andhadgeneratedmorethan11,000sharesonFacebookand9,000tweets,aswellas115
commentsdirectlyontheblogbetweenAugust3and10.

Figure1.3.13thCenturyswordwhosephotographwaspublishedbytheBritishLibrary5.
Foracolorversionofthefigure,seewww.iste.co.uk/andro/libraries.zip
InordertomobilizeInternetusers,culturalinstitutionspossesssolidadvantages.Theyoften
alreadyhavesolidexperienceinmobilizingvolunteersandorganizingcontests,readinggroups
andeventsandeveninthe“adoption”ofbookswhosepurchaseisfinancedbyreadersor


patrons.Furthermore,theseinstitutionsenjoyapositiveimageamongthepublicandare
consideredtobetrustworthytoworkforthegeneralinterestandwhosegoalsarecultural,not
financial.Thesegoalsarethereforelikelytoattractvolunteersandelicitcontributions.
Crowdsourcingintheserviceofdigitallibrariesisalsothemeansofturningthesometimes
thanklessworkrequiredofasingleemployeeintoaworthwhileactivityofferedtoan
indefinitegroupofvolunteerInternetusersand“workerbees”whowouldliketoactively
contributetothedevelopmentoftheculturalWeb.Thedocumentsdigitizedandputonlineare
thustheobjectofaparticipativeredocumentarization,aremediationmakingitpossiblefor
newandcollaborativeprocessingofcollectionsofdocumentsbycallingandsometimeson
testimonyandmemory,andsometimesontheexpertiseandknowledgeofInternetusers.The
collectionsarethusrevisited,reinventedandreimagined.

1.1.3.Growinginterestfrompoliticians,Internetusersand
academics
ThesuccessofcrowdsourcingprojectsandtheinterestintheseprojectsfromInternetusers,
politiciansandacademicresearchers,isincreasing.AsSarrouy[SAR14]reports,a2011
studybymassolutions.comestimatedthecrowdsourcingmarketatmorethan300million
dollarswithagrowthrateofmorethan75%between2010and2011.In2012,anotherstudy
by[MCK14]evaluatedthegainsinproductivity,calculatingsocialmediaandcrowdsourcing
platformsinconsumergoods,financialservices,advancedproductionandprofessional
servicesat25%.Finally,attheendof2013,theGartnerfirmanticipatedthatby2017,more
thanhalfofproducersofconsumergoodswillbasemorethan75%oftheirresearchand
developmentoncrowdsourcing.Intheareaofcitizenscienceinvolvingbiodiversityalone,
researchersattheUniversityofWashingtonestimatethatthein-kindcontributionsofthe1.3–
2.3millionvolunteerswouldhaveaneconomicvalueofmorethan2.5billiondollars.
Crowdfunding,inparticular,wouldhavebeenabletofinanceamillionprojectsin2012and
raise2billioneuros[ONN13].Althoughthefinancingofprojectsbyprivateindividualsin
itselfisnothingnew,theInternetmakesiteasiertodoandtogivesanewscopeto
participatoryfinancingthatalreadyrepresentsamarketofthreebilliondollarsworldwidein
2012andwhosegrowthisexponential.
ByusingtheserviceGoogleTrends,whichistosaythetracesleftinvoluntarily6byInternet
userswhoperformGooglesearches,wealsoobservethat,beyondpolitics,moreandmore
Internetusersenteredtheword“crowdsourcing”,whichhasveryfewtranslationsintomodern
languages,intotheGooglesearchenginestartingin2006,whenthetermwaspopularizedby
JeffHowe.Inabase100system,thecountrieswhoseInternetuserscarriedoutthemost
searchescontainingthewordcrowdsourcingareinorderasfollows:theNetherlands(100),
Portugal(60),Germany(60),Spain(56),Singapore(55),Austria(54),Switzerland(54),the
UnitedStates(48),Brazil(43)Denmark(38)andtheUnitedKingdom(31).


Figure1.4.Changeinthenumberorsearchesfortheword“crowdsourcing”onGooglefor
eachcountry,accordingtoGoogleTrends.Foracolorversionofthefigure,see
www.iste.co.uk/andro/libraries.zip

Figure1.5.CountriesrepresentedinthesurveyconductedbyOCLCaboutsocialmetadata,
from[SMI11].Foracolorversionofthefigure,seewww.iste.co.uk/andro/libraries.zip
AninvestigationofcrowdsourcingprojectsappliedtolibrarieswascarriedoutbytheOCLC


[SMI11];itshowedthat,amongtheprojectsstudiedwhoseleadersweresoughtforthe
investigation,60%wereAmerican,19%Australian,10%Englishand5%NewZealander,and
only7%werefromothercountriesoftheworld.

Figure1.6.ChangeinthenumberofpublicationsoncrowdsourcingindexedbyGoogle
Scholarappliedtothedigitizationoflibraries.Foracolorversionofthefigure,see
www.iste.co.uk/andro/libraries.zip
Crowdsourcingappliedtodigitizationprojectsthereforeshouldnotbeconsideredapurely
Anglo-Saxonphenomenon.


Theinterestofscientificresearchworldwideinthephenomenonofcrowdsourcingisgrowing,
especiallywithregardtoitsapplicationstodigitizationoftheheritagepreservedinlibraries.
Thisstatementcanbesupportedbyobserving,forexample,thenumberofarticlesindexedin
GoogleScholaraboutthisspecificsubject.

1.2.Origin,definitionandscopeofcrowdsourcing
Crowdsourcinghaslongbeenapragmaticprofessionalpracticewellbeforeitwas
conceptualizedandbecameasubjectofacademicresearch.Undertheseconditions,itsorigin,
definitionandscopecanbedifficulttoestablish.Beforebecomingabuzzword,theterm
“crowdsourcing”wasfirstusedbyJeffHoweinthetitleofanarticlepublishedinWired
MagazineinJune2006,whichwasentitled“TheRiseofCrowdsourcing”.Accordingto[SCH
10],thetermhad,however,beenusedbyananonymousInternetuserinaforum.Otherauthors
prefertotalkabout“openwork”or“fair-tradework”.
Inthecaseofdigitallibraryprojectswhoseactualcontributorsareonlyanactiveminorityof
volunteersandcannot,inanycase,beassimilatedintoacrowd,certainauthorsprefertouse
thetermnichesourcingorcommunitysourcing,preferringthemorespecificword
“community”tothatofamoreindeterminate“crowd”.Itinvolvesnotsomuchusingthepublic
thanrecruitingvolunteersmotivatedbyaspiritofcollaboration,cocreationandcoconstruction.ThisideaisrelatedtotheonelaidoutbyJakobNielsen7,accordingtowhom
80%ofInternetusersarepassiveconsumersand20%areactivecontributorsandproducersof
contentontheWeb.AccordingtoHollyGoodier8,theseproportionswouldhavechangedsince
thenandwouldnowmorelikelybe25%ofInternetuserswhoareinactive,45%whocomment
andenrichand30%whoproducecontent.Whenitcomestodigitallibraries,theterm
communitysourcingseemsthemostjudicioustous.Weneverthelesswillusetheterm
crowdsourcing,whichismorecommon,willmakeourwritingmoreintelligibleandwill
allowustoavoidresortingtocomplexjargon.
Theauthorsof[EST12],whoseworkisauthoritative,havesoughttoworkspecificallyonthe
questionofthedefinitionofcrowdsourcingbycollecting,intheliterature,thediversityof
definitionsthatarefoundthere.Nolessthan40citationsin32articlespublishedbetween
2006and2011werecollectedinthisstudythathascategorizedthedifferentelements
necessaryfortheconstructionofasummarydefinition.


Table1.1.Multicriteriadefinitionsofcrowdsourcing
Whomakes
upthe
crowd?
Whatdoesthe
crowddo?
Whatdoesthe
crowdgetin
return?
Whoinitiates
it?
Whattypeof
processis
involved?
Whatmedium
isused?

Amateurs.

Itvoluntarilyandconsciouslyaccomplishestasksandmicrotasksinorderto
solveproblems.
Distraction,pleasure,thedevelopmentofskills,experiences,knowledge,the
sharingofknowledge,theloveofacommunity,economiccompensation,social
recognitionorbetterself-esteem.
Publicorprivatecompanies.
Aproductionprocess,aneconomicmodel,participativeoutsourcingofatask
afterarequestthatisopentoeveryone.
TheInternet.

Basedontheseelements,hereisthedefinitionwhichtheseauthorshavecomeupwith:
“Isatypeofparticipativeonlineactivityinwhichanindividual,aninstitution,anonprofitorganization,orcompanyproposestoagroupofindividualsofvaryingknowledge,
heterogeneity,andnumber,viaaflexibleopencall,thevoluntaryundertakingofatask.
Theundertakingofthetask,ofvariablecomplexityandmodularity,andinwhichthe
crowdshouldparticipatebringingtheirwork,money,knowledgeand/orexperience,
alwaysentailsmutualbenefit.Theuserwillreceivethesatisfactionofagiventypeof
need,beiteconomic,socialrecognition,self-esteem,orthedevelopmentofindividual
skills,whilethecrowdsourcerwillobtainandutilizetotheiradvantagethatwhattheuser
hasbroughttotheventure,whoseformwilldependonthetypeofactivityundertaken”.
[EST12]
ThequestionofthevoluntaryorinvoluntarynatureoftheparticipationofInternetuserscan
neverthelessbediscussed.Indeed,ifwebelievethatthecontributionisnecessarilyvoluntary
asthisdefinitionasserts,weexcludethefieldofcrowdsourcingonsitessuchasYouTube,
OCRcorrectionresultingfromreCAPTCHAandalargepartoftheprojectsthatcollect
contributionsofInternetusersintheformofgames(gamification).Ifwerecognizethatthis
contributionisnotnecessarilyvoluntary,thescopedefinitelyexpandsconsiderably.Inevery
case,excludingnotfullyconsciousformsofparticipationfromthefieldwouldatleastdeserve
justification,whichseemsdifficult.Maybeitisthereforepreferable,fromourpointofview,to
speakratherofexplicitcrowdsourcingwhenthecontributionofInternetusersisvoluntaryand
implicitcrowdsourcing(orinvoluntarycrowdsourcingorpassivecrowdsourcing)whenitis
not[HAR13].Renault[REN14b]alsoconsidersthisdefinitiontobesomewhatnaive,since
therearemanycontributorstocrowdsourcingwhoarenotawareoftheircontribution.


Tài liệu bạn tìm kiếm đã sẵn sàng tải về

Tải bản đầy đủ ngay

×

×