edinburgh research explorer · introducing the ablative framework chris speeda*, jon oberlanderb...

Edinburgh Research Explorer

Designing from, with and by Data: Introducing the ablativeframework

Citation for published version:Speed, C & Oberlander, J 2016, Designing from, with and by Data: Introducing the ablative framework. inProceedings of DRS 2016 International Conference: Future–Focused Thinking. Proceedings of DRS 2016,vol. 1, pp. 2991-3004, 50th Anniversary Design, Research, Society Conference: Future–Focused Thinking,Brighton, United Kingdom, 27/06/16. <http://www.drs2016.org/proceedings>

Link:Link to publication record in Edinburgh Research Explorer

Document Version:Peer reviewed version

Published In:Proceedings of DRS 2016 International Conference: Future–Focused Thinking

General rightsCopyright for the publications made accessible via the Edinburgh Research Explorer is retained by the author(s)and / or other copyright owners and it is a condition of accessing these publications that users recognise andabide by the legal requirements associated with these rights.

Take down policyThe University of Edinburgh has made every reasonable effort to ensure that Edinburgh Research Explorercontent complies with UK legislation. If you believe that the public display of this file breaches copyright pleasecontact openaccess@ed.ac.uk providing details, and we will remove access to the work immediately andinvestigate your claim.

Download date: 21. Feb. 2021

Copyright©2016.Thecopyrightofeachpaperinthisconferenceproceedingsisthepropertyoftheauthor(s).Permissionisgrantedtoreproducecopiesoftheseworksforpurposesrelevanttotheaboveconference,providedthattheauthor(s),sourceandcopyrightnoticeareincludedoneachcopy.Forotherusespleasecontacttheauthor(s).

Designingfrom,withandbyData:IntroducingtheablativeframeworkChrisSpeeda*,JonOberlanderbaCentreforDesignInformatics,UniversityofEdinburghbCentreforDesignInformatics,UniversityofEdinburgh*Correspondingauthore-mail:c.speed@ed.ac.uk

Abstract: This paper introduces a framework for designers in which existingmethodologiescanbeplacedinordertobetteracknowledgehowtheyworkwithdataindifferentways to support theirpractice.Thepaper startsbydistinguishing threekindsofvalueassociatedwithdata:(i)rawmeasurements;(ii)commercialandsocial;and (iii) moral and ethical. We then note that changes in computing andcommunications technologies serve tode-emphasisecomputersasdevices,and re-emphasisetheflowofdatabetweenpeople,machines,andthings;thus,wesharetheviewthathuman-datainteractionisakeychallengefordesigners.Inaddressingthechallenge,weintroducetheframeworkfordesignerstodistinguishdesignfrom,with,andbydata.Wenotethat informaticsprovidesthetheoryfor,andtechnologiesof,information processing, while design provides the methods to adapt and createproductsandservices.Thepaperusescasestudiestoillustrateourapproach.

Keywords:Design,Data,Informatics,Framework

1.Introduction.Designhasusedqualitativeandquantitativedatatoinformthedevelopmentofproducts,servicesandsystemsformanyyears.Frommarketanalyticstoobservationalanalysis,andquestionnairestodesignprobes,designersunderstandimplicitlytheneedtowatch,listenandlearnfromthedatathatisgatheredbyprototypesbeforeandduringthedesignprocess.However,whilstthemethodsforgatheringdatahavegrowntoreflectresearchthroughdesignapproaches,therehasbeenlittleclassificationofthekindsofdatathatweareencounteringinanageofbigdata,nortoframehowwedesignalongsideit.

Thispaperintroducesaframeworkfordesignerstoreflectontheirexistingmethodsofworkingwithdata,inordertoanticipateitsabilitytotransformdesignprocessasitslevelofperformativityincreases.Thepaperbeginsbyoutliningthreekindsofvaluethatdataisinvolvedinmediatingandthenestablishesacomplexityinwhichqualitativeandquantitativedatabecomesentangledacrosssocial,economic,moralandethicalvalues.The

Speed&Oberlander

2

secondpartofthepaperintroducesanemergingfieldofenquirythatsupersedesHumanComputerInteraction,thatofHumanDataInteraction(HDI).HDIdemandsthatseriousattentionisnowrequiredtoaddressthesystemsthatplacestressonconventionalethicalandmoralmodelsofhandlingpersonaldata.Ourpapertakesthismantleandproposesthatdesignersplayavitalroleinthedesignoffuturesystemsinwhichpeople,thingsandcomputersco-existintheproductionofdata.

However,inordertounderstandbetterhowtodesignalongsidedata,theauthorsgoontointroduceaframeworkforrecognisinghowexistingandemergingresearchmethodsaddresstheincreasingperformativityofdata.Thepapercloseswithreflectionsonthethreecasesofdesigningfrom/with/bydata,andthenexplorestheimplicationsfortheframework.

2.Datainvolvesatleastthreekindsofvalue.Acollectionofdatacanbethoughtofasasetofvaluesforsomevariables,acquiredoriginallybymeasurementsofsomekind.Underanappropriateinterpretation,datacountsasinformation,andinformationprocessingcanrefine(relatively)rawdataandmakeituseful,bycapturing,transformingandcommunicatingit.

Inthepast,andstilltoday,almostalldataisimpersonal;measurementsintheLargeHadronCollider,orintheSquareKilometerArrayaimtoprovideextraordinarynumbersofvaluesforvariableseveryday.Ofcourse,inthepast,atleastsomedatawaspersonal,asinpopulationcensuses.However,anincreasingamountofdataispersonal.Thatis,becausetheirpreferences,attitudesandbehaviourcanbemeasuredonlineinmanyways,peoplenowadaysgeneratelotsofdata,bothconsciouslyandunconsciously.This“bigdata”ofapersonalnaturecapturesaspectsoftheirbehaviourasconsumers,communicators,andashealthyorunhealthyphysicalandsocialbeings.

Sothefirstsetofvalues,thedatavaluesthataremeremeasurements,canbecomeentangledwithtwootherimportantkindsofvalue.

Thesecondkindofvaluearisesbecausebyaggregatinganykindofdataatscale,corporationsandagenciescangeneratenewcommercialorsocialvalue:theycancreateproductsandserviceswhichincreaseindividualorcollectiveutility,andwhichcanbemonetisedinatleastsomecases.

Thethirdkindofvaluearisesbecausethewaysinwhichcorporationsandagenciestreatallkindsofdata(butespecially,personaldata)reflectsasetofmoralorethicalvalues,including:theprotectionorviolationofprivacy;thepromotionorpreventionofreciprocityinrelationships;respectorrejectionofthecustomsandattitudesoflesspowerfulpeoples–suchastheirattitudestotime,diet,orsexuality;andtheenhancementorerosionoffairnessinsocietiesmostgenerally.

Designingfrom,withandbyData:Introducingtheablativeframework

3

3.ThefallofcomputersandtheriseofdataThepervasivenessoftheinternet,andofwirelessnetworking,haveenabledwidespreadadoptionofcloudcomputingservices.Forourpurposes,whatmattersaboutthecloudisthatitopensupgapsbetweentheplaceswheredataisgenerated,processed,andactedupon.Inthepast,thecapture,transformationandcommunicationofinformationmightallhavehappenedinoneplaceononematerialdevice:acomputerwithsuitableperipherals.Now,wefrequentlydonotcarewherethecomputingtakesplace.Inlightofthis,someaspectsofhumancomputerinteractionarebetterframedintermsofhumandatainteraction.Ifhumancomputerinteractionstudiesthewaysinwhichhumansinteractwith,andthrough,computers,wemightnowde-emphasisethematerialdevicesdoingthecomputing,andfocusmoreattentiononthewaysinwhichhumansinteractwith,andthrough,data.

Onegroupofresearchersconcernedwiththeprocessingofpersonaldatahavealreadyusedtheterm“humandatainteraction”tocoverthe“theindividualandcollectivedecisionsthatwemakeandactionswetake,asusersofonlinesystems,orassubjectsofdatacollectionpractices”(Mortieretal.2014).Theypointtotheneedto“makedataandanalyticsalgorithmsbothtransparentandcomprehensibletothepeoplethedataandprocessingconcerns”,andtogivepeople“thecapacitytoactwithinthesedatasystems,toopt-inortoopt-out,tocontrol,informandcorrectdataandinferences”.Onthisaccount,theproperstudyofhumandatainteractiongoeswellbeyondtraditionalinterestsindatavisualisation,toexploresocial,legalandethicalaspectsofpersonaldataprocessing.Thus,thethreekindsofvalueintroducedaboveareallimplicated.

Buttwoothertrendsrelatingtodataandinteractionareworthyofnote.Thecloudacceleratestheharvestingofpersonaldata,tobesure.Butitalsoenablesothernewdataflows,throughboththeInternetofThings,andsystemswhichsupportsocialcomputing.TheInternetofThings(IoT)is“thesetoftechnologies,systemsandmethodologiesthatunderpinstheemergingnewwaveofinternet-enabledapplicationsbasedonphysicalobjectsandtheenvironmentseamlesslyintegratingintotheinformationnetwork”(UKInternetofThingsSIGRoadmap,March2013).Socialcomputingiswheresocialbehaviourmeetscomputationalsystems.Itencompassescurrentonlinesocialinteraction,butalsogeneratespeople-poweredcomputation,withapplicationsfromonlineauctionstorecommendationsystems,fromelectionmonitoringtocitizenscience.

Mortieretal.’sconceptofhumandatainteractionisfocussedonpersonaldata,andtheproblemsandneedsassociatedwithit.TheIoTandsocialcomputingintroduceatleasttwonewoptions.ConsidertheIoT.First,weneedtointeractwithdata,andperhapswecanusethingstohelpusdothat.Butsecondly,wewillsometimesneedtointeractwiththethingsthemselves,andwewillthereforelikelyneedtotransformIoTdataintoformswithwhichwecaninteract.Thirdly,anynewinteractionwithdataorthingscanitselfgeneratefurtherdata,givensuitableinstrumentation.Finally,itwouldbenaturalforlevelsofaccesstodependuponontherolesindividualactorsplaywithrespecttocollectionsofthings.Inthese

Speed&Oberlander

4

respects,socialcomputingisanalogous:peoplecanusesocialcomputingsystemstointeractwithdata;theycanusedatatointeractwiththesystems;theirinteractionsgeneratefurtherdata;andwhattheycandowilldependupontheirroleinthelargersystem.

Drawingthesepointstogether,weseethattheyaresimplyfacetsofaworldofdistributedcomputinginwhichthecloudhelpsseparatethephysicalmechanismsofsensing,storing,processing,communicatingandactinguponinformation.Somemechanismsarelocal,othersremote.Somemechanismsareobviouslycomputers,otherslookjustlikethings,andyetothersarepeople.Thispicturemultipliesthenumbersandtypesofagentsatlooseintheworld,butitisobviousthatallthedataflowsandinformationprocessingarestillentirelysupervenientonphysicalmechanisms.Butsomeofthemechanismsareoutofsightofthepeopleinvolvedinthedataflows,andsoitisquiteunderstandablethattheydistinguishthematerial,visiblethingsfromtheimmaterial,andsometimesinvisibledataflows.

Someofthedatawhichpeopleinteractwithcanbeconsidered“researchdata”,inthesensethatitiscollectedtoinformthedesignofproductsandservices;atthesametime,sometimesdata(bigorsmall)isitselfamajorpartofaproductorservice.Intheformercase,themainpeopleinteractingwithdataaredesigners;inthelattercase,itisenduserswhodomostoftheinteracting(thankstothedesigners).Sodataplaysmultiplerolesindesignresearch.Moreover,theproblemsofhumandatainteractionidentifiedbyMortieretal.areimportant,buttheyarenotinfactspecifictopersonaldata;theyapplyalsototheotherdataflows,includingthoseinvolvingIoTdata,andsocialcomputingdata.Thisbeingso,howcantheseproblemsbetackledbydesignersoffuturesystemsofpeople,thingsandcomputers?

4.AframeworkfordesignersWithanestablishedhistoryinthedevelopmentofcreativemethodstowardthegatheringofempiricaldata,designershavemadesignificantcontributionstohowquantitativeandqualitativedatasupportamoreuser-centreddesignofproductsandservices.Howevertheadventofmobileandubiquitouscomputingpresentsthedisciplinewithamorecomplexarrayofdataformsthataremediatedindifferentwaysandassuch,theydemandthatwethinkabouthowdesignersdesignarounddata.Inlookingforameansofdistinguishingbetweentheformsofdatathatdesignersarenowfacedwithengagingwith,theauthorsidentifiedanincreaseintheperformativityofdata.Fromtypesofstabledatathatremainimmutable,throughdatathatistransformedwiththenetworksthatitisassociatedwith,todatathatisbeginningtoproduceitsowndata,thereisacontinuuminwhichdatabeginstospeakforitself(Cox2014).PerformativityisacomplextermthatDewsburydescribesas“thegap,therupture,thespacingthatunfoldsthenextmomentallowingchangetohappen.”(2000),andtraditionallyperformativityisusedtoexplainthecapacityofspeechandgesturestoactandofferemergentstructures.ThetermisattributedtothelanguagephilosopherAustinwhoestablishedthatwordscanbeusednotonlytodescribesomething,butcanusedtodosomething.Hismostpoignantexampleofwhathecoinedas

Designingfrom,withandbyData:Introducingtheablativeframework

5

‘performativeutterances’beingwhenweusethewords“Ido”toinstantiateanaction(suchasmarriage)(Austin1962).

Acknowledgingthatdataisstartingto‘do’things,weturnedtotheablativecaseinLatinthatindicatesanagent,instrument,orsourcewithinarelationshipexpressedby‘by’,‘with’,or‘from’.Ifdesignersarehavingtoadapttohowtheyderiveknowledgethroughdata,theablativecasemightbestdescribehowthedatathattheyareworkingwithisincreasinginitsperformativequalities.Byreversingthetraditionalablativecaseinwhich‘by’isgivenagency,‘with’isco-producedand‘from’istaken,itispossibletoexpresstheshiftinpracticesthatdesignershavebeguntodevelopasdatamovesfrombeingsomethinglikeasourcetodesign‘from’,toacomplexandfluidsettingtodesign‘with’,andfinallytoaconditioninwhichdesignisproduced‘by’dataitself.

4.1DesignfromdataDesignfromdata:whensystemsaredesignedbypeople,wheretheyareinspiredbymeasurablefeaturesofhumans,computers,things,andtheircontexts.

Therearemanymethodsthatdesignersusetoelicitdatafromsocial,technicalandenvironmentalsettings:fromestablishedethnographicmethodsfromuserobservations(Abrams2000,Stempfle2002andKawulich2005)andinterviews(Bernard2000,Byrne2001,Rubin2005);tomoredesignerlymethodsincludingculturalprobes(Gaveretal1999),technologyprobes(Hutchinson2002)andContextualmapping(Stappersetal2005).

CriticisedbyNormanifsolelyusedatthebeginningofadesignprocess(2006),userandparticipantobservationshelpdesignersgatherdatafrompeopleinspecificsituations.From‘flyonthewall’approachestotheuseofvideo,stillphotographandnotetaking,thegatheringdatafromcontextsinwhichpeoplearecarryingouteverydaypracticesorusingprototypes,isafamiliarmethodfordesignerstounderstandsocialpractices.Similarly,theuseofstructured,semi-structuredandun-structuredinterviewsalsooffersavaluablemethodtogatherdataabouttheperceptions,behaviourandopinionsofpeoplewhoareengagedintheconsumption,useorinteractionwithparticularproductsandcontexts.Whilstparticipantobservationandinterviewsareextendedfromestablishedethnographicmethods,culturalprobesandcontextmappingaremoreuniquetodesignanduseartefactsandmaterialstogatherdata.Packsconsistingofvariouselementssuchasdiaries,disposablecameras,postcardsanddrawingmaterialsthataredistributedtoprojectparticipants,encouragethemtodescribetheirexperienceswithoutthepresenceofthedesignresearcher.Useofgraphics,metaphorsandpersonalisedtouchescansupportparticipantstoofferimaginativematerialtoinspirethedesignprocess.Inthedevelopmentoftechnologyprobes,Hutchinsonetal.acknowledgehow“probeswillchangethebehaviourofourusers”(2002)andsubsequentlydevelopedaprobethatusestechnologytofosteraco-adaptiverelationshipwiththeuserinwhichthedeviceprovokesandpromotesinteractionsfromwhichunderstandingsofuseandcontextcanbeelicited.Explicitlynotaprototype,technologyprobesstimulateuseoveraperiodoftime,andallowresearcherstoreflecton

Speed&Oberlander

6

thisuseinordertogatherinformationabouttheusersaswellasinspireideasfornewtechnologies.Contextmapping,alsoadesigntechnique,usesaseriesofphasesthatbeginwiththecaptureofthedesigners’preconceptionsforasetting,followedbytheuseofavarietyofstimuli(includingquestionsandculturalprobes)tohelpparticipantsreflectonacircumstanceorsituation.Sessionsareusuallyrecordedtosupporttheidentificationofpatternsinlanguage,experienceandpractice.

Thevarietyofmethodsforgatheringdataisnotlimitedtothefourexamplesabovebutextendstoallprocessesinwhichdataisgathered‘from’settingsbeforebeinganalysedandusedtoinformsubsequentdesigndecisions.Throughthemulti-disciplinaryEquatorproject,agooddealwasestablishedabouttheappropriatewaysthatdatacanbegatheredandusedtoinformdesign.Hemmingsetal.listsevenstepstowarddesign:1.Planning;2.RecruitingParticipants;3.SelectingVolunteers;4.AssemblingDomesticProbes;5.DeployingDomesticProbes;6.RetrievingandAnalysingProbes,before7.SpeculativeDesign(2002).Thisorderofdatacaptureultimatelyendsinthestudio,wherethedesignercanlearnanddesign‘from’thematerials.

Figure1 TheHaggle-O-Tronwasdevelopedusingacombinationofdesignfromdatamethods

includingvideoethnographyandparticipantobservationthroughtheuseofatechnologyprobe.

AnexampleofhowtheauthorshavedevelopedaDesignFromDataapproachisintheirdevelopmentoftheHaggle-O-Tron(Speedetal2014).TheHaggle-O-TronisaninteractivekettlethatwasdevelopedforplacementwithinanOxfamsecondhandshoptoexplorehow

Designingfrom,withandbyData:Introducingtheablativeframework

7

haggling(apracticecurrentlyprohibitedinOxfamshops)mightbehelpfulinrevealingsecondhandgoods’financial,moral,social,andaestheticproperties.Visitorstotheshopwereinvitedtousethekettletohaggleoverthepriceofanarticlethattheywereinterestedinbuying.Amemberofthedesignteamwhowaslocatedintheshop’sbackroomandwasconnectedtotheHaggle-O-Tronviaawebcameraandmicrophone.This‘WizardofOz’techniqueallowedustosimulatethekettle’ssentience,inordertosustainarealistichaggle.Fromreviewingfootageandidentifyinginteractionsbackinthestudio,theresearchersgainedabetterunderstandingofbargainingtactics,theuseofincentivesandtheeffectivevocabularythatwouldsupportOxfam’swidercharitableprojects,whilstofferingthemaninsightintohowtheymightchangetheirin-storepolicies.

4.2DesignwithdataDesignwithdata:whensystemsaredesignedbypeople,wheretheytakeintoaccounttheflowsofdatathroughsystems,andtheneedtosustainandenhancehumanvalues.

Asthenetworksocietyhasdeveloped,ethnographyinturnhasdevelopedmeansofexpandingitspracticestoutilisesocialmedia,telecommunicationsandinternetcommunicationsinordertogatherdata.Virtualethnography(Hine2000),netnography(Kozinets2006),cyber-ethnography(Keeley-Browne2011)andonlineethnography(Wilson2002)allrefertoonlineresearchmethodsthathaveadaptedtraditionalethnographicmethodstostudyparticipantsthroughcomputer-mediatedsocialinteractions.Whilstthesemethodslargelygathermaterialandreport‘from’sourcesbeforeanalysis,easyaccesstoubiquitouscomputingtechnologiesisenablingresearcherstosustainalink‘with’aparticipantorcommunitytobetterunderstandhowdata-centricprototypes,productsandserviceshaveanimpactontheuser.Wedescribethisemergingresearchscenario,inwhichinformationcanflowinmorethanonedirection,asoneinwhichitispossibleto‘designwithdata’.

Theconstantconnectiontotheinternetbetweenproductssuchasmartphonesorservicessuchasenergythroughsmartmetersinhomes,istransformingtheindustryofdesign.Nolongeraredesignerssimplycontributingtostagesinavaluechainasaproductmovesfrommanufacture,packaging,distributiontoconsumption;designersareretainedtomediatethevalueofproductsandserviceswithinacomplexnetworkofsocialandenvironmentalconnections.CoinedbyNormannandRamirez(1994),theterm‘value-constellations’describestheeconomicsystemsthatemergedattheendofthe20thcenturyasglobalisationandnewtechnologiesinfluencedthewaythatvaluewassustained.Recognisingtheroleofco-createdvaluewithinnetworks,NormannandRamirezhighlightthat“successfulcompaniesconceiveofstrategyassystematicsocialinnovation:thecontinuousdesignandredesignofcomplexbusinesssystems”(1994).Withinavalue-constellation,thevalueofaserviceisconstantlymediatedaccordingtotheflowsofdatathatallowusersandstakeholderstosustainthevaluepropositionassociatedwithaproduct,serviceorexperience.Thesemoredynamicmodelsofvaluecreationandrelationrepresent

Speed&Oberlander

8

adifferentopportunityfordesigntoretainarelationshipwithusersthroughouttheirengagementwithproducts(Speed&Maxwell2015).Theopportunityfordesignersto‘designwithdata’thatisderivedfromtheinteractionsofusersenablesadifferentunderstandingofhowthefeedbackfromusercommunitiesaffectsthevalueofaproductorservice.

Figure2 Fiveinternetconnectedtoiletrollholderswaitingfordeployment,andascreencaptureof

datastreamedtotheinternetannotatedwithreflectionsfromoneowner.

Anexampleofhowtheauthorsareinvolvedindesigningwithdataarisesinthedeploymentoffiveinternetconnectedtoiletrollholdersthatfedbackdatatotheirowners.Thedesignoftheflowofdatawasrelativelysimple:eachdeviceconcurrentlymeasuredthemass(andhencelength)ofremainingtoiletpaper,andstreamedthevaluestoadesignatedrecipient.

ThedesignsolutionwasdevelopedforanInternetofThingsresearchprojectthatprovidesaplatformforownersofconnecteddevicestolayclaimtothedatathattheyproduceandbegintoexplorewaysinwhichtotradewithit.CurrentbusinessmodelsforIoTdevicesinvolvethecustomerpurchasingadevicethatsupportsparticularnetworkfunctions,butoftenstreamsdatabacktothemanufacturerwhomaysellthedatatothirdparties,oruseittoinformtheirowneconomicstrategies.TheHubofAllThingsproject(www.hubofallthings.com)seekstoprovideaplatformforpeopletomanagetheuseoftheirowndataandinturnidentifyvaluefromitbyeitherchoosingtoprotectit,shareitorpotentiallysellit.

OriginallyidentifiedbytheresearchteamasarelativelyeasyInternetofThingsdevicetodesign(comparedtofridgesandotherdomesticappliances),thetoiletrollisatthecentreofhighlypersonalpracticesthattakeplacebehindlockeddoorsandexemplifiesthetypeofpersonaldatathatpeoplemaywanttomanage.Throughthegraphthatisfedtoapersonaldatastoreandvisibleinabrowser,itispossibletoclearlyidentifyeventsthatusesignificantamountsoftoiletpaperfromwhichitisfurtherpossibletoinferparticulartoiletactivities;seeFig.2.Uponfurtheranalysis,thegraphalsorevealedaseriesoflesslikelyevents

Designingfrom,withandbyData:Introducingtheablativeframework

9

includingcleaningupaftercats,therunningoutoftoiletpaper,extrahouseguests,andsomebodyhavingarunnynose.Theperformativenatureofthedataemergesasfamiliesbegintointerpretthedatatoinferdomesticpractices,andinonecaseidentifythepresenceofastrangerinthehouse,whilstafamilywereawayonholiday.

Designingwithdataacknowledgesthatdataisnotacoldresourcetobetakenbacktothelaborstudioforexamination,butaconditioninwhichdesignersshouldanticipatethedisruptivepotentialthatisproducedfromstreamsoflivedatafromnetworkedartefacts.Trust,privacy,identityandsecurityareconceptsthatashumans,wedeterminethevalueofwithincomplexsocialandmaterialpractices.Inorderfordesignerstounderstandthebreachesanddisruptionsinvolvedinthehumandatainteractionsbetweeninternetconnectedthings,wewillneedtodevelop‘designwithdata’methodsinordertounderstandthevalueconstellationsthatareproducedandco-producedtosupportbettermanagement.

4.3DesignbydataDesignbydata:whensystemsaredesignedbyothersystems,largelyautonomously,wherenewproductsandservicescanbesynthesisedviathedata-intensiveanalysisofexistingcombinationsofhumans,computers,things,andcontexts.

Thefinalareaisintheemergingprospectthatdataitself,supportedbyanalgorithm,willbecomeadesigner.Suchacircumstanceisnotsofaraway,accordingtoGartner,whopredict:“By2017,asignificantdisruptivedigitalbusinesswillbelaunchedthatwasconceivedbyacomputeralgorithm.”(Gartner2014)

Thescaleofdatathatisbeingproducedandco-producedthroughmachinetomachineandmachinetohuman/humantomachineinteractionshasproventobeexponential.Ithasbeenobservedthatapproximately90%ofallofthedataintheworldhasbeenproducedinthepast2years(Arthur2013);whetherthisisinfacttruenow,theexactproportionisperhapsirrelevant.Astheflowofdatamovesfromwebbasedapplications,throughmobiledevicestonetworkedobjects,thedatathatisproducedbecomestheprimaryassetwithwhichtosustainthevalueofproductsandservices.Iftheinformationthatisderivedfromthedataandreturnedtotheuserdoesnotdemonstrategoodvalue,thentheusermaydroptheproduct.Inordertoidentifyvaluableinformation,machinelearningisbeingusedacrossawidevarietyofdatabasestoidentifypatternsinordertoelicitnewinsights(Bandyopadhyay&Sen2011).Designbydatasuggeststhatasthesealgorithmsbecomefasterandbetteratidentifyingnewopportunitiestosustainoraddvaluetoproductsandservices,itwon’tbelongbeforedata-drivenobjectsbegintobecomedesignerswithinourlives.

Speed&Oberlander

10

Figure3 TheThingTankprojectadoptsadesignbydataapproach.Camerasattachedtodomestic

objectsallowedhumanresearcherstoidentifyactivities,whilstdatafrominternetconnecteddevicesallowedmachinestolearnabouthumanactivity.Thiscombinationenabledtheresearcherstoidentifymorethanhumanactivitiesthatweredonewhilewaitingforthewatertoboil(fromakettle’sperspective).

Thecasestudythattheauthorshavecontributedtothatbestexemplifiesascenarioof‘designbydata’istheThingTankproject.Theprojectwasfundedtoexplorethepotentialforidentifyingnovelpatternsofusewithindatathatisstreamedthroughtheinteractionbetweenpeopleandthings,andthingsandthings.Throughanunderstandingofwhatdatacantellusabouthowweuseobjectsinpractice,theprojectpositedthatnewmodelsofusewouldemergeandreinvigoratetheroleofthingsandpeoplewithindesignandmanufacturing.Inthepast,manyInternetofThingsprojectshaveusedthenetworkconnectionofartefactstoidentifycostsavingandprocessefficiencies(e.g.,vehiclemanufacturers),ortotrackgoodswithinlargenetworks(e.g.,logisticscompanies),ortomonitorthehealthandsafetyofsystems(e.g.,aircraftmanufacturers).Suchprojectslookforregularpatternswithindatasetswhichsuggestefficienciesthatwillreinforcetheidentityofaproductorservicebymakingitsfunctioneasiertouseormoreeconomical.Bycontrast,theThingTankprojectproposedthatlookingforanomaliesandoutliersindatasetscouldsuggestmoreradicaldesignopportunities.Duringstudies,theresearchteamdevelopednon-anthropocentricmethodsbygatheringandstreamingdatafrombothmaterialobjectsandhumansthatwereinvolvedinadomesticrelationship,tobetterunderstandhowmachinescouldidentifypracticesthatwentunidentifiedbyhumanresearchers(Giaccardietal2016).

Althoughthemajorityofususeproductsasintended,manyofusalsoinventnovelusagesofobjectsbyadaptingorusingthemforunintendedpurposes.Byscanninglargedatasetsforevidenceofmis-useandthenusingthemtobuildnewassemblages,theThingTankproject

Designingfrom,withandbyData:Introducingtheablativeframework

11

proposesthatalgorithmsmayexploitdatatodesignthingsthathumandesignerscouldhaveneverhaveconceived.

5.ReflectionsandImplicationsCollectively,wetermthesethreeclassesofdesigningfrom,withandbydataasthe“AblativeFramework”fordesigninformatics,referencingtheablativegrammaticalcaseinLatin,whichisusedtocover“by,withorfrom”.Theflowofdata,andthegenerationofdifferingformsofvalue,arethecentralconcernsandallowdesignerstoreconfigureexistingpracticesandmethodstobetterunderstandtheincreasingperformativityofdata.Theframeworkseesdesignfromdataasestablishedmethodsfordesigners,anddesignbydataasstillhighlyemergent;whilstdesignwithdataistheimportantspaceofenquirythatrequiresurgentresearchtoaddressthefullextentofHumanDataInteractions.

TheFrameworkaimstoofferameansoforganisingbothexistingmethodsbutalsoofanticipatingemergingmethodsthatrecognisetheincreasingperformativequalitiesofdata.TheFrameworkisplacedwithinanetworksocietyinwhichdesignersareworkingalongsideawiderangeofdisciplinestomediatevaluewithinaconstellationofstakeholdersincludingalgorithms.TheFrameworkidentifiesdifferentrelationshipsbetweendesignersanddata,andhelpsusseewhentheuseofestablishedethnographicanddesignerlymethodsforgatheringdatafromisrequired,orwhenthesustainedflowsofdatarequireadesignwithdata.TheThingTankexampleinfactdemonstratesthisneatly:thedesignerspursuedtraditionaldesignmethods,andthenconstructedflowsofdatafromdevicesembeddedinusers’practicesandvalues,andthenengagedmachinelearningtoidentifyoutliers,whichpointstowardstheincreasingautomationofnewproductdesign.Sotheprojectinvolvedallthreerelationsbetweendesignanddata;theFrameworkdoesnotinstantiateahierarchyforthethreerelations,andacknowledgestheimportanceandinteractionofallthreewithindesignresearch.

Usesofthethreecasescanbeunderstoodintermsoftheneedforsomedesignprojectsthatdepartfromthestandarddoublediamondofdesign,withitspipelineoffourstages:discover,define,developanddeliver.Suchanapproachtypicallyidentifiesthebehavioursandconventionsthathavetobeobserved,andfindswaysofsustainingthem.Withtheadventofdesigningalongsidedata,thereislimitedchancetofreezethediscoveranddefinestages,becausedatawillcontinuetobereceivedfromusersandcommunitiesthatadjustthevaluepropositionoftheproductorservicethathasbeendelivered.

Designisadjustingfromprovidingservicesthataddvaluealongthetraditionalvaluechain,towardsplayinganactiveroleinthemediationofvaluewithinaconstellationinwhichdataprovidesfeedback,oreventakescontrol.Frameworkssuchastheoneproposedhereprovidetoolswhichhelpusunderstandwhichmethodstoadopt,andwhen.Someofthosetoolsshouldbeverysimple,takingtheformofchecklistsforpractitionersthatrespondtothefollowingquestions:

Speed&Oberlander

12

1. Incontextsinwhichhumans,computersandartefactsareincloseinteraction,howcandesignersidentifymeasurablefeaturesfromwhichdatacanbeelicitedtobetterunderstandthevaluesinplay,andhowcantheydesigninterventionstocapturedatainamannerthatissensitivetohumanvalues?

2. Incontextsinwhichdataisflowinginsuchawaythatitisperformative,informingandaffectingthebehaviorofhumansandartefacts,howcanthedesignteamdevelopsystemsthatcapturetheexistingflows,andofferinterventionsthatsupportandenhancehumanvalues?

3. Incontextsinwhichsystemsaredesignedbyothersystems,howcandesignersmediatethedevelopmentofproductsandservicesthataresynthesisedbydataprocesses,toensurethatthevaluesofthesystemsarecommensuratewiththevaluesofthehumanandmorethanhumanparticipants?

Infuturework,weaimtoexercisetheseprinciplesandmakethembroadlyavailablefordesignresearch.

1. Acknowledgements:Designresearchforthispaperoriginatedfromthefollowinggrants:TheInternetofSecondhandThingsprojectfundedbyanEPSRCgrant(EP/K012819/1)andOxfamUK;TheHubofAllThingsprojectfundedbyanEPSRCgrant(EP/K039911/1);andtheThingTankprojectfundedbytheSkoltechFoundation,Russia.Thanksgotothemanyparticipantswhosupportedtheresearch,andwearegratefultoourreviewersfortheirconstructivesuggestionsforimprovingthepaper.

5.ReferencesAustin,J.L.(1962)HowtoDoThingswithWords.Oxford:ClarendonPress.

Bandyopadhyay,D.&Sen,J.(2011)InternetofThings:ApplicationsandChallengesinTechnologyandStandardization,WirelessPersonalCommunications:AnInternationalJournalarchive,58(1),49-69.

Bernard,R.(2000).SocialResearchMethods:QualitativeandQuantitativeApproaches.ThousandOaks,CA:Sagepublications.

Cox,G.(2013)SpeakingCode:CodingasAestheticandPoliticalExpression.MITPress,Cambridge,MA.

Dewsbury,J.(2000)PerformativityandtheEvent:EnactingaPhilosophyofDifference.EnvironmentandPlanningD:SocietyandSpaceAugust200018:473-496,

Norman,D.(2006)Whydoinguserobservationsfirstiswrong.Interactions13,4(July2006),50-ff.

Hemmings,T.,Clarke,K.,Crabtree,A.,Rodden,T.andRouncefield,M.(2002)Probingtheprobes,Proceedingsofthe2002ParticipatoryDesignConference,pp.42-50,ComputerProfessionalsforSocialResponsibility.

Hutchinson,H.,Mackay,W.,Westerlund,B.,Bederson,B.B.,Druin,A.,Plaisant,C.,Beaudouin-Lafon,M.,Conversy,S.,Evans,H.,Hansen,H.,Roussel,NandEiderbäck,B.(2003)Technologyprobes:inspiringdesignforandwithfamilies.InProceedingsoftheSIGCHIConferenceonHumanFactorsinComputingSystems(CHI'03).ACM,NewYork,NY,USA,17-24.

Designingfrom,withandbyData:Introducingtheablativeframework

13

Hine,C.(2000)VirtualEthnography.London:Sage.

Gartner(2014)GartnerRevealsTopPredictionsforITOrganizationsandUsersfor2015andBeyond.Garner.Weblink:http://www.gartner.com/newsroom/id/2866617Accessed10thNovember2015.

Giaccardi,E.,Speed,C.,Caldwell,M.&Cila,N.(2016)WhenObjectsBecomeCo-ethnographers:PotentialsofaThing-CentredApproachinDesignandAnthropology,in(Eds)Binder,T.,Kjærsgaard,M.&CharlotteSmith,R.DesignAnthropologicalFutures,Bloomsbury.

Kawulich,B,B.(2005)ParticipantObservationasaDataCollectionMethod.ForumQualitativeSozialforschung/Forum:QualitativeSocialResearch,[S.l.],v.6,n.2,May.

Keeley-Browne,E.(2011).Cyber-Ethnography:TheEmergingResearchApproachfor21stCenturyResearchInvestigation.InG.Kurubacak,&T.Yuzer(Eds.)HandbookofResearchonTransformativeOnlineEducationandLiberation:ModelsforSocialEquality(pp.330-238).Hershey,PA:InformationScienceReference.doi:10.4018/978-1-60960-046-4.ch017

Kozinets,RobertV.(2006),Netnography2.0,inHandbookofQualitativeResearchMethodsinMarketing,ed.Belk,R.W.Cheltenham,UNandNorthampton,MA:EdwardElgarPublishing,129-142.

leCompte,M.&Schensul,J.(1999).EssentialEthnographicMethods(vol.2):EthnographersToolkit.NewYork,NY:Altamirapress.

Miller,D.,andSlater,D.(2000).TheInternet:Anethnographicapproach.Oxford;NewYork:Berg.

Mortier,R.,Haddadi,H,.Henderson,T.,McAuley,D.&Crowcroft,J.(2014)Human-DataInteraction:TheHumanFaceoftheData-DrivenSociety.AvailableatSSRN:http://ssrn.com/abstract=2508051orhttp://dx.doi.org/10.2139/ssrn.2508051

Normann,R.&Ramirez,R.(1994)DesigningInteractiveStrategy:FromValueChaintoValueConstellation.JohnWiley&Sons.

Speed,C.,Hartswood,M.,Laurier,E.,Magee,S.,deJode,M.&Hudson-Smith,A.(2014)TheHaggle-O-Tron:designinterventioninsecondhandretail.Proceedingsofthe2014companionpublicationonDesigninginteractivesystems.Pages137-140.ACMNewYork,NY,USA.ISBN:978-1-4503-2903-3doi:10.1145/2598784.2602802

Speed,C.&Maxwell,D.(2015)Designingthroughvalueconstellations.interactions22,5(August2015),38-43.DOI=http://dx.doi.org/10.1145/2807293

Wilson,S.M.,&Peterson,L.C.(2002).TheAnthropologyofOnlineCommunities.AnnualReviewofAnthropology31:449–467.

AbouttheAuthors:

Chris Speed is Professor of Design Informatics within EdinburghCollege of Art, University of Edinburgh. Chris is Co-Director of theDesign Informatics Research Centre that is home to researchersworking across interaction design, temporal design, anthropology,softwareengineeringandcryptocurrencies.

Jon Oberlander is Professor of Epistemics within the School ofInformatics,UniversityofEdinburgh.Heworksongettingcomputers

Speed&Oberlander

14

to talk like individual people, so he studies how people expressthemselves and develops systems that can adapt themselves topeople.