rea2: a unified formalisation of the resource-event-agent...
TRANSCRIPT
REA2: A unified formalisation of the Resource-Event-Agent Ontology
LAURIER, Wim, KIEHN, Jesper and POLOVINA, Simon <http://orcid.org/0000-0003-2961-6207>
Available from Sheffield Hallam University Research Archive (SHURA) at:
http://shura.shu.ac.uk/19016/
This document is the author deposited version. You are advised to consult the publisher's version if you wish to cite from it.
Published version
LAURIER, Wim, KIEHN, Jesper and POLOVINA, Simon (2018). REA2: A unified formalisation of the Resource-Event-Agent Ontology. Applied Ontology, 13 (3), 201-224.
Copyright and re-use policy
See http://shura.shu.ac.uk/information.html
Sheffield Hallam University Research Archivehttp://shura.shu.ac.uk
1
Title page
Title
REA2:AUnifiedFormalisationoftheResource-Event-AgentOntology
Authors
1.WimLaurier([email protected])
UniversitySaintLouis,Bruxelles
BoulevarddujardinBotanique43,1000Bruxelles,Belgique
AND
GhentUniversity,FacultyofBusinessAdministration,DepartmentofBusiness
InformaticsandOperationsManagement
Tweekerkenstraat2,9000Gent,Belgium
AND
inno.comInstitute
Heiststeenweg131,2580Beerzel
2.JesperKiehn([email protected])
3.SimonPolovina([email protected])
ConceptualStructuresResearchGroup,CommunicationandComputingResearchCentre½DepartmentofComputing,SheffieldHallamUniversity,Sheffield,UKS12NU
2
3
Abstract (200 words)
ThroughaproofofconceptinSWI-Prolog,thispaperdemonstratesabusiness
transactionmodelbywhichthetradingpartnerscanderivetheirown,personal
perspectivefromshareddata.Thedemonstrationisaninnovativeformalisationofthe
Resource-Event-Agent(REA)ontologyasitallowsforswitchingviewpointsinreal-time
betweenonetrading-partner’sperspectiveandthatofatrading-partnerwithan
opposingview(i.e.customerorsupplier),oratrading-partnerindependentperspective
(e.g.trustedthird-party).ThebusinesstransactionmodelisachievedbyunitingREA
withtheOpen-EDIBusinessTransactionOntology(OeBTO).Theresultingunified
formalisationoftheREAontology(REA2)alsohighlightsimplicationsforthefuture
developmentofa)enterpriseinformationsystems(EIS)inthecloud,b)social-media-
basedEIS,c)blockchainEIS,andd)EISinteroperabilityacrossbusinessparadigms.The
EISinteroperabilitysuchasbetweentraditionalEIS(whichtypicallyusesatrading-
partnerperspective),andEISforthecollaborativeeconomy(whichtypicallyusesa
trading-partnerindependentperspective)isparticularlyhighlightedasitbecomesmuch
moretransparentthanpreviously.
Keywords
Resource,Event,Agent,REA,Ontology,Prolog,DependentView,IndependentView,Open-ediBusinessTransactionOntology(OeBTO),CollaborativeEconomy,blockchain
4
Body of text (divided by subheadings)
Introduction
Businessescollaboratewithpartnerstocreatevalueandcompetewiththosesame
partnerstoearnthelargerpartofthevaluetheycreatedtogether(Brandenburger&
Nalebuff,1996).Collaborationrequiressharinginformationwithtrading-partners,
whereasinformationsharingmighthampermaximisationofearningsinacompetitive
environment(Letaifa,2014).Meanwhile,thirdparties(suchasgovernments,trusted
thirdparties)requestaccesstotradedatasothattheycanprocessthisinformationto
reducehealthandsafetyrisksthroughfood-traceability,e-customs,orinnovelformsof
businessinteractionssuchasblockchains(Laurier&Poels,2012;Steiner&Baker,2016;
Tan,Niels,Klein,&Rukanova,2010).Traditionalenterpriseresourceplanning(ERP)
andaccountinginformationsystems(AIS)arelimitedintheircapabilitytoprovidethis
third-partyorvaluenetworkperspective,duetotheirinternalfocus("ValueNetwork
Software(NRP),"2016).
Toaddresstheseissues,businessontologyneedstobeinaformthatisconciseand
simpleenoughtobeimplementedinthemultitudeofmodellingandprogramming
languagesthatreflectthetechnologiesthatthetrading-partnersorthirdpartiesmaybe
using.Atthesametimetheontologyhastopermitaninteroperable,simultaneous
representationoftheopposingperspectivesofcollaboratingtrading-partnersaswellas
aneutralthird-partyperspectiveonbusinesstransactionsandoperationstosupport,for
example,traceability("ValueFlowsvfvocabs,"2016).
5
Thepaperisstructuredasfollows:Thenextsectionoutlinesthecapabilityofthe
Resource-Event-Agent(REA)ontologyforformalisingbothatrading-partnerdependent
andtrading-partnerindependent(orneutral)perspective.Thethirdsectionpresentsa
unifiedREA-ontologycalledREA2thatformalisesboththedependentandindependent
dimensionsinoneview.Thefourthsectionaddressestheresearchmethodologyand
relatedresearch.Thefifthsectionintroducesthemeta-modelthatextendsthe
traditionalinterpretationoftheREAontology.Subsequently,thesixthsection
introducesthesemanticpatternformalisedasadata-model.Theseventhandeighth
sectionspresentanexamplethatisusedastheexemplarthatactsastheproofof
conceptforREA2.Theninthsectiondiscussesthelimitationsandimplicationsofthe
proofofconcept.Thisdiscussioniscomplementedwithdirectionsforfutureresearch,
andconcludeswiththevaluethatREA2bringstoREA.
Resource-Event-Agent Ontology
TheResource-Event-Agent(REA)isanontologythathasalreadybeenusedtodocument
boththeperspectiveofatrading-partnerandthatofathird-party(Geerts&McCarthy,
2002;ISO/IEC,2007).REAthusremainsasthefocalontology.Thispaperargues
howeverthattheREAontologyhasthecapabilityofmodellingatrading-partner
dependentandindependent(orneutral)perspectivesimultaneouslyifandonlyifthe
REAprimitivesareassignedanidentifiabletrading-partner.
GivenREA’sorigininaccounting,REA’sdependentview,whichisalsocalledthetrading-
partnerview,isenterprise-centric.Inonedimension,REAfocussesonthesemantics
requiredfordescribingtheperspectiveofasingleorganisation(McCarthy,1979,1982).
Inanotherdimension,REA’sindependentview,whichisalsocalledhelicopterview,is
6
supply-chaincentricandfocussesonthesemanticsfordescribingbusinessfromthe
perspectiveofaneutralthird-party(Haugen,2007).Thevalidityoftheindependent
viewhasbeenrecognisedbytheInternationalOrganisationforStandardisation(ISO)as
astandardforeffectiveelectronicdatainterchange(EDI)(ISO/IEC,2007).Inpractice,
thedependentviewofREAhasmanyimplementationsinnumerousfields.Ithasbeen
mappedtootherenterprise-centricontologies,andhasbeenidentifiedasarecurring
patterninthemarket-leadingSAPERPsoftware(Gailly&Poels,2009;Geerts,2011;
O'Leary,2004).ThereisthusaneedtobringanindependentviewtoREAthatthan
elegantlyco-existwiththeseexistingenterprise-centricimplementations,bringingthem
intolinewiththeISOstandard.
REA2
Tomeettheabove-identifiedneed,thispaperpresentsaunifiedREA-ontologycalled
REA2thatformalisesboththedependentandindependentviews.Fig.1visualisesthe
intendedapplicationcontextoftheREA2formalisation,inwhichdataavailablein
ISO/IECFDIS15944-4oracompatibleview-independentdataformatsupports
interoperabilitybetweentheinformationsystemsoftrading-partnersharinga
collaborationspace.SuchacollaborationspaceisdefinedbytheISOOeBTOstandardas
abusinessactivityspaceinwhichaneconomicexchangeofvaluedresourcesisviewed
independentlyoftheperspectiveofanytrading-partner(ISO/IEC,2007).ISO/IECFDIS
15944-4isoneoftheearliesandstillanessentialsourceforREA’sindependentview.
<<InsertFig.1here>>
7
Fig. 1. REA2 view integration: application context
Intheintendedapplicationcontext,thetrading-partnerswouldcontinuetouseREA’s
dependentview(oracompatibleview-dependentdataformatsuchasSAP(O'Leary,
2004),sincedatainanISO/IECFDIS15944-4compatibleformatwouldrequirenoora
trivialtransformation).REA2willbeshowntoallowforanautomatedtransformationof
view-dependentdataintoview-independentdataandviceversa.Thisautomated
transformationwillallowtrading-partnersfromtheirdependentviewperspectiveto
sharetheirdatainthecollaborationspace,wherethetransformationallowsthemto
reusethedatasharedbytheircounterparttradingpartners,whichmaintaintheir
counterpartdependentview.
8
Toavoidreadabilityissueswiththevisualisations,theREA2formalisationhasbeen
decomposedinandpresentedas3separatemodels:1)ameta-model(M2),2)adata-
model(M1),and3)anexample(M0)thatactsastheexemplarthatdemonstratesthe
valueofREA2.Toassistunderstanding,theinstantiationsemanticsthatrelatethese
modelstoeachotheraccordwiththeM2-M1-M0semanticsoftheObject-Management
Group’s(OMG)Meta-ObjectFacility(MOF)((OMG),2016).
Themeta-model(M2)usesOntoUMLstereotypes,topresentaminimalsetofREA
constructsthatallowforasimultaneousrepresentationoftheopposingperspectivesof
trading-partners—i.e.dependentview—andtheviewpointofanindependentthird-
party—i.e.independentview.ThegeneralisationsemanticsintheREA2meta-modelwill
beusedtoderivetheindependentviewfromthedependentview.OntoUMLisof
interestasithasbeendevelopedtocountersilo-basedinformationarchitectures,which
hamperansweringcriticalquestionsthatcanonlybesolvedbyconnectinginformation
thatisscatteredoverthesesilos.(Guizzardi,2014).Thesecriticalquestionsincludethe
traceability,profitabilityandequitabilityquestionsthatrequireaview-independent
perspective.As this paper focusses on ontological aspects, all the models below abstract
from context-specific aspects such as cardinalities that model business rules (e.g. the upper
and lower boundaries for the number of economic-agents that can be related to an economic
event). Consequently, no cardinalities will be shown on the diagrams in this paper. As the
REA ontology addresses a pattern in conceptual modelling, it requires us to focus on the role
and relationship semantics of the pattern, inherently abstracting from the kinds of things that
can play the resource, event and agent roles. This abstraction and subsequent focus on roles is
essential to the contribution of this paper.
9
AssistedbyMOF,thedata-model(M1)implementsthemeta-modeltoshowasemantic
patternthataddressestrading-partnerdependentandindependentsemanticsina
collaborationspaceasdescribedearlier.Thissemanticpatterncaninturnbeusedto
mapthedependentviewtotheindependentviewandviceversa.Extensionsof,and
variationson,thisarchetypaldata-modelcanbefoundinmodel-drivenbusiness
patternsasdocumentedbyHrubyetal.,(2006).ThisM1modelisanarchetypal
implementationoftheREApatterndescribedbythemeta-model(M2).This
implementationdoesnotrequireaparticularsyntax.Consequently,itusesastandard
UMLsyntax,instantiatingthestereotypedclassesofthemeta-modelasassociations
withanREAassociationstereotype,beingclasseswithanREArolestereotypeanda
namereferringtoakind.TheprincipleisalsoappliedtotheclassesintheM1diagram
(i.e.fig.5),whichareassignedM2levelREArolestereotypes(i.e.instancesoffig.2-4)
wheretheirnamesrefertoinstancesofakind.
Theexemplarillustratesthearchetypalimplementationscenariowithexampledata
(M0)thatweredevelopedfortestingthestatementsthataddressthesemanticchanges
inducedbytheswitchingbetweenthedependentandindependentviewpoints.The
exemplarwillbediscussedindetailshortly.Fornow,wecanexplainthatthetest
scenariogeneratesindependentviewstatementsfromthedataofasingletrading-
partnertakingthedependentviewi.e.thetrading-partner104infig.6.Thegenerated
independentviewstatementsarethentransformedbackintodependentview
statementsfortrading-partnerswithaviewopposingthatofthetrading-partnerof
whichthedatawereusedtogeneratetheindependentviewdatai.e.thetrading-partner
106–104’ssupplier–and107–104’sclient–infig.6.
10
Research Methodology & Related Research
Tothebestofourknowledge,thispaperisthemostrecentinahistoryofpapersthat
endeavourtoelucidatethefinersemanticsoftheREAontology.Thebest-known
originatefromtheanalysisofREAinthelightofSowa’sfoundationalontology(Sowa,
2000),withonefurtherstudythatsupplementsSowa’sontologywithabstraction
mechanismsinREA(Geerts&McCarthy,2000b,2002,2005,2006;Sowa,2000).A
numberofauthorshaveformalisedREAtomakeitcomputer-readable,whilstnumerous
othershavetriedtofindthebestdefinitionsforREAconceptsthatmosthelpthehuman
endeavourinapplyingREA(Gailly&Geerts,2013;Gailly,Laurier,&Poels,2008;Hruby
&Kiehn,2006;Ito&Vymětal,2013;Jaquet,2006;Laurier&Poels,2014;WilliamE.
McCarthy,GuidoL.Geerts,&GrahamGal,2016;W.E.McCarthy,GuidoL.Geerts,&
GrahamGal,2016;Zdravkovic,Zikra,&Ilayperuma,2011).
Ourapproachforthispaperinscribesitselfinthedesignsciencetradition,whichhas
beenthedominanttraditioninREAresearch(Geerts,2011).Designscienceinvolvesa
relevance,arigorandadesigncycle.Therelevancecyclecontextualizesthedesign
artefactbyreferringtoapracticalproblemintherealworld.Thispracticalproblem
informstheevaluationoftheartefact.Therigorcycleaddressesthefoundationsand
innovativeaspectsofthedesignartefactreferringtothestateoftheart.Thestateofthe
artisdefinedbyscientificandprofessionalliteratureandpresentedastheexpertiseof
specialistsinvolvedintheartefact’sdesign.Thedesigncycleproducesdesignartefacts
throughaniterativeprocessofradicalinnovationandcontinuousimprovement,
informedbytherelevanceandrigorcycle(Hevner&Chatterjee,2010)Hevner&
Chatterjee,2010�).
11
Asthedesign,rigorandrelevancecyclearetypicallynotaddressedsimultaneouslyin
theREAliterature(Dunn,Gerard,&Grabski,2016),ourpaperwillfocusontherigor
cycleaddressingthecurrentstateoftheREAontologyknowledge.Toaminorextent,we
willalsofocusonthedesigncyclebypresentinganartefact—namelytheexemplar—
thatwillserveasourproofofconceptandallowforascenario-basedanalysisofour
approachpresentedshortly.Thispaperaddressesresearchrigourbyrefactoringthe
Prolog1codefoundinGeerts&McCarthy(2000a)withnotionsfromtheREA
formalisationdevelopedbyGaillyetal.(2008).Thatformalismisstereotypedinturn
withOntoUMLstereotypes(Guizzardi,2014)insteadofOWLstereotypes,andnotions
fromtheoriginalResource-Event-Agent(REA)datamodelpublishedbyMcCarthy
(1982)andHrubyetal.’s(2006)bookonREA.Theproofofconceptaddressesthe
designcyclethroughademonstrationoftheapproachthroughtheexemplarbeingthe
archetypalexample.TheexemplarisasetofstatementswritteninPrologableto
transformdependentviewPrologstatementsintoindependentviewPrologstatements
andvice-versa.Toemphasisethispaper’sembeddingintheREAliterature,thecodein
appendixCandEalsoshowshowanopposingandindependentperspectivecouldbe
derivedfromGeerts’andMcCarthy’s(2000a)code.ThesereferenceolderREAartefacts
butnonethelessillustratethispaper’sembeddingofREA’sdesignsciencetradition;the
illustrationisaccordinglyreinforcedbyaninformedargument.
The Meta-Model
Themeta-modelextendsthetraditionalinterpretationoftheREAontology.For
readability,themeta-modelwillbevisualisedasthreeseparatefiguresabstractingfrom
cardinalityconstraints,eachfocusingonarelatordefinedbyaREAaxiom.A relator
1 Prolog is a logic-based programming language (Colmerauer & Roussel, 1993)Colmerauer & Roussel, 1993�).
12
mediates between the roles in a material relation. A material relation relies on the relata,
which are the entities that are being related, irrespective of their intrinsic properties (e.g. Bob
likes Alice). A formal relation is reducible to intrinsic properties of the relata (e.g. Bob is
taller than Alice) (NEMO, 2015).
Fig.2addressesthestock-flowaxiom—i.e.“At least one inflow event and one outflow event
exist for each economic resource; conversely inflow and outflow events must affect
identifiable resources.” (Geerts & McCarthy, 2005). Stock-Flows are shown as relators. The
relata in fig. 2—i.e. economic-resource, economic-event and its subtypes—have all been
stereotyped as rolemixins. A rolemixin is a generalisation of a set of roles, and is depicted in
the fig. 2 as roleMixin. A role is defined as an anti-rigid sortal type that is connected to a
characterising relation and specialises a unique kind (Guizzardi, 2014). Through this
generalisation, rolemixins can be assigned to instances of different kinds through a joint
specialisation of the kind and rolemixin, where roles can only be assigned to instances of a
single kind though specialisation of this kind. Therolemixineconomic-resourcecanbe
assignedtoanentity(orthing)thatisscarceandiscontrolledbyatrading-partner,to
whomithasvalue(e.g., a right, services and a good can all be economic-resources).
(ISO/IEC,2007)Bydefininganeconomic-resourceasarolemixin,thispaperdefendsthe
opinionthatthingscanonlybeconsideredaneconomic-resourcewhentheyare
valuabletoatrading-partner.Forexample,acaronlyhasvaluetothepersonthatcan
driveit.So,mycarisaresourcetome,thecarnexttoitintheparkinglotisnota
resourcetome-despitebeingaresourcetoitsowner-evenifitskind(e.g.VWGolf)
identicaltothatofmycar.Thisinterpretationofthetermeconomicresourcediffers
significantlyfromthetraditionalinterpretationasthetraditionalinterpretationcanbe
readas:aneconomicresourceisanythingthathasthepotentialofbeingconsideredan
13
economicresourcebyatleastoneperson.Consequently,anyobjectofthekindVWGolf
isaneconomicresourceinthetraditionalinterpretation.Therolemixineconomic-event
canbeassignedtoaphenomenonthatreflectschanges2inthevalueorthequantityof
economicresources(McCarthy,1982).Bydefiningeconomic-eventsasrolemixins,this
paperarguesthatsomekindsofeventcanpotentiallyhaveaneconomicimpactwithout
havingto.Forexample,astormcanhaveaconsiderableeconomicimpactwhenit
damagesgoods.However,astormoftheexactsamekindcanhavenoeconomicimpact
atall(e.g.,whencrossinglandwithnoeconomicvalue).Moreover,eventsthathavean
economicimpactbydefinitiondonotneedtohaveaneconomicimpactoneveryperson
thatcanobservethem.Forexample,lookingatacashtransfer(e.g.,fromathird-party
perspective)doesnotmakemeanyricherorpoorer,althoughthetrading-partners
involvedwillobservetheeconomicimpactofthecashtransfer.Thisperson-bound
interpretationofthenotioneconomiceventisalsorelatedtothispaper’scontribution
asthetraditionalinterpretationofeconomiceventisabsoluteandshouldbereadas“an
eventthathasthepotentialofhavinganeconomicimpactonatleastoneperson”.
Fig. 2 reads the “inflow event” and “outflow event” in the stock-flow axiom as an event
playing the increment event role in an inflow and an event playing the decrement event role in
an outflow respectively, as inflow and outflow are considered subtypes of stockflow by Gailly
etal.(2008)Asstatedbythestock-flowaxiom,theeconomiceventrolemixinhastwo
sub-types.Theincrement-eventrolemixinisconnectedtoaninflowrelator,wherethe
decrement-eventrolemixinisconnectedtoanoutflowrelator,bothinflowandoutflow
aresub-typesofthestock-flowrelator.Theincrementanddecrementrolesare2 In the accounting literature this change is always instantaneous. Even when this change is part of a lengthy process, a single temporal part – typically the first or last – of this process is labelled as the change. For transfers this implies that ownership is transferred either at the start or at the end of the transport.
14
consideredmutuallyexclusiveinasingleperspective—i.e.intheuniverseofdiscourse
ofasingletrading-partneraneconomic-eventisassignedeitheranincrementor
decrementrolebutnotboth.
Fig. 2. The REA2 meta-model for stock-flows
Withinthecontextofthispaper,thesecondREAaxiom—i.e.“All events effecting an
outflow must be eventually paired in duality relationships with events effecting an inflow and
vice-versa.”(Geerts & McCarthy, 2005)—hasbeenmodelledasincrementand
decrementeventrolemixinsthatareconnectedtoadualityrelator.Bymodellingduality
asarelatorconnectingtwoevents,thispapersubscribestotheinterpretationofa
dualityasasocialconstruct(i.e.amaterialrelationship),whichprevailsinaccounting
(Fisher-Pauzenberger&Schwaiger,2017).Forexample,whenstealingacarwearenot
defeatingthelawsofphysics,butmerelybreakingcivillaw,whichisasocialconstruct,
asitisphysicallypossibletotakeacarwithouthavingtogiveanythinginreturn.Inthe
universeofdiscourseofasingletrading-partner,adualitypairsanincrement-eventtoa
⌧roleMixin�economic-resource
⌧roleMixin�economic-event
⌧relator�stock-flow
⌧roleMixin�increment-event
⌧relator�inflow
⌧roleMixin�decrement-event
⌧relator�outflow
15
decrement-event.Thisunilateraldefinitionofdualityasperceivedbyasingletrading-
partnerrelatestothedefinitionofavalue-interfaceinthee3valueontology,which
modelsthedualityconceptfromtheperspectiveofeachactor—i.e.trading-partner
(Gordijn,2002).Asincrementanddecrementeventsarerolemixins,theycanbe
assignedtodifferentkindsofevents(e.g.,transfersandtransformations),showninfig.
3.Moreover,asdualitiesaddvalueforeachofthetrading-partnersinvolved,theadded
valueforeachtradingpartnercanbeconsidereda“quaindividual”.Thisaddedvalue
inheresintherecipientanddependsontheprovider.Forexample,inanexchangeof
cookiesformoneybetweenElmoandCookiemonster,forCookiemonstertheadded
valueoftheexchangedependsonElmotransferringthecookiesandinheresin
Cookiemonstervaluingthecookiesmorethanthemoneythatisspent.FromElmo’s
perspective,theaddedvalueoftheexchangedependsonCookiemonstertransferring
themoneyandinheresinElmovaluingthemoneymorethanthecookiesthatare
transferred.Inthecaseofatransformationduality–asopposedtothetransferduality
describedabove–thisaddedvalueinheresintheownervaluingtheproducedresources
higherthantheconsumedresources.
16
Fig. 3. The REA2 meta-model for duality
Thekindsofthings(entities)thattheeconomicresourceandagentrolemixinscouldbe
assignedtowouldbeaninterestingside-discussion.However,tomaintainthefocusof
thispaper(andasreflectedininfig.3)wecentreourattentiononthekindsofevents
thattheeconomic-eventrolemixincouldbeassigned,asitisrelevantforformalising
trading-partnerviewpoints.Theincrementanddecrementeventrolemixinscanbe
assignedtoeithereventsofthetransformationortransferkind.Atransformationadds
valuebychangingresourceproperties,whereasatransferaddsvalueinamarket
⌧roleMixin�economic-event
⌧roleMixin�increment-event
⌧roleMixin�decrement-event
⌧relator�duality
⌧kind�transfer
⌧role�take
⌧role�give
⌧relator�transfer-duality
⌧kind�transformation
⌧role�produce
⌧role�consume
⌧relator�transformation-duality
17
transactionbetweentrading-partners(W.E.McCarthyetal.,2016).Theproducerole
specialisesboththekindoftransformationandtheincrement-eventrolemixin.The
consumerolespecialisesboththetransformationkindanddecrement-eventrolemixin.
Asaresult,onlytransformationeventscanbeassignedproduceandconsumeroles.
Whentransformationsareinstantaneous,produceandconsumerolescanbeassignedto
thesameevent.Forexample,whenblendingfruitsformakingasmoothie,all
ingredientslosetheiridentitysimultaneously(i.e.consume)andtheimmediateresult
(i.e.produce)oftheblendingprocesscanbeidentifiedasasmoothie.However,when
describingacomplexprocesswithdiscretesteps,produceandconsumerolescanbe
assignedtodifferentevents.Forexample,whenaddingraisinstodoughtomakeraisin
bread,theraisinslosetheiridentitywhenbeingadded(i.e.consume)althoughtheyonly
changethepropertiesofanintermediatestate(i.e.dough)oftheproduct,whilethefinal
product(i.e.raisinbread)willonlybeobtainedattheendofthebakingprocess(i.e.
produce).Thesamereasoningappliestoaddingextrasaltorbuttertothedough,or
seasoningit.InREA,suchacomplexprocessismodelledasaconfigurationofproduce
andconsumeeventsinterlinkedwithtransformationdualities.Theinterlinkedconsume
andproduceeventscanbediverseinnature(e.g.anumberofinstantaneousconsume
eventscombinedwithaproductionprocesswithaconsiderableduration,andan
instantaneousproductionofaby-productattheend).
Continuing,thetransformation-dualityrelatorspecializesthedualityrelator,connecting
aproduceroletoaconsumerole.Ontheotherhand,atransfer-dualityrelator
specialisesthedualityrelatorbyconnectingataketoagiverole.Thetakerole
specialisesboththetransferkindandincrement-eventrolemixin,whereasthegiverole
specialisesthetransferkindandthedecrement-eventrolemixin.
18
Finally,thethirdREAaxiom—i.e.“Each exchange needs an instance of both the inside and
outside subsets.” (Geerts & McCarthy, 2005)—addresseshowaparticipationrelator3
connectsaneconomic-agenttoaneconomic-event.Theeconomic-agentrolemixincanbe
assignedtoanaturalorlegalpersonparticipatingineconomicevents.Economic-agents
suchasemployeescanparticipateineventsasinside-partyonbehalfofthemselvesor
anothereconomicagent.Inaninside-partyrelation,economicagentsdealwithanother
person(legalornatural)inatrading-partnerrole.Inatrading-partnerrole,aneconomic
agentexperiencestheeffectofaneconomicevent.Alleconomicagentsthatrelatetoan
economic-eventthroughanoutside-partyrelatoraretrading-partners.Theeventkindin
whichtheyparticipateisatransferbetweentrading-partners.Fromandtoarethe
subtypesofoutside-partyrelator.Atrading-partnerconnectedtoafromrelatorassigns
atakeroletoatransferevent.Atrading-partnerconnectedtoatorelatorassignsagive
roletoatransferevent.Fig.4summarisestheabove.
3 This participation relator represents an REA participation, not an OntoUML participation, which is a formal relationship that holds between an event (economic or not) and all objects involved in it.
19
Fig. 4. The REA2 meta-model for participations
AswithREA’seconomic-resourceandeconomic-eventprimitives,thispaperarguesthat
REA’seconomic-agentshouldbemodelledasrolemixinsassignedtoaperson(i.e.a
kind).Adiscussionconcerningthekindsofpersonsthatcanorcannotbeassignedan
economic-agentroleisconsideredoutsidethescopeofthispaper.Aconsequenceof
modellingeconomic-agentsasrolesisthatthesameordifferentpersonscanplay
multipleinsideandoutsideroles.Forexample,whenIbuypizzatoJoe’sPizza,Icanact
onbehalfofmyselfasinsideandoutsideagent.Inthatcase,IamJoe’strading-partner
andinvolvedinbothtransferevents(i.e.paymentanddelivery)throughaninside-and
outside-partyrelation.IcouldalsoaskmydaughtertopickupthepizzaandpayJoe.In
thatcase,IamstillJoe’strading-partner,butthedeliveryandpaymentareexecutedby
mydaughteronbehalfofme.Consequently,I’mlinkedtobothtransfersthroughan
mirror⌧formal relation�
⌧roleMixin�economic-event
⌧roleMixin�economic-agent
⌧relator�participation
⌧roleMixin�trading-partner
⌧relator�inside-party
⌧relator�outside-party
⌧role�take
⌧relator�from
⌧role�give
⌧relator�to
20
outside-partyassociationandmydaughterthroughaninside-partyassociation.
Similarly,Joecoulddeliverthepizzadirectlytomeordecidetosendapizza-boy.Inthe
formercase,Joeisbothmytrading-partner(i.e.outside-party)andtheperson
participatinginthepaymentanddelivery(i.e.inside-party).Inthelattercase,Joeisstill
mytrading-partner(i.e.outside-party)bythetransfereventsareexecutedbythepizza-
boy(i.e.insideparty)onbehalfofJoe.
TheREA2ontologyformalisationasindicatedearlierallowsforasimultaneous
documentationofboth—i.e.sellerandbuyer—perspectivesonanexchange.AsREA2
alsoaccountsfortheindependentview,besidesbothdependentviews,thesymmetry
betweentheuniversesofdiscourseofbothtrading-partnersneedstobecapturedbythe
formalisation.Inthemeta-model,thissymmetryiscapturedbytheformalrelation
namedmirror,akintoamirrorimage,whichisareflectedduplicationbutisreversedin
directionjustlikeinamirror.Amirrorisshowninfig.4.Aformalrelationisarelation
thatcanholddirectlybetweenitsrelataandisreducibletointrinsicpropertiesofthe
relata,suchas‘olderthan’.Themirrorrelationaddressesthefactthatatransferalways
involvesastock-flowfromonetrading-partnertoanother.Theformalmirrorrelation
formalisesthatfromandtorelatorsaretwosidesofthesamecoin,thecoinbeingthe
transfer.Forexample,whenJoesendspizzatoacustomer,itisatransferfromhimto
thecustomer.NopizzacanbesentbyJoewithoutgoingtoacustomer,norcanany
customerreceivepizzawithoutitbeingshipped(i.e.from).Thesamelogicappliesto
moneytransfers.Nopaymentcanoccurwithoutapersonspendingthemoney(i.e.from)
andapersonreceiving(i.e.to).
21
Collaboration Space: A semantic pattern Whereasthemeta-modelinfigs.2,3and4formalisestheconceptsoftheREAOntology
usingOntoUMLstereotypes,thesemanticpatterninfig.5simultaneouslyvisualisesa
buyer,sellerandindependentperspective4.Theacquisitioncycleinblue5showsthe
buyerperspectiveonaneconomicexchange,whichtypicallyinvolvesapurchaseofraw
materialsfromavendorpairedindualitywithacashdisbursementtothevendor.The
revenuecycleinred6,ontheotherhand,showsthesellerperspectiveonaneconomic
exchange,whichtypicallyinvolvesasaleoffinalproductstoacustomerpairedin
dualitywithacashreceiptfromthatcustomer.
REAalsosupportstheconversioncycle,whichtransformsproducts(i.e.rawmaterials)
acquiredthroughtheacquisitioncycletoproductsthatcanbesoldthroughtherevenue
cycle(McCarthy,2003).Besidesthepresenceofatransformation-dualityinthemeta-
model,theconversioncycleisanotherside-discussionthatisthusnotaddressedinthis
paper.Thestereotypesinfig.5relatedata-modelconstructstothemeta-model.To
embedfig.5intheREAliterature,eventsfollowthenamingconventionsoftheISO
OeBTOstandard7ontheindependentviewandthetrading-partnerperspectiveonthese
eventsfollowsthenamingconventionsofMcCarthy’swork8.
4 Each of these perspectives should be seen an additional viewer-specific semantic layer on top of a shared objective reality, where this paper explicitly abstracts from modelling the objective reality. 5 The blue refers to Cookiemonster in W.E. McCarthy’s iconic slides on exchanges (McCarthy)McCarthy) 6 The red refers to Elmo in W.E. McCarthy’s iconic slides on exchanges (McCarthy)McCarthy) 7 Fig. 11 of (ISO/IEC, 2007)ISO/IEC, 2007) 8 Fig. 7 of (McCarthy, 1982)McCarthy, 1982(W. E. McCarthy et al., 2016)�)(W. E. McCarthy et al., 2016)
22
Fig. 5. Collaboration space semantics
Theformalrelationmirrorpredicatesthatanopposingviewexistsoneachstock-flow
connectedtoatransferinagiveortakerole.Itshouldbenotedthattheindependent
viewdoesnotallowforthesesimultaneousinflow-outflowsemantics,asthesamestock-
flowisaninflowforonetrading-partnerandanoutflowfortheother.Consequently,
viewindependentgeneralisationsofstock-flowshavebeenincludedinfig.5.Adelivery
hasasaleandpurchaserolethatcannotbeseenfromanindependentperspective,and
cannotbeseensimultaneouslyfromadependentperspective.
Fromthebuyer’sperspective,thedeliveryeventisaninflowofproducts;intheseller’s
eyesthissamedeliveryisanoutflowofthesameproducts.Thedeliveryroleconnected
toaproductinflowiscalledapurchase,whereasthedeliveryroleconnectedtoa
productoutflowiscalledasale.Apaymenthasacash-disbursementandcash-receiptrole
⌧stock-flow�
⌧stock-flow�
⌧inflow�
⌧outflow�
⌧transfer-duality�
⌧inside�
⌧inside�
⌧to�
⌧from�
⌧stock-flow�
⌧stock-flow�
⌧inflow�
⌧outflow�
⌧transfer-duality�
⌧inside�
⌧inside�
⌧to�
⌧from�
mirror
mirror
⌧economic-resource�cash
⌧economic-resource�product
⌧transfer�payment
⌧give�buyer::cash-disbursement
⌧take�seller::cash-receipt
⌧transfer�delivery
⌧take�buyer::purchase
⌧give�seller::sale
⌧trading-partner�buyer::seller
⌧trading-partner�seller::buyer
23
thatareinvisiblefromanindependentperspective,andcannotbeseensimultaneously
fromadependentperspective.Thepaymentisaninflowofcashfromtheperspectiveof
theseller—i.e.cash-receipt—andanoutflowofcashfromtheperspectiveofthebuyer—
i.e.cash-disbursement.
Fig.5highlightstwodistincttransfer-dualityrelatorsbetweenadeliveryandapayment
intheirrespectivegiveandtakeroles—i.e.oneforeachperspective.Fromthebuyer
perspective,thedeliveryisperceivedasaninflowofproducts—i.e.purchase—thatis
dualtothepayment,whichisperceivedasanoutflowofcash—i.e.cashdisbursement—
whilethesellerperceivesthemasadualsale—i.e.outflow—andcashreceipt—i.e.
inflow.Sincethedualityaxiomrequiresinflowandoutflowsemantics,itdependsonthe
perspectiveofatrading-partner.Thecomplianceofdatawiththedualityaxiomthus
cannotbeverifiedwithoutreferencetotheperspectiveofaspecifictrading-partner
(Geerts&McCarthy,2000a,2005).
Intheindependentview,thesemanticsoftherelatorsareindependentofthe
perspectiveofthetrading-partners.Rather,theyrepresenttheperspectiveoftheentire
tradecommunityoranindependentthird-partynottakingpartinthetransaction(such
asasupplychainmanager,governmentagency,regulatorybodyor—indeed—anentire
tradingcommunity).Wherethedependentviewfocusesonthedirection—i.e.inor
out—ofresourceflowsforasingletrading-partners,theindependentviewdistinguishes
twokindsofparticipationrelators—i.e.toandfrom—thatsignalthedirectionofthe
stock-flowsconnectedtotransfers,whichareconsequentlydeprivedoftheirinflowor
outflowsemanticscomparedtothedependentperspective(ISO/IEC,2007).
Accordingly,weneedtoswitchbetweenbothperspectives.
24
The Exemplar Theexemplarthatwillbeusedtodemonstratehowthemodelsintroducedaboveenable
computerstoproducetheindependent,buyerandsellerperspectiveisnowpresented
indetail.Althoughswitchingbetweenthebuyer,sellerandthird-partyscriptsis
intuitiveandeasyforhumans,itislessobviousforcomputersandalgorithms.The
exemplarwillrevealhowthischallengeisovercome.Forthispurposeandeaseof
understanding,wewillrefertothedataofasingletrading-partnerinthesupplychain.
Theexampledataareassembledintotheexemplarthatisthepaper’sexampleandproof
ofconcept.
Theexemplarandproof-of-conceptapplicationwerecodedinSWI-Prolog.Thecodecan
befoundintheappendix.Sincethecodecoverstheentirevaluechain,itismore
elaboratethanthePrologcodeinGeerts&McCarthy’s(2000a)intensionalreasoning
paper,whichexclusivelycoverstherevenuecycle.Thetestscenariowillbeintroduced
andvisualisedintheremainderofthissection.
Theexemplarshownbyfig.6coversthedataoftrading-partner104,whoassignsREA
rolemixinsasmodelledinfig.5tothings(entities)101to120.Thetestscenario
containsacquisition,revenueandconversioncycledata.Theacquisitioncycleconsists
oftrading-partner106,delivery102,payment113,cashaccount114andproduct101,
allseenfromtheperspectiveofagent104.Therevenuecycleconcernsproduct112,
delivery115,trading-partner107,payment117andcashaccount114.Theconversion
cycleconsistsofconsumeevent105,product101,produceevent110,andproduct112.
25
Fig. 6. The evaluation scenario
Thenewlyintroducedformalrelationmirrorandthesemanticpatternpresentedinfig.
5allowforderivingthevendorperspectiveontrading-partner104’sacquisitioncycle
fromthedatasharedby104.Fig.6revealsthattrading-partner106willperceive
trading-partner104asabuyer,whichisformalisedasthetag“106::buyer”(i.e.
trading-partner104isabuyertotrading-partner106).Thesemanticpatternalso
predicatesthatevent102,whichisapurchasetotrading-partner104—i.e.tag
“104::purchase”—isasaletotrading-partner106—i.e.tag“106::sale”—whereas
event113willbecategorisedasacash-receiptbytrading-partner106—i.e.tag
“106::cash-receipt”—asitisacash-disbursementtotrading-partner104—i.e.tag
“104::cash-disbursement”.
Fig.6alsoshowstheperspectiveoftrading-partner107ontrading-partner104’s
revenuecycle.Trading-partner107willperceivetrading-partner104asaseller—i.e.
tag“107::seller”.Sale115asperceivebytrading-partner104—i.e.tag
⌧from�
⌧transfer-duality�
⌧to�
⌧to�
⌧from�
{104::inflow,106::outflow}⌧stock-flow�
⌧outflow�
⌧inside�
⌧transformation-duality�
⌧inside�
⌧inflow�
{104::outflow,107::inflow}⌧stock-flow�
⌧from�
⌧stock-flow�
{104::outflow,106::inflow}
⌧stock-flow�
{104::inflow,107::outflow}
⌧transfer-duality�
⌧to�
⌧to�
⌧from�
106:trading-partner{104::seller}
102:delivery{104::purchase,106::sale}
113:payment{104::cash-disbursement,106::cash-receipt}
104:trading-partner{106::buyer,107::seller}
101:product
105:transformation{104::consume}
110:transformation{104::produce}
112:product
115:delivery{104::sale,107::purchase}
114:cash
117:payment{104::cash-receipt,107::cash-disbursement}
107:trading-partner{104::buyer}
26
“104::sale”—willbeperceivedasapurchasebytrading-partner107—i.e.tag
“107::purchase”.Economicevent117willbeperceivedasacash-disbursementby
trading-partner107—i.e.tag“107::cash-disbursement”—asitisacash-receipt
fortrading-partner104—i.e.tag“104::cash-receipt”.
InanREAindependentviewscript,whichtakestheperspectiveofanindependentthird-
party,trading-partnerscannotbeassignedcustomerorvendorsemanticsasthose
dependonatrading-partner’sperspective.Consequently,inthesemanticmodel
derivedfromagent104’sdata,agents104,106and107aretobecategorisedastrading-
partners.Trading-partneristhestereotypeassignedtobuyerandsellerinfig.5.
Similarly,event102and115arecategorisedasdeliverieswheretheirdependentview
semanticsaresaleandpurchase,whicharesubtypesofdeliveryinfig.5.
Correspondingly,event113and117areassignedpaymentsemanticsinthe
independentview,whiletheyarecashreceiptsanddisbursementsintheirrespective
dependentviews.
Equivalently,itisimpossibletoassigninflowandoutflowsemanticstostock-flows
connectedtotransferroles—i.e.giveandtake—as,intheindependentview,asingle
transferisconnectedtoaninflowforonetrading-partnerandanoutflowfortheother.
Sincetheindependentviewalsorequiresinformationabouttheflowofresources
(typicallyoppositelydirectedproductsandmoneyflows)betweenthefromandto
participations,itcanconveytheinflowandoutflowsemanticsthatcharacterisethe
dependentview.Ifatrading-partnerroleconnectsatransfertoaninflowinhis/her
27
dependentview,resourcesareflowingtowards(to)him/her,whereresourceswillbe
flowingawayfromhim/herwhenhe/sherelatesthetransfertoanoutflow.
Consequently,delivery102entailsastock-flowofproduct101fromtrading-partner106
totrading-partner104.Thedualeventisapaymentinvolvingacashtransferfrom
trading-partner104totrading-partner106.Thendelivery115resultsinastock-flowof
product112fromtrading-partner104totrading-partner107.Therequitingpaymentis
cashtransfer117fromtrading-partner107totrading-partner104.Werethecase
exemplarshowstraditionalproduct-forcashexchanges,thesamelogiccanbeappliedto
bartertrade(i.e.productforproductexchange)andmoneyformoneyexchanges,since
transferdualitiesalsoapplytodelivery-deliveryandpayment-paymentpairs.
Table1summarisesthecaseexamplarvisualisingallsemanticsassignedtoanobjectin
differentviews.Itshowsthatinthecaseexemplar,resourcesemanticsareuniversalfor
theentiretradecommunity.Consequently,nospecificdependentviewsemanticswere
assignedtoresources101,112and114.Althoughviewdependentsemanticsexist(e.g.
raw-material,final-productforproducts)adiscussionwasirrelevantforthispaper’s
contribution.Asaresultofthisdecision,resources101and112arerecognisedas
productsandresource114ascashbytheentiretradecommunityincludingtrading-
partners104,106and107.Ontheotherhand,economic-eventsemanticsdiffer
accordingtothetrading-partnerperspectivetaken.Forexample,delivery102isa
deliveryfortheentiretradecommunity,butalsoapurchasefortrading-partner104
aloneandasalefortrading-partner106alone.Economic-agentsalsoreceive
perspectivedependentsemantics.Forexample,trading-partner104isseenasatrading-
partnerbytheentiretradecommunity,butadditionallyasabuyerbytrading-partner
28
106andasasellerbytrading-partner107.Inturn,trading-partner104willperceive
106and107asatrading-partnersliketheentirethetrade-community,butalsoasa
sellerandbuyerrespectivelyinhisparticularperspective.InREA’sdependent
perspectivetrading-partnersarenotawareofthemselvesastheyarenevermodelled
explicitly.However,since104ispartofthetradecommunity,heknowsthathe’sa
trading-partner.Whenapplyingthelogicpresentedabove,104caninferthatheisa
buyertotrading-partner106inthecontextofexchange103-112andasellertotrading-
partner107inthecontextofexchange115-117.Themissingobjectnumberswere
assignedtoinsideagentsinamoreelaborateversionofthiscaseexemplar.Theywere
omittedinthisversiontofocusontheactualcontributionofthispaper.
Object Independent 104’s view 106’s view 107’s view
101 Product
102 Delivery Purchase Sale
104 Trading-partner Buyer Seller
105 Transformation Consume
106 Trading-partner Seller
107 Trading-partner Buyer
110 Transformation Produce
112 Product
113 Payment Cash-
disbursement
Cash-receipt
114 Cash
115 Delivery Sale Purchase
29
117 Payment Cash-receipt Cash-
disbursement
Table1:Summaryofthesemanticsoftheevaluationscenario.
The Proof of Concept: Prolog code ThePrologcodeinappendixformalisestheinferencerules(appendixA),tests
(appendixD)andcounterpartsoftheexemplarintroducedabove(appendixB).Wenow
introducethestatementsandinferencerulesthatarerequiredforderivingthe
independentviewfromdependentviewdataandan(opposing)dependentviewfrom
independentviewdata.Thetranslationprocesstotrading-partnerviewtothe
perspectiveofatrading-partnerwithanopposingviewismediatedbytheindependent
view,whichissemanticallyequivalent.InthePrologcode,thistranslationprocessis
decomposedintwophases.First,trading-partnerviewdataaretranslatedto
independentviewdata.Inthiscase,trading-partner104’sdata(appendixB)aremade
viewindependentbytheinferencerulesinappendixA.Appendix1containsasetof
rea_latticepredicatesthatisequivalenttotheinheritanceandinstantiation
semanticsinfigures2to5.Thesepredicatesarethenusedtoassignindependent-view
semanticstothedependentviewdatainappendixBthroughis_astatements.Through
thisprocess,theobjectsoftheexemplarareassignedbothtrading-partnerand
independent-viewsemantics.Subsequently,thetrading-partnerviewsemanticsare
hiddenandtheindependent-viewsemanticssharedasoebto_roleand
oebto_relatorpredicates.Thefilteringtechniquethatisusedtohidetrading-
partnerviewsemanticscouldbegeneralisedtoprotectinformationthatprovides
competitiveadvantage(e.g.,bysharingthecomponentsofaproduct,withoutsharing
thestepsoftheproductionprocess).Theindependent-viewoebtostatementarethen
convertedtotrading-partnerviewstatementsbydependent_role,
30
dependent_relator,agent_patternandevent_patternstatements.The
opposingviewsoftrading-partner106and107obtainedthroughthestatements
describedabovearethentestedinappendixD.
Theis-asemantics—i.e.inheritanceandstereotypes—foundinfigs.2-4havebeen
summarisedinasemanticlattice(i.e.apartiallyorderedsetofcategories(Sowa,2000)
).Thelatticeisusedtoderiveindependentviewsemanticsfromdependentviewdata.
Theis-arelationsareformalisedaspredicates,ofwhichasubsetisshownbelowincode
snippet1,thecompletesetofstatementscanbefoundinsectionA.4andA.5ofappendix
A.Theis-aargumentintherea_latticestatementshasbeenaddedforclarity.Ithasno
computationalvalue.Thefirstlineincodesnippet1specifiesthateveryincrement-
eventisaneconomicevent.Thesecondlinespecifiesthateveryproduceisanincrement
event,andhenceaneconomicevent.Thisstatementisindentedbecauseitmodelsa
secondlayerofthesemanticlattice.Theindentationhasnocomputationalvalue.
rea_lattice(increment-event,is_a,economic-event). rea_lattice(produce,is_a,increment-event). rea_lattice(produce,is_a,transformation). rea_lattice(take,is_a,increment-event). rea_lattice(take,is_a,transfer). rea_lattice(delivery,is_a,transfer). rea_lattice(sale,is_a,delivery). rea_lattice(sale,is_a,give). rea_lattice(purchase,is_a,delivery). rea_lattice(purchase,is_a,take).
Code Snippet. 1. As-is statements in the semantic lattice
TheinferencerulesinsectionA.1.,shownincodesnippet2,resultinis-astatements.
SubsectionA.1.1.containsthecodethatprocessesthetransitivityofis-asemanticsat
thetypelevel,whichcanbefoundintheinthesemanticlattice.Forexample,ifa
rea_latticestatementpredicatesthatapurchaseisadelivery(i.e.
31
rea_lattice(purchase,is_a,delivery).)andanotherstatementstatesthata
deliveryisatransfer(i.e.rea_lattice(delivery,is_a,transfer).),thena
purchaseisalsoatransfer(i.e.is_a(purchase,transfer)).Thefirstinferencerule
underA.1.1.istheidentity(e.g.everyeconomic-eventisaneconomic-event).Thesecond
statementoperationalisesis-asemanticswiththestatementsofwhichcodesnippet1is
asubset.
/*A.1. Inference Rules*/ /*A.1.1. Navigating the REA Ontology Primitives and their Super- and Sub-Types*/ is_a(X,X). is_a(X,Y):- rea_lattice(X,is_a,Y). is_a(X,Z):- is_a(X,Y), rea_lattice(Y,is_a,Z). /*A.1.2. Inherited Semantics*/ is_a_role(X,is_a,Z,to,P):- rea_role(X,is_a,Y,to,P), is_a(Y,Z). is_a_relator(X,is_a,Z,to,P):- rea_relator(X,is_a,Y,to,P), is_a(Y,Z).
Code Snippet. 2. Processes the is-a statements in the semantic lattice
SubsectionA.1.2.showsthecodethatcombinesthesetype-levelis-asemanticswiththe
is-asemanticsofinstantiationsuchasshownbyfig.6.Forexample,ifitisgiventhat102
isapurchase(i.e.rea_role(102,is_a,purchase,to,104)),then102mustalso
beadelivery(i.e.rea_role(102,is_a,delivery,to,104)),andatransfer(i.e.
rea_role(102,is_a,transfer,to,104)).Thelattertwopredicatescanbe
derivedfromthefirstrea_rolestatementwiththehelpoftheinferencerulesunder
A.1.2,whichcombinethefirstrea_rolestatementswithis-astatements.For
computationalreasonsthisassignmenthadtobeoperationalisedastwoseparate
statements(i.e.oneforrolesoneforrelators).
32
TheinferencerulesinsectionA.2,shownincodesnippet3,usetheis-astatementsof
sectionA.1.2.toderivetheindependentviewfromgivendependentviewdata.Thecode
insectionA.2.1.definesavocabularyfortheindependentviewasasubsetofthe
rolemixinsandrelatorsincludedinthetherea_lattice.Thisvocabularyisalsoasubsetof
thetermsfoundintheOeBTOstandard(ISO/IEC,2007).
/*A.2. The Open-edi Business Transaction Ontology*/ /*A.2.1. The OeBTO vocabulary as a subset of REA*/ oebto_vocabulary([product,cash,delivery,payment,trading-partner,stock-flow,transfer-duality,from,to]). oebto_interface(X):-oebto_vocabulary(L), member(X,L). /*A.2.2. Defining the visibility of OeBTO compliant REA roles and relators*/ oebto_role(X,is_a,Y,to,Z):- is_a_role(X,is_a,Y,to,P), oebto_interface(Y), access(Z,to,P,data). oebto_relator(X,is_a,Y,to,Z):- is_a_relator(X,is_a,Y,to,P), oebto_interface(Y), access(Z,to,P,data).
Code Snippet. 3. Derives the independent view from the dependent view
SectionA.2.2.containstheoebtostatementsthatgivespersons(suchascustomer,
supplier,government)accesstoasubsetofatrading-partner’sdata.Thissubsetis
constrainedbythevocabularydefinedinA.2.1.andtheaccessisrestrictedthroughthe
accessstatement.Inabsenceofanaccessstatement,apersonhasnoaccesstoatrading-
partner’sdata.Whenapersonhasaccess,heonlyseesthesemanticshehasaccessto.
Forexample,event102isapurchaseto104,sincepurchaseisnotpartoftheOeBTO
vocabularydefinedinA.2.1.,thetermdeliverywillbesharedwithatrading-partnerthat
hasaccessto104’sdata(i.e.oebto_role(102,is_a,delivery,to,106)),as
33
deliveryisasuper-typeofpurchase(i.e.
is_a_role(102,is_a,delivery,to,104))andpartofthevocabularydefinedin
A.2.1.(i.e.oebto_interface(delivery))Thethirdandfinalconditionconstrains
theaccessoftrading-partnerstodata.Althoughthedeliverysemanticsofentity102can
besharedinthecollaborationspace,trading-partner007willnothaveaccesstothe
datainabsenceofanexplicitaccessstatement(i.e.access(007,to,104,data)).As
withappendixsectionA.1.2.thefiltersforrolesandrelatorshavebeenoperationalised
separatelyforcomputationalreasons.
ThestatementsinsectionA.2.3.,shownincodesnippet4,addfromandtostatements
involvingthemodeller,sincethemodellerisnotself-awareinthedependentview,and
requiredintheindependentview.Theaddedfromandtostatementscomplywiththe
formalrelationfrom-mirrors-toshowninfig.4.TheCodeinA.2.3.mirrorsallfromand
tostatementsidentifiedbythemodeller.Forexample,iftrading-partner106is
connectedtopurchase102throughafromrelatoraccordingtomodeller104(i.e.
rea_relator([102,106],is_a,from,to,104)),then104willbeconnectedto
102throughatorelator(i.e.oebto_relator([102,104],is_a,from,to,106)).
Foraviewindependentexample,entity106canbereplacedwithentity000.The
statementsinsectionA.2.3.1.formalisetheformalrelationmirror.Thestatements
underA.2.3.1.definethesymmetricnatureofthemirrorsrelation.
/*A.2.3. Mirror (Formal relation)*/ oebto_relator([X,Y],is_a,Z,to,P):- access(P,to,Y,data), rea_relation(Alpha,mirrors,Z), rea_relator([X,_],is_a,Alpha,to,Y). /*A.2.3.1. Opposing Semantics in Opposing Views*/ rea_relation(from,mirrors,to).
34
rea_relation(to,mirrors,from). Code Snippet. 4. Adds the modeller in de independent view
TheinferencerulesinsectionA.3,shownincodesnippet5,derivedependentviewfrom
givenindependentviewdata.Sincetheproofofconceptoptsforaoneversionofthe
truthapproach,dependentviewstatementsarederivedfromOeBTOstatements,which
areintheirturnderivedfromthedependentviewdataofthemodellerasshownabove.
Intheindependentview,fromandtostatementsconveythegiveandtake(inflowand
outflow)semanticsthatarecharacteristictothedependentview.Thestatementsin
sectionA.3.1.usetheinformationconveyedbytoandfromstatementstoassigninflow
andoutflowsemanticstostock-flows.Anoutflowisastock-flowconnectedtoatransfer
fromthemodeller.Aninflowisastock-flowconnectedtoatransfertothemodeller.The
firststatementunderA.3.1.transformsfromsemantics,whicharetypicalforthe
independentviewperspective,tooutflowsemantics,whicharetypicalforthe
dependentviewperspective.Iftherelatorbetween102and106isafrom(i.e.
oebto_relator([102,106],is_a,from,to,106))andthereisastock-flow
relator101-102accordingto106(i.e.oebto_relator([102,101],is_a,stock-
flow,to,106)),thenthisstock-flowmustbeanoutflowto106(i.e.
dependent_relator([102,101],is_a,outflow,to,106)).Thesecond
statementunderA.3.1.doesthesameforinflowandtosemantics.
/*A.3. Deriving a Trading-partner view from the Independent view*/ /*A.3.1. inflow and outflow for transfers*/ dependent_relator([X,Y],is_a,outflow,to,P):- oebto_relator([X,Y],is_a,stock-flow,to,P), oebto_relator([Y,P],is_a,from,to,P). dependent_relator([X,Y],is_a,inflow,to,P):- oebto_relator([X,Y],is_a,stock-flow,to,P),
35
oebto_relator([Y,P],is_a,to,to,P). Code Snippet. 5. Derives inflow and outflow relators from to and from semantics
Similarly,buyerandsellersemanticsandsale,purchase,cash-receiptandcash-
disbursementsemanticscanbederivedfromthefromandtostatementsinthe
independentview.ThestatementsinsectionA.3.2.3.,shownincodesnippet6,assign
sellerandbuyersemantics.Ifatrading-partnerisconnectedtoadeliverythroughato
relator,itmustbeabuyer,itatrading-partnerisconnectedtoadeliverythroughafrom
relator,itmustbeaseller.ThestatementsunderA.3.2.3.operationalisetheagent
patternsunderA.3.2.3.1.Ifentity115isadeliveryto107(i.e.
oebto_role(115,is_a,delivery,to,107))andrelator115-104isafromto
107(i.e.oebto_relator([115,104],is_a,from,to,107)),104mustbeaseller
to107inthecontextofrelator115-104(i.e.
dependent_role_event(104,is_a,seller,to,107,in,[115,104])).
/*A.3.2.2. Event semantics*/ dependent_role_event(E,is_a,X,to,P):- oebto_role(E,is_a,Y,to,P), oebto_relator([E,P],is_a,Z,to,P), event_pattern(X,is_a,Y,Z). /*A.3.2.2.1. Event Patterns*/ event_pattern(sale,is_a,delivery,from). event_pattern(purchase,is_a,delivery,to). event_pattern(cash-receipt,is_a,payment,to). event_pattern(cash-disbursement,is_a,payment,from). /*A.3.2.3. Agent semantics*/ dependent_role_agent(A,is_a,Z,to,P,in,[D,A]):- oebto_relator([D,A],is_a,Y,to,P), oebto_role(D,is_a,X,to,P), agent_pattern(X,Y,Z). /*A.3.2.3.1. Agent patterns*/ agent_pattern(delivery,to,buyer). agent_pattern(delivery,from,seller).
Code Snippet. 6. Derives event and agent roles from to and from statements
36
Thestatementsinsection3.2.2.,shownincodesnippet6,assignsale,purchase,cash-
receiptandcash-disbursementsemanticstodeliveriesandpayments.Ifatrading-
partnerisconnectedtoadeliverythroughafromrelator,thedeliverymustbeasale
fromhisviewpoint.Ifatrading-partnerisconnectedtoadeliverythroughatorelator,
thedeliverymustbeapurchasefromhisviewpoint.Ifatrading-partnerisconnectedto
apaymentthroughatorelator,thepaymentmustbeacash-receiptfromhisviewpoint.
Ifatrading-partnerisconnectedtoapaymentthroughafromrelator,thepaymentmust
beacash-disbursementfromthisviewpoint.ThestatementunderA.3.2.2.
operationalisesthesemanticpatternsdefinedunderA.3.2.2.1.Forexample,entity102is
asaletoentity106(i.e.dependent_role_event(102,is_a,sale,to,106))if
102isadelivery(i.e.oebto_role(102,is_a,delivery,to,106))and102-106
isafromaccordingto106(i.e.
oebto_relator([102,106],is_a,from,to,106)).
Therea_roleandrea_relatorstatementsinappendixBidentifytherolesandrelatorsas
theyareassignedbytrading-partner104.ThestatementsinsectionB.1ofappendixB
areequivalenttotrading-partner104’sperspectiveinfig.6.Forexample,113isacash-
disbursementto104,102isapurchaseto104,andtherelatorbetweenthemisa
transferdualityin104’sperspective.Sincetrading-partner104isnotself-awareinthe
exemplar,trading-partner106and107identify104asatrading-partner.AppendixB
alsocontainstheaccessstatementsthatgivetrading-partner106and107,aswellasan
independentthirdperson—i.e.000—accessto104’sdata.
37
AppendixD,containstheteststhatevaluatewhethertheopposingperspectivesof
trading-partner106and107,andanindependentperspective(asvisualisedinfig.6)
canbederivedfromtrading-partner104’sperspectiveasformalisedbythestatements
inappendixB.SectionD.1.teststheperspectiveoftrading-partner106,andsectionD.2.
testtheperspectiveoftrading-partner107.SectionD.3teststheperspectiveofan
independentthird-party.SectionD.4testsafull-disclosureapproachtotheindependent
view.Thisalternativeindependentviewhasnoaccessrestrictions,whichmeansthatall
personscanaccesstheinformation,andincludestransformationrolesandrelatorsinits
vocabulary,whichmeansthattraceabilityofproductcomponentsisenabledthroughthe
publicationofproductiontransformationprocesses.Finally,inorderprovethatthe
approachpresentedabovecanbeappliedtotraditionalREAdata,appendixEtests
whetheranindependentandopposingdependentviewcanbederivedfromtheProlog
codefoundinGeerts&McCarthy’s(2000a)intensionalreasoningpaper.Therelevant9
partofGeert’s&McCarthy’sPrologcodecanbefoundinappendixC.AppendixE
demonstratesthatthelogicintroducedinthispapercantransformdependentviewdata
intoindependentviewdatathatcaninturnbetransformedintothedependentviewofa
trading-partnerwithanopposingperspective(i.e.theothertrading-partner).Toachieve
this,itisonlyrequiredtoidentifythetrading-partnertowhichthedependentview
belongsandmapthevocabularyoftheproprietarydataformattothatofthedependent
viewmodelpresentedabove.Sincethemodelpresentedaboveisbasedonrole
semantics,itdoesnotrequireinterferingwiththesemanticsoftheproprietarydata
model.
9 As we are only interested in the data, all Prolog statements relating to the definition of the conceptual schema in the intensional reasoning paper have been omitted.
38
Discussion
Theinnovationintheformalisationthatwehavepresentedinthispaperisthe
identificationofatrading-partner’scentralroleinREAmodels,astrading-partners
definetheirownuniverseofdiscoursebyassigningREA-rolesandrelatorstoinstances
ofdifferentkinds.ThroughthisformalisationofREAconstructsasrolesandrelators,the
constructsbecomecontextdependenttagsassignedtoinstancesofakind.Akindisa
rigidsortal,whichmeansthattheirinstancescannotchangetheirkindduringtheir
lifespan.Forexample,anaturalperson,whichisaninstanceofakind,becomesanagent
inthecontextofanexchange,orland(kind)isaresourcewhenitiscultivated.This
distinctionbetweenasubjectiveandobjectiverealityisrelevanttoallhuman
interaction.Thisdistinctionisimportantinmanyeconomicdisciplines(suchas
contracts)andtakentoanextremeininductivegametheory,inwhichitispossibleto
abstractfromthemeaningassignedtoanobjectivereality(M.Kaneko&Kline,2008;
MamoruKaneko&Kline,2010).Theabilitytomodelaneconomicrealitywithoutthe
needforanobjectiverealitymightbeextremelyusefulformodellingREA’sunhappy
paths.Forexample,anexpectedsalethatdidnotgothrough.Modellingsuchanunhappy
pathcanbeverydifficultifweassumethatvalueisanintrinsic(objective)attributeof
thingsirrespectiveoftheirutilitytoaspecificperson.However,whenworkingwith
subjectiverealities,itispossibletoaccountforasuddenlossofutility–andhencevalue
–ofaproductforapersonwithoutobservinganychangeintheobjectivepropertiesofa
product.Whenmodellingresources,eventsandagentsasrolesinsteadofkinds,wecan
accountforthesubjectiveandcontextualcharacterofvalue.Suchsubjectiveeconomic
modelsshouldallowustomodelandsimulatebothrationalandirrationalcustomer
behaviour.
39
Thispaperalsodemonstratesthatitispossibletoformalisethedependentviewasa
specialisationoftheindependentview.Consequently,is-asemanticsoperationalisedin
object-orientedprogrammingcouldeasilyimplementthelogicthatobtainsthe
independentviewbystrippingtheview-dependentsemanticsfromtheoriginaldata.
Thepaperalsoshowsthatthefromandtosemanticsintheindependentvieware
meaningfulenoughtoderivetheperspectiveofoneormoretrading-partnersfroman
independentview.
Thisdiscoveryopensnewperspectivesforblockchainaccountinginwhichtheblock-
chaintakesanindependentperspectiveandindividualtrading-partnersderivetheir
ownperspectivefromtheblock-chain.Suchablockchainwouldneedacleardefinition
ofitsscope,whichcouldbelimitedtotransfersonlyorcouldinvolvetransformationsas
well.ThefilteringprincipleshowninsectionA.3.couldbeusedinbothsituations.The
vocabularyinsectionA.3.thatislimitedtotransfers,andalsoinA.6.allowsforsharing
informationabouttransformations.Inamorerestrictedimplementation,thesharing
couldbelimitedtodeliveriesorpaymentsonly.Consequently,thisfilteringprinciple
mightbeusefuliftrading-partnerswanttoshareenoughinformationforcollaboration
withoutsharingtoomuchinformationaboutprocessesthatprovidecompetitive
advantage.
GiventhatREA2usestheexactsameprimitivesastheREAontologythatmapsittoother
interpretationsandimplementationsofREA,ontologiesthathavebeenmappedtothe
originalinterpretationoftheREAontologyshouldbestraightforward.However,
alternativesfororextensionsoftheREAnomenclatureusedinthispaperexist.For
example,Hrubyetal.(2006)defineproviderandrecipientforparticipationrelators
relatedtoincrementordecrementevents.Thatstated,LaurierandPoels(2014)explain
40
howtheseprovider-recipientsemanticsrelatetotheinside-outsidesemanticsusedin
thispaper.
Limitations ThispaperrefrainsfromaddressingtheconstraintsonassigningREArolestoinstances
ofkinds.Forexample,canacomputerplaytheroleofeconomicagent?Whilstof
interest,suchdiscussionswoulddistractfromthecentralcontributionofthispaper.The
paperalsolimitsitselftoasetofcoreREAconcepts.Itdoesnotaddresscommitments,
types,groups,responsibilityrelatorsandmanyotherREAconstructs,soastofocuson
eventrolesandtheirrelators.Inacollaborationspace,asingletransfercanbeassigned
differentsemanticsbytrading-partnersandthird-parties.Thesameappliesto
commitments,whicharelikeevents.Asaresult,anREAexpertshouldbeabletoapply
theprinciplesshownabovetocommitments.NexttoaddressingonlycoreREAconcepts,
thepaperalsorefrainsfromaddressingthesimilaritybetweenREArolesandrelations
andthematicrolesandrelations.Thediscussiononrolesandrelationshipsaboveis
solelyfocusedonaneconomiccontext,abstractingfromlinguisticaspects.Athorough
analysisoftheinteractionbetweenthelinguisticandeconomicaspectsofREAcould
contributetonaturallanguageprocessingincommerce.
Wheretheproduceandconsumeroleandthetransformationdualityareshowninfig.3,
theirsemanticsinthedependentandindependentperspectivehavenotbeendiscussed.
Themainreasonfortheirpresentexclusionisthattheirsemanticsareunderspecifiedin
theREAliterature,andthattheformalisationwouldinvolveaconsiderableamountof
speculation(W.E.McCarthyetal.,2016).Giventheseremarks,athoroughdiscussionon
transformationsisconsideredoutofscope.
41
Althoughvisualisedinfig.4,adiscussiononinside-partyrelatorsisalsoconsideredout
ofscope.Intheformalisationpresentedabove,atransferdualityismodelledbyone
trading-partner.Subsequently,thedualityissharedintheindependentviewandneeds
tobeacceptedbythetrading-partnerwithanopposingview.Futureresearchshould
addressthepresenceandabsenceofinteractionsbetweenonetrading-partners
subjectivereality,theobjectiverealityandthesubjectiverealityofatrading-partner
withanopposingview.Thisresearchshouldmitigatethecurrentdominanceofthe
modellingtrading-partnerinthetrianglesubjectivereality,objectiverealityand
opposingsubjectivereality.
Asthereis–tothebestofourknowledge–nomodellinglanguagethatallowsfora
multi-layerobservant-dependentsemantics,wehavechosenOntoUMLtoformaliseour
modelsasitssemanticscloselyrelatetothesemanticsrequiredforshowingamulti-
perspectiverealitywithlayeredsemantics.However,wearefullyawarethattheuseof
OntoUMLasdemonstratedinthispapergoeswell-beyonditsintendeduse.Forexample,
thedefinitionofnamespacesinsidetagsisanovelpracticethatwasrequiredfor
showingmultipleperspectivesinasinglemodel(i.e.theperspectivesoftradingpartner
104,106,107and000).TheuseofOntoUMLroles,rolemixinsandrelatorsassocial
constructspartofanobserver’ssubjectiverealityisnotincludedinOntoUML’sintended
use.Consequently,thispaperintroducesafirstusecase–andhencetheneedfor–
modellinglanguagesthatsupportmultiple(subjective)realitiessimultaneously.
Implications TheREA2formalisationintroducedaboveallowsforanautomatedtransformationfrom
dependentviewinformationtoindependentviewinformationandviceversa.This
approachisexpectedtofacilitatetheintegrationbetweentraditionalERPandAIS,of
42
whichmosttakethedependentview,andNetworkResourcePlanning(NRP)systems,
whichtaketheindependentview("ValueNetworkSoftware(NRP),"2016).Basedon
thesetwotypesonenterpriseinformationsystems(EIS),REA2thusaddressesEIS
interoperabilitybetweentraditionalEIS(whichtypicallyusesatrading-partner
perspective),andEISforthecollaborativeeconomy(whichtypicallyusesatrading-
partnerindependentperspective).
ThroughaunifiedformalisationoftheREAontology,themodelandmeta-model
presentedabovebridgethegapbetween(accounting)informationforobtaininga
competitiveadvantagethatisclassifiedandcannotbesharedsuchasthecompany’scost
structure,businessprocesslayouts,employeeexpertise,andunclassifiedtrade
informationthatisessentialforcollaborativevaluecreation.Insuchasetting,classified
informationcouldbedocumentedusingREA’sdependentview,wheretheunclassified
informationcouldbefilteredandsharedinthecollaborationspaceusingREA’s
independentviewderivedautomaticallyfromthedependentviewbystatementslike
thoseinsectionA.2.Thevocabularythatdefinesthefilteringcouldbeextendedor
reducedaccordingtotherequirementsandpreferencesofthetrading-partners
involved.
SincetheREAontologyfindsitsorigininaccountingtheviewintegrationtechnique
shownabovecouldallowforanintegratedtypeofaccountingthatcouldtakeadvantage
ofthecloudtechnologythatiswidelysuccessful.Theabilitytoshareaccountingdata
automatically,whichisexpectedtodecreasethecostofcomplianceandwithouttherisk
ofexposinginformationofstrategicimportancemightincreaseanorganisations’
willingnesstosharetradeinformation.Suchinformationcanhelpimplement‘e-
43
customs’andimplementsupply-chainmonitoringsystemssuchaswithblockchains.
Similarly,thehithertohiddeninformationexplicatedbyourapproachcouldincrease
foodsafety,fightcounterfeitandmoneylaundering.
Future research Thispaperhasmainlyfocussedonconstructionrigoranddesign.Whilsthavingthe
ambitiontobepartoftheREAresearchembeddedinthedesignscienceparadigm,
relevanceevaluationwillfollowinalaterphaseofthisproject.Therelevancewouldbe
testedbygeneratingandtestingaprototypeinacommercialprogramminglanguage
(e.g.,C#,Java,Python,Ethereum).Thetestscenariosofthisinitialprototypewillalso
includetrackingandtracing(Laurier&Poels,2012).Aftertestingthisinitialprototype,
commitments,typesandgroupswillbeaddedtotheprototype.Aftertestingthis
prototypeforcloud-basedandblockchainaccounting,weintendtoaddcross-company
planningtotheprototype.Cross-companyplanningistherationalefortheview
integrationtechniqueintroducedaboveandisexpectedtoadvancethestate-of-the-art
considerably,sinceitisourgoaltobuildaprototypethatplansanentirevaluenetwork
(supplychain)whiletheprocessingshouldbelocalanddecentralised.Afterbuilding
andtestingthecross-companyplanningprototype,itisourgoaltobuildlibraries,
prototypesandtestthecommercialprototypesinareal-worldsetting.Weplantodraw
onexistingexperiences.Forexample,anREA-basedintra-companyplanningsystem
thathasalreadybeenbuiltinJava(Buysse&Jonnaert,2012).
ThispaperprimarilyusesMcCarthy1982andISOOeBTO15944-4asasourceforit’s
REAvocabulary.Inthelightofthefutureworkabove,moregeneralormorespecific
termsmightberequired.Forexample,itmightbeconvenienttohavemoreelegantbut
perspectiveindependentgenerallyaccceptedtermsforsellerandbuyer,vendorand
client.Itwouldalsobeconvenienttofindgenerallyacceptedtermsforsalesand
44
purchases,andcash-receiptandcashdisbursementingoodsdominantlogic,
servicedominantlogic,rightsdominantlogic,bartertradeandfinance(i.e.money-for-
moneyexchanges).
Conclusions
ThispaperhasintroducedaninnovativeformalisationoftheREAOntologycalledREA2,
whichallowsforviewintegrationacrosstrading-partnersandthirdparties.REA2uses
REAconstructsasrolemixinsassignedbyanexplicitlyidentifiedtrading-partnerasthe
modellerinsteadofrigidkindsassignedbyanimplicitmodeller.Viewintegrationis
definedastheintegrationoflocalviewsintoaglobalmodel,specifyinghoweachof
theselocalviewscanbederivedfromtheglobalview(McCarthy,1982).Incontrastto
McCarthy’sdefinitionofviewintegration,theglobalmodelthatisREA2andpresented
hereiscollaboration-spacewideandnotlimitedtotheenterprise,whereasthelocal
viewscanremaindependentontheperspectiveofatrading-partner.
TheREA2formalisationalsoallowsustodistinguishbetweenadataformatfor
informationthatisrelevanttotheentiretradecommunityandthirdparties—i.e.the
independentview—andadataformatforinformationthatisonlyrelevanttomembers
oftheorganisation—i.e.thedependentview.Furthermore,atechniqueisdemonstrated
thatcandirectlyorindirectlygenerateinformationrelevanttoanindependentthird-
partyoratrading-partnerwithanopposingperspectivefromtheinformationsharedby
asingletrading-partner.Thistechniqueshowsthattheomissionoftheperspective
definingtrading-partnerimpedesanautomatedtransformationofthedependentto
independentview,orfromonetrading-partner’sviewtotheopposingviewofanother
trading-partnerascomputerscannotdealwiththeambiguity.Infig.6,theperspective
45
definingtrading-partnerdefinesanamespaceinthetagssuchas“106::sale”.In
AppendixA,thisperspectiveofthedefiningtrading-partnerisshownasvariablePinthe
statements.Humanscaneasilydealwiththisambiguityastheyareawareofthetrading-
partnerthatdefinestheperspectivetakenbyaninformationsystem.Humansarealso
familiarwiththerealitythat,whatisgivenawaybyonetrading-partner,mustbe
receivedbyatrading-partnerwithanopposingview.Givencomputerscannotdothe
same,wehavebroughtcomputerproductivitytothisotherwisepurelyhuman
endeavour.REA2raisesawarenessthatmakingthisview-determiningtrading-partner
explicitindependentviewmodelsallowsforautomatedtransformationsbetween
opposingtrading-partnerviewsandbetweenoneormoretrading-partnerviewsand
theindependentview.
Initswidestsense,thispaperhighlightsREA2’simplicationsforthebeneficialfuture
developmentofa)enterpriseinformationsystems(EIS)inthecloud,b)social-media-
basedEIS,c)blockchainEIS,andd)EISinteroperability.Themodestcontribution
demonstratedbyREA2thusfarmovesREAclosertowardsthataim.
Acknowledgements
WewouldliketothankW.E.McCarthy,whohelpedusrealizethattheworkon
integratingREA’sdependentandindependentviewwedidearliercouldbesimplified
whilemakingitmorepowerful.WewouldalsoliketothankRobertHaugen,Pavel
Hruby,PaulJohannessonandananonymousreviewerfortheirvaluedcomments,and
KarlDawsonfromAmazon.comfortheinspiringdiscussiononpotentialusecases.
46
References
(OMG),O.M.G.(2016).MetaObjectFacility.Needham,MA02494USA:ObjectManagementGroup.
Brandenburger,A.,&Nalebuff,B.(1996).Co-opetition(1sted.).NewYork:Doubleday.Buysse,F.,&Jonnaert,T.(2012).HetontwikkelenvaneenMRP(II)toepassingopbasisvan
hetREAreferentiemodel.(MasterofscienceMasterThesis),GhentUniversity,Ghent.Retrievedfromhttp://lib.ugent.be/fulltxt/RUG01/001/893/296/RUG01-001893296_2012_0001_AC.pdf
Colmerauer,A.,&Roussel,P.(1993).ThebirthofProlog.PaperpresentedattheThesecondACMSIGPLANconferenceonHistoryofprogramminglanguages,Cambridge,Massachusetts,USA.
Dunn,C.L.,Gerard,G.J.,&Grabski,S.V.(2016).Resources-Events-AgentsDesigneory:ARevolutionaryApproachtoEnterpriseSystemDesign.CommunicationsoftheAssociationforInformationSystem,38(5),554-595.
Fisher-Pauzenberger,C.,&Schwaiger,W.S.A.(2017).TheOntoREAAccountingModel:Ontology-basedModelingoftheAccountingDomain.ComplexSystemsInformaticsandModelingQuarterly,54(11),20-37.doi:10.7250/csimq.2017-11.02
Gailly,F.,&Geerts,G.L.(2013).Ontology-DrivenBusinessRuleSpecification.JournalofInformationSystems,27(1),79-104.doi:10.2308/isys-50428
Gailly,F.,Laurier,W.,&Poels,G.(2008).PositioningandFormalizingtheREAenterpriseontology.JournalofInformationSystems,22(2),219-248.
Gailly,F.,&Poels,G.(2009).UsingtheREAOntologytoCreateInteroperabilitybetweenE-CollaborationModelingStandards.InP.Eck,J.Gordijn,&R.Wieringa(Eds.),AdvancedInformationSystemsEngineering(Vol.5565,pp.395-409):SpringerBerlinHeidelberg.
Geerts,G.L.(2011).Adesignscienceresearchmethodologyanditsapplicationtoaccountinginformationsystemsresearch.InternationalJournalofAccountingInformationSystems,12(2),142-151.doi:10.1016/j.accinf.2011.02.004
Geerts,G.L.,&McCarthy,W.E.(2000a).AugmentedIntensionalReasoninginKnowledge-BasedAccountingSystems.JournalofInformationSystems,14(2),127.
Geerts,G.L.,&McCarthy,W.E.(2000b).TheOntologicalFoundationoftheREAEnterpriseInformationSystems.Retrievedfromhttp://www.msu.edu/user/mccarthy4/Alabama.doc
Geerts,G.L.,&McCarthy,W.E.(2002).Anontologicalanalysisoftheeconomicprimitivesoftheextended-REAenterpriseinformationarchitecture.InternationalJournalofAccountingInformationSystems,3(1),1-16.
Geerts,G.L.,&McCarthy,W.E.(2005).TheOntologicalFoundationofREAEnterpriseInformationSystems.UniversityofDelaware,Newark;MichiganStateUniversity,EastLansing.Retrievedfromhttp://www.msu.edu/user/mccarth4/Tulane.doc
Geerts,G.L.,&McCarthy,W.E.(2006).Policy-LevelSpecificationsinREAEnterpriseInformationSystems.JournalofInformationSystems,20(2),37-63.
Gordijn,J.(2002).e3-ValueinaNutshell.VrijeUniversiteitAmsterdam.Amsterdam.Guizzardi,G.(2014).OntologicalPatterns,Anti-PatternsandPatternLanguagesfor
Next-GenerationConceptualModeling.InE.Yu,G.Dobbie,M.Jarke,&S.Purao(Eds.),ConceptualModeling:33rdInternationalConference,ER2014,Atlanta,GA,USA,October27-29,2014.Proceedings(pp.13-27).Cham:SpringerInternationalPublishing.
47
Haugen,R.(2007).BeyondtheEnterprise:TakingREAtoHigherLevels.Hevner,A.,&Chatterjee,S.(2010).DesignScienceResearchinInformationSystems
DesignResearchinInformationSystems:TheoryandPractice(pp.9-22).Boston,MA:SpringerUS.
Hruby,P.,&Kiehn,J.(2006).RevisedClassificationoftheEnterpriseInformationArchitectureElements.Paperpresentedatthe3rdInternationalConferenceonEnterpriseSystemsandAccounting(ICESAcc’06),SantoriniIsland,Greece.http://www.itu.dk/people/hessellund/REA2006/papers/HrubyKiehn.pdf
Hruby,P.,Kiehn,J.,&Scheller,C.V.(2006).Model-drivendesignusingbusinesspatterns.Berlin:Springer.
ISO/IEC.(2007).Informationtechnology-BusinessOperationalViewPart4:Businesstransactionscenario-Accountingandeconomicontology.ISO/IECFDIS15944-4:2007(E).
Ito,S.,&Vymětal,D.(2013).TheformalREAmodelattheoperationallevel.AppliedOntology,8(4),275-300.doi:10.3233/AO-140129
Janssen,O.(2011).DevelopingaconsolidationapplicationusingtheREAontology.(MasterofScienceMasterThesis),GhentUniversity,Ghent.Retrievedfromhttp://lib.ugent.be/fulltxt/RUG01/001/788/494/RUG01-001788494_2012_0001_AC.pdf
Jaquet,M.(2006).APropertyDrivenApproachtowardsDescribingSemanticsofREAEntities.Proceedingsofthe2ndInternationalREATechnologyWorkshop.
Kaneko,M.,&Kline,J.J.(2008).Inductivegametheory:Abasicscenario.JournalofMathematicalEconomics,44(12),1332-1363.doi:10.1016/j.jmateco.2008.07.009
Kaneko,M.,&Kline,J.J.(2010).TwodialoguesonEpistemicLogicsandInductiveGameTheory.AdvancesinMathematicsResearch,12.
Laurier,W.,&Poels,G.(2012).TrackandTraceFuture,Present,andPastProductandMoneyFlowswithaResource-Event-AgentModel.InformationSystemsManagement,29(2),123-136.
Laurier,W.,&Poels,G.(2014).AnEnterpriseOntologyBasedConceptualModelingGrammarforRepresentingValueChainandSupplyChainScripts.InternationalJournalofConceptualStructuresandSmartApplications(IJCSSA),2(1),18-35.doi:10.4018/ijcssa.2014010102
Letaifa,S.B.(2014).Theuneasytransitionfromsupplychainstoecosystems:Thevalue-creation/value-capturedilemma.ManagementDecision,52(2),278-295.doi:doi:10.1108/MD-06-2013-0329
McCarthy,W.E.(Producer).AnREAmodelofaneconomicexchange.[MSPowerpointpresentation]Retrievedfromhttps://www.msu.edu/user/mccarth4/cookie--elmo--basic\%20REA.ppt
McCarthy,W.E.(1979).AnEntity-RelationshipViewofAccountingModels.AccountingReview,54(4),667.
McCarthy,W.E.(1982).TheREAAccountingModel:AGeneralizedFrameworkforAccountingSystemsinaSharedDataEnvironment.AccountingReview,57(3),554-578.
McCarthy,W.E.(2003).TheREAModelingApproachtoTeachingAccountingInformationSystems.IssuesinAccountingEducation,18(4),427-441.
McCarthy,W.E.,Geerts,G.L.,&Gal,G.(2016).CongruentenMeronymicConstellationsintheREAOntology.Paperpresentedatthe10thInternationalWorkshoponValueModelingandBusinessOntologies,Trento,Italy.
48
http://www.loa.istc.cnr.it/vmbo2016/wp-content/uploads/2016/02/VMBO2016_paper_15.pdf
McCarthy,W.E.,Geerts,G.L.,&Gal,G.(2016).TheEconomicStructuresofExchangesvs.ConversionsintheREAEnterpriseOntology.PaperpresentedattheUnderstandingtheNotionofValueintheServiceEconomy,Trento,ViaS.Croce77.http://www.loa.istc.cnr.it/vmbo2016/wp-content/uploads/2016/02/conversion-paper-trento-22-FEB.-embedded-figures.pdf
NEMO.(2015).OntoUMLSpecification1.0:daFonseca,LucasBassettiR.O'Leary,D.E.(2004).OntherelationshipbetweenREAandSAP.InternationalJournalof
AccountingInformationSystems,5(1),65-81.Sowa,J.F.(2000).Knowledgerepresentation:logical,philosophical,andcomputational
foundations.PacificGrove:Brooks/Cole.Steiner,J.,&Baker,J.(2016).Blockchain:thesolutionfortransparencyinproduct
supplychains.2016,fromhttps://www.provenance.org/whitepaperTan,Y.-H.,Niels,B.-A.,Klein,S.,&Rukanova,B.(Eds.).(2010).AcceleratingGlobalSupply
ChainswithIT-Innovation:ITAIDEToolsandMethods:Springer.ValueFlowsvfvocabs.(2016).fromhttps://valueflo.wsValueNetworkSoftware(NRP).(2016).fromhttp://mikorizal.org/software.htmlZdravkovic,J.,Zikra,I.,&Ilayperuma,T.(2011).AnMDAMethodforServiceModelingby
FormalizingREAandOpen-ediBusinessFrameworkswithSBVR.InJ.Ralyté,I.Mirbel,&R.Deneckère(Eds.),EngineeringMethodsintheService-OrientedContext:4thIFIPWG8.1WorkingConferenceonMethodEngineering,ME2011,Paris,France,April20-22,2011.Proceedings(pp.219-224).Berlin,Heidelberg:SpringerBerlinHeidelberg.