A SEMIOTIC APPROACH TO WEB SERVICE DESCRIPTION

Adrian Benfell and Kecheng Liu

Informatics Research Centre, University of Reading, Ground Floor

Building 42, Whiteknights, Reading RG6 6WB, U.K.

Keywords: Semiotics, Web Services, Affordance.

Abstract: The discovery of suitable web services is a demanding challenge for organisations that plan to benefit from

this technology. Strategic objectives, organisational structures, business processes and technology placed

in a climate of constant change impact the normative behavioural patterns of people working in all kinds of

organisations. Such dynamic conditions can have a profound influence over the discovery of appropriate

web services. Advocated in this paper is a semiotic approach to web service description that configures a

solution to take into account the dynamic conditions affecting web service discovery. The semiotic

approach merges the articulation of dynamic conditions with web service description whilst facilitating the

engagement of service providers and consumers in joint actions. Framed by affordance, joint actions capture

the changeable normative behavioural patterns of people so that web service utilisation can be harmonised

with organisational contexts.

1 INTRODUCTION

Based upon the principle of building software made

accessible across organisational boundaries, web

service technology is an approach to leveraging

existing software to fulfil business objectives

(Berkem, 2008; Carey 2008; Li et al, 2007). Web

service technology is also a development of the

‘software engineering’ paradigm aligned to object

oriented programming. Encapsulation is a concept

found within web service utilisation, as a key aim is

to present to a service consumer a highly cohesive

data type parameter based interface whilst adhering

to a notion of low coupling. Web services are

particularly amenable to such a situation; a web

service is a functioning software unit that is located

on the internet and responds to incoming messages

using web technology. Web services are therefore

an instalment of software that can be reused by a

multitude of participants as service consumers. A

significant challenge emerges when service provider

organisations attempt to describe web services

effectively so that service consumer organisations

can use those descriptions to achieve the fusing of

web services with their own specialised

organisational contexts. Organisations encompass

contextualised social parameters established by

strategic objectives, organisational structures,

business processes and technology in a flux of

constant change resulting in normative behavioural

patterns that are only understood by people

(Berkem, 2008, BMM, 2008, OASIS, 2008). To

overcome the challenge of matching web services to

organisational contexts, the purpose of this paper is

to highlight the limitations of current approaches to

web service description, show how semiotic theory

can be used to enhance web service description, and

propose a concept that draws semiotics and web

service description and discovery into a unifying

paradigm.

2 CONTEMPORARY WEB

SERVICE DESCRIPTION

Web Service Description Language (WSDL 1.1)

files are based upon the Extensible Mark-up

Language (XML) and are constructed to list data

type definitions, message types, port types

(operations) and bindings (WSDL, 2001) whilst

focusing upon the communication protocols that

enable packets of data as Simple Object Access

Protocol (SOAP) messages to traverse a network

architecture (SOAP, 2007). Conventionally, a

service provider publishes web service(s) in a

service registry to a standard such as Universal

Liu K. and Benfell A.

A SEMIOTIC APPROACH TO WEB SERVICE DESCRIPTION.

DOI: 10.5220/0003270000680076

In Proceedings of the Twelfth International Conference on Informatics and Semiotics in Organisations (ICISO 2010), page

ISBN: 978-989-8425-26-3

Description Discovery and Integration (UDDI)

(Atkinson et al, 2007; UDDI, 2004). The service

registry provides the references to the web services

available and the links to description files. The bind

details of the web service enable a service provider

and consumer to have some form of interaction

where messages structured by the SOAP protocol

are exchanged (Erl, 2008; Huang et al, 2008;

Pastore, 2008). The syntactic nature of WSDL 1.1

files that facilitate the discovery of web services

limit the identification of appropriate web services,

as the connection properties of a web service are the

only properties made visible. Implementation logic

encapsulated within a web service is hidden from a

web service consumer precluding the possibility of

detailed examination. To overcome this limitation,

the concept of Semantic Web Services, a branch of

the Semantic Web, provides a way for service

providers to describe web services more effectively

using XML based ontology mark-up, and for service

consumers to assess web services more accurately

(Il-Woong et al, 2007; OWL-S, 2004; Papazoglou et

al, 2007; Shadbolt et al, 2006).

The Semantic Web, Berners-Lee et al (2001), is

founded on the idea that meaning can be obtained

from data using standardised mark-up to represent

ontologies. Prime examples of ontology languages

particular to web service description are WSDL-S

(2005), WSML (2008) and OWL-S (2004). These

ontology languages describe web services from a

data orientated perspective, and align web services

to orchestrate business processes based upon clearly

defined interfaces and the calling of various web

service operations. However, these types of

ontologies have restrictions when considering the

dynamic conditions belonging to organisations

(deMoor, 2005; Schoop et al 2006). While the

description of data in a static structure is necessary

when navigating organisational boundaries of all

kinds to ensure consistency, data that are uniformly

described may be applied uniquely under dynamic

conditions. Rule based descriptions such as the

Semantic Web Rule Language (SWRL) (2004) can

be added to ontology languages like OWL-S to

elucidate further the orchestration of business

processes. For example, each OWL-S process

(OWL-S, 2004) is based upon an Input Output

Process Result (IOPR) model (Redavid et al, 2007).

The inputs represent the information required to

execute a process and the outputs are the result.

Preconditions are imposed upon the inputs to invoke

the process successfully. An OWL-S process may

have several results with corresponding outputs

whilst the result entity of the IOPR model provides a

means to specify this situation. However, with

SWRL, inferences made are based upon Modus

Ponens and Modus Tollens, but according to Beller

(2003, 2008) people are able to draw inferences

based upon Denial of the Antecedent and

Affirmation of the Consequent under ‘exhaustive’

conditions (where people consider all causally

relevant factors to the situation) and a ‘closed-world’

principle (where people believe they have

considered all causality until evidence proves

contrary). A dynamic condition is the chronological

representation of people’s ability to draw inferences

exhaustively and in ‘closed-world’ normative

behavioural patterns that influence organisational

activities (Liu and Benfell. 2009).

2.1 Problem Definition and Solution

The syntactic nature of WSDL 1.1 and the extended

semantic web service descriptions offer different

ways to describe web services but are not effective

when considering ‘pragmatic web service

description’. Pragmatic is defined here as the

dynamic conditions affected by contextualised social

parameters set by strategic objectives, organisational

structures, business processes and technology in a

climate of constant change that influence the

normative behavioural patterns of people. The

chance of meeting the needs of service consumers

without any form of communication, questions

current methods of web service description (Crasso

et al, 2008; Il-Woong et al, 2007; Papazoglou et al

2007; Singh, 2002). The challenges addressed using

pragmatic web service description in this paper are:

1. Organisational context: organisations

encompass complexities and fine distinctions

that must be catered for when documenting web

services prior to consumption. People

belonging to organisations have a deep

understanding of the normative behavioural

patterns they operate in and are therefore

ultimately responsible for the web service

descriptions they provide and consume.

2. Consumers before providers: the emphasis of

pragmatic web service descriptions must focus

upon the effect of information about web

services on service consumers, rather than on

the intended meanings supplied by service

providers defined in static object type

ontologies.

3. An abundance of signs: many different types

are available within web service description and

the promotion of them to ensure an accurate

description of web services may be

A SEMIOTIC APPROACH TO WEB SERVICE DESCRIPTION

accomplished by placing web service

description into a semiotic framework. The

signs inherent in various web service

descriptions and their meanings must be

communicable by any number of people

‘verbally and nonverbally’ and ‘horizontally

and vertically’ across organisational boundaries.

4. Dynamic conditions: the execution of a web

service in a dynamic condition is important as

such a context will determine the meanings held

in various ontologies that are time dependent.

Sign meaning is dynamic so all people (as

participants) must share in normative

behavioural patterns. Changes made to the

meaning of signs must be accounted for by

developing a temporal vocabulary that can be

pooled and understood by all participants

engaging with a web service that support

exhaustive and closed-world principles.

5. Interaction is vital: the dynamic interpretation

of web services affects communication as the

meanings generated in the process of interaction

between service providers and consumers

cannot be dependent on post deployment

semantic and syntactic descriptions only.

To address the first three challenges listed

above, semiotic theory by Peirce (1931-1958)

positions a solution in this paper to add pragmatic

descriptions to web services. Taking this approach,

a unifying paradigm is specified to draw service

providers and consumers into joint actions. To

achieve this objective, Stamper’s work on ontology

and affordance is used. Stamper (1985) recognises

three different types of ontology, whereby the first

relates to the recognition of symbols typically found

in any standard presentation format, and the second

identifies distinct objects and object type

classification. To satisfy the last two challenges in

this paper, the third type of ontology is particularly

relevant to web service description as it is based

upon the view that the world known to a person

consists of only the actions a person can carry out in

their environment – called affordance. Web services

are deployed in activity orientated situations, and

people carry out activities commensurate with their

expected duties, defined as normative behavioural

patterns. People invoke web services to complete a

specific activity; the description of web services

therefore should follow this pattern to enable web

service providers and consumers to fully appreciate

the normative behavioural patterns that affect web

service consumption.

3 SEMIOTICS AND WEB

SERVICES

Peirce’s particular semiotic theory of signs is

applicable to web service description due to its

triadic grounding. According to Everaert-Desmedt

(2010) Peirce’s version of semiotics is general (it

accounts for the emotional, practical and intellectual

experience of sign users), triadic (owing to the three

foundational philosophical categories that Peirce

created namely, firstness, secondness and thirdness),

and pragmatic (in that it takes into consideration the

dynamic context in which signs are produced and

consumed). Peirce’s version of semiotics also draws

together three terms that constitute a sign:

representamen; object; and interpretant. Firstness,

secondness and thirdness are used to illuminate

further each of these sign constituents. Peirce

developed three semiotic accounts, ‘early’, ‘interim’

and ‘final’. The early account includes some

fundamental concepts that also appear in the interim

and final accounts: representamen, object and

interpretant, and also illustrates how Peirce

establishes semiosis. Peirce uses different terms that

relate to the triadic nature of signs, for instance often

used is the term sign in place of representamen as

one of the three components of a sign. For lucidity

here, the term representamen is used in place of sign

(to help clearly define the triadic relationship), and

sign refers to the collective purpose of all three

parts. The following texts describing Peircean

semiotic theory are used (Atkin, 2006; Chandler,

2002; Commens, 2010; Short, 2007; Sowa, 2000) to

place web service technology into a semiotic

framework.

3.1 Early Account and Web Service

Description

Peirce formulates the triadic nature of semiosis

based upon three elements of a sign: representamen,

object and interpretant. Peirce suggests that a

representamen generates an interpretant in three

different ways, as an icon, an index and a symbol. A

sign is an icon when a quality is shared between a

representamen and its object, for example a portrait.

When a representamen is causally linked to its

object in some way it is an indexical sign. If a sign

user applies convention to understand a sign, such as

the rules of some language, a sign is symbolic. In

line with this account, web service description

includes two types of signs, indexical and symbolic.

For example, the word operation is an indexical sign

ICISO 2010 - International Conference on Informatics and Semiotics in Organisations

as it points to an actual web service operation, and

the word operation has a symbolic meaning in web

service description. These two types of signs are an

important classification for web service description,

but for a complete analysis they fall short when

considering the role of interpretant signs for web

service description. The process of semiosis Peirce

describes in this account is used to structure ‘shared

semiosis’ between service providers and consumers.

3.2 Interim Account and Web Service

Description

Building on his early account, Peirce devised a

phenomenological theory based upon three

categories: firstness as the conception of being

independent of something else, that is a

representamen distinguished by its own

phenomenological category; secondness as the

concept of a representamen being linked to or

having a reaction with its object; and thirdness as a

concept of mediation, where a first and second are

brought in relation in which an interpretant is

assigned to the way a representamen denotes its

object. Peirce’s phenomenological theory sets the

format for describing web services as the

representamen, object and interpretant are each

divided into these three phenomenological

categories. Following Peirce’s principle of

hierarchy amongst categories, a representamen

(firstness) cannot belong to a higher category than

its object (secondness) and in thirdness, an

interpretant cannot be in a higher category than its

object. Peirce yields ten mechanisms of sign

meaning that can be used to understand phenomenon

of all kinds that can equally be applied to web

service description.

The Representamen

For the successful signification by a representamen

of its object, qualisigns, sinsigns and legisigns are

used by Peirce to divide the representamen based

upon the three phenomenological categories.

Qualisign – firstness (material quality) is a

representamen that does not appear in web service

descriptions. Sinsign –secondness (material index) is

a representamen that relies upon an existential

connective with its object. These types of signs are

present within web service description, for example

they would constitute the actual existence of all

kinds of web service description files. Legisign –

thirdness, (material convention), is a representamen

based upon a law or habit, and in terms of web

service description, these signs are the expected

conventions, the syntax, of any web service

description file.

The Object

The object is the notion of the representamen

interacting with its object. In terms of web service

description, the object provides the meaning

associated with the syntax contained in a description

file. Iconic signs – firstness (relational quality), are

interpreted by some shared quality – a likeness to

something as an interpretation by a sign user, for

example the ‘file icon’ on a computer operating

system. Indexical signs – secondness (relational

index), are signs interpreted by causal connections.

Example indexical signs in terms of web services

include the actual existence of operations identified

by their names and the endpoints that can be

connected to. Indexical signs are found within the

semantic descriptions made possible by ontology

languages. Symbolic signs – thirdness (relational

mediation), are linked to their representamen by

knowing the conventional or habitual rules

applicable to the representamen. For example, by

practice a software programmer would accept that

‘double’, ‘decimal’ or ‘float’ would give a data type

for floating point arithmetic.

The Interpretant

The interpretant represents the concept of mediation,

where the representamen and object are brought into

a relation in which the representamen’s interpretant

is linked to the way a representamen denotes its

object. In essence, the interpretant is the reaction of

someone’s mind when a connection is made between

a representamen and an object and the resultant sign

meanings are expressed in a natural or artificial

language (Sowa, 2000). Rheme – firstness (formal

quality), the interpretant focuses a person’s

understanding of a sign based upon its (quality) in

that a representamen determines its object by its

quality only – for example a classifier for a set of

objects. Dicent – secondness, (formal index) the

interpretant focuses a person’s understanding on the

existential features of an object through proposition,

for example an operation name to index an actual

operation whilst the operation name used suggests

its purpose. Argument – thirdness, (formal

mediation) the interpretant focuses a person’s mind

on a rule of inference to derive an argument by

applying some kind of convention or law. Current

web service description does make use of such signs

but may entail web service descriptions to indicate

applicability under certain dynamic conditions.

A SEMIOTIC APPROACH TO WEB SERVICE DESCRIPTION

Table 1: Sign Mechanisms and Web Services.

Firstness Secondness Thirdness

Qualisign

(A quality)

Sinsign

(An existent thing)

1. A WSDL file, an

OWL-S file

Legisign

(A convention or law)

2. Description file

syntax.

Icon

(A similarity)

Index

(Causal connection)

1. A web service.

2. An actual

operation.

2. An actual data type.

2. Proper noun – an

object.

Symbol

(refers to its object by

convention or law)

2. Operation naming

convention.

2. Data type

conventions.

Rheme

(Quality only – a

common noun)

2. Class names.

2. Data type

variables : age for

example.

Dicent

(An sign of actual

existence – a

sentence)

1. The web service

used for a general

task.

2. An actual operation

invoked for a sub task

Argument

(An inference from

dicent signs in context)

The interpretant (table 1) provides a semiotic

frame that can be used to address the limitations of

matching web services to the dynamic conditions of

organisational contexts. For example, rhematic-

index-legisigns provide the classification of things

belonging to a web service that would normally be

found in a WSDL file (labelled as ‘2’ in table 1).

Dicent-index-sinsigns are available within WSDL

(denoted by ‘1’ in the table 1) and are also present if

web services are described using semantic mark-up

(denoted by ‘2’ in table 1). For example, a web

service operation exists (legisign as a firstness), an

actual operation by a name (secondness as an index),

and that a named operation can be called upon to

carry out a task, such as a process descriptor or

workflow in OWL-S for example (dicent as a

thirdness). Evidently missing from the framework in

table 1 is the sign classification argument-symbol-

legisign. This particular sign classification

synchronises the full mechanism of the interpretant

for all signs and therefore for pragmatic web service

description. To specify how pragmatic web service

description can work using argument-symbol-

legisigns for web service description, Peirce’s final

account is referred to.

3.3 Final Account and Web Service

Description

In this version Peirce divides the object and

interpretant to take into account a chronological

process of inquiry. This approach is applicable to

web service description when the dynamic

conditions of organisational contexts affect accurate

web service description and discovery. Peirce

introduces two important considerations with regard

to dividing an object and dividing an interpretant.

The terms Peirce uses are the ‘immediate’ and

‘dynamic’ object. The immediate object is the

object as a person would know it to be an object at

any instance in time. The dynamic object is the

object as it is known to be at the end of ‘exhaustive’

inquiry. The static representations within WSDL 1.1

and semantic ontology languages must be validated

against a process of inquiry to ensure that a web

service fits a changeable organisational context –

achievable as a dynamic object. However, to

explain fully what a web service realises, the

mechanism Peirce uses to divide the interpretant into

three, ‘immediate’, ‘dynamic’ and ‘final’ is also

particularly important. The dynamic interpretant is

an understanding of the relation between a

representamen and a dynamic object at any stage,

and the immediate interpretant is a generalised

understanding of the relationship between a

representamen and a dynamic object. The final

interpretant is the complete understanding of a

dynamic object that all people would agree to. In

the case of web service description, the final

interpretant is the agreement reached between

service providers and consumers about the

pragmatic nature of using web services in

organisational contexts. For instance, the reaction of

the dynamic object with the final interpretant

determines how an argument-symbol-legisign is

arrived at and requires a service provider and

consumer to join forcefully in a process of inquiry.

The argument-symbol-legisign is a norm or standard

that can be derived from a line of inquiry into the

applicability of a web service in various dynamic

conditions. Peirce’s ‘pragmatic maxim’ - three

grades of clarity Peirce (1931-1958) is applied to

arrive at argument-symbol-legisigns for the web

service description and discovery process. The first

grade of clarity is to have an unreflective grasp of

the structure of textual web service description –

immediate interpretant. The second grade of clarity

is being able to define the generalised concepts

within textual descriptions agreed dynamically

between a service provider and consumer – dynamic

interpretant. The final grade of clarity determines

what effects that are held in relation to the concepts

of study that are considered to be true, for example a

list dynamic conditions that both the service

provider and consumer agree to be true – the final

interpretant. The pragmatic maxim ensures that the

effects of web service information on consumers are

understood by service providers. The pragmatic

conditionals (as dynamic conditions) for web service

description aligned to texts, is linked to Peirce’s

account of modality. Possibility and necessity are

ICISO 2010 - International Conference on Informatics and Semiotics in Organisations

based upon the epistemological facts in relation to

the meaning of signs within a text. To say

something is necessary is to confirm that something

must be the case by a service provider or consumer.

To say it is possible is to say that under varied

dynamic conditions a service provider and consumer

know something to be the case. This reduces the

signs, made by an author of a textual web service

description explicable and translatable by a

consumer using modal representations. For

example, from the perspective of a service provider,

alethic modal operators to explain what may happen

in certain dynamic conditions, and for the service

consumer deontic modal operators to intimate the

normative behavioural patterns that must be adhered

to when activating and calling various components

of a web service under the same set of dynamic

conditionals.

4 SETTING PRAGMATIC WEB

SERVICE DESCRIPTION

Returning to Stamper’s (1985) overview of

ontologies, attaining interpretant signs such as

argument-symbol-legisigns, requires that all

participants understand the dynamic conditions in a

domain that a web service belongs to, referred to by

Stamper as affordance. The theory of affordances

originates from Gibson (1977) and can be extended

to the study of the real world for understanding

normative patterns of human behaviour (Liu, 2000),

additionally aligning itself to the ontology

specification provided by Stamper (1985). Society

as an environment makes many patterns of

behaviour possible; should a person (participant) be

separated from its environment, the repertoire of

behaviour the participant owns would cease to exist.

The purpose of identifying ‘affordance’ (see figure

1) is to provide the contextual setting for web

service description for all interpretant signs. At an

abstract level an affordance, for web service

description, shares some characteristics similar to

business capability modelling, Ulrich (2006).

Affordance is also applied similarly in other

situations such as Customer Relationship

Management, Finnegan and Currie (2010), and on-

line communities, Welser et al (2009). The purpose

of affordance is to place shared semiosis into a

framework. For example, semiosis for a service

provider starts when authoring syntactic or semantic

description files by soliciting description from other

existing texts – called intertextuality, Chandler

(2002). Semiosis for a service consumer occurs

when examining preliminary description files to

assess the functional properties and capabilities of a

web service. These two activities comply with

Peirce’s first grade of clarity, as they initiate shared

semiosis to form the representamen that is required

for joint actions to be effective:

1. The representamen – the elements (composition

of web service description) designating the codes

as dynamic objects in a shared vocabulary (first

grade of clarity);

2. The dynamic objects as they relate to the

semantics describing the codes contained in a

shared vocabulary. Dynamic objects also

correspond to the data structures and processing

capabilities of a web service (second grade of

clarity);

3. The final interpretant signs that describe all

features and capabilities of a web service in

dynamic conditions using high-level descriptions

based upon various modalities (third grade of

clarity).

With reference to figure 1 below, shared

semiosis profiles the representamen between a

service provider and consumer to understand the

various codes related to web service description

whilst working as the starting point in a

chronological series of joint actions. Codes hold the

syntactic and semantic meanings that describe web

service elements. Arrow 1 in figure 1 illustrates that

syntactic documentation such as computer program

source code, analysis and design specifications and

any other suitable text (also shown later in table 3)

can be used to describe an existing web service.

Intertextuality (relationship to other texts) and

encoding work together to create the codes, hence

codes are generated, as a first grade of clarity, that

encompass various elements of a web service.

Figure 1: Shared Semiosis.

The codes produce an initial shared vocabulary

as part of shared semiosis (formalised in table 2

Interpretant

Affordance

(no rm s)

Object

Representamen

SHARED SEM IOSIS

Intertextuality and encoding -

negotiated code and reading

Textual documents

related to description

or specification

Determine the codes

used for description

and discovery

2.Decoding

(Connotative)

1.Encoding (Denotative)

signifies

A SEMIOTIC APPROACH TO WEB SERVICE DESCRIPTION

below). Shared semiosis then moves onto

‘negotiated code and reading’ (Chandler, 2002) to

agree the pragmatic meanings, between service

providers and consumers, of the codes contained in a

shared vocabulary. The shared vocabulary is an

XML mark-up file structured by an XML Schema

(XSD) that a service provider and consumer have

access to. The codes have dependencies of different

kinds, first, with the elements of a web service, and

second with normative behavioural patterns. Norms

(Boella, 2006; Stamper, 1996; Young, 2008) (as

normative behavioural patterns) are dependent upon

affordances (figure 2 below), thus final interpretant

signs (table 1) are contingent upon the codes that

describe a web service linked to the norms contained

within affordances. The dependencies between the

codes, norms and the affordances facilitate the

matching of web services with dynamic

organisational contexts captured by modal

descriptors. For example, a service provider may

submit a description based upon some dynamic

conditions, specified as affordance qualifiers, to

suggest a ‘possible’ mode of interaction. A service

consumer, who has a set dynamic conditionals for

that particular affordance may have some related

‘obligatory actions’, the matching of obligations to

affordance qualifiers that may be deemed as

‘possible’ harmonises web service description and

discovery. Affordance qualifiers are part of the

semantic description as codes, for example a

workflow (refer to figure 2 for code) that is

structured in accord with a qualifier but could over

time change according to the organisational context.

The implication of this approach is that qualifiers do

not need to relate purely to the calling of web

service operations; they illustrate a complete

affordance context. Affordance qualifiers are also

captured and represented in a chronological format,

thus allowing the representation of normative

behavioural patterns to evolve. The valid-time of a

norm in the ‘real-world’ and any transaction-times

Liu (2000) and Stamper (1996) when a ‘temporal’

web service description is updated configures the

time dependent character of affordances. To

summarise, the semiotic branches advocated by

Morris (1938) are used in table 2 to show that web

service description is based upon the process of

shared semiosis and informed by affordance (also

refer to figure 2 below).

Table 2: Formalising Shared Semiosis.

Semiotic

branch

Intent and Real World

Effect

Semiosis

Syntactic

Encoding

(denotative

sign).

Capture through existing

texts the elements to form

codes that structure the

syntactic and semantic

features of a web service

description file.

Representamen – Textual code

(intertextuality and encoding)

Source code, analysis and design

specifications – narrative and

diagrammatic models.

Semantic

Decoding

(connotativ

e sign).

Comprehension by

consensus (dynamic objects)

the web service elements

symbolised as codes in

relation to their functions

and capabilities that belong

to an affordance.

Dynamic object – Connotative

sign (negotiated code and

reading) Ontological

dependencies linked to the

contextualised interpretation by

an Interpreter (Participant).

Pragmatic

(All

interpretant

signs).

Linking the interpretations

of the codes with potential

contexts and effects on all

participants and specifying a

meaning of all codes

congruent with all

participants.

Final interpretant –

Connotative signs (argument-

symbol-legisigns) linked to the

social parameters of a business

organisation defined as

affordances and structured using

norms and amplified using modal

operators.

Affordances convey intent as participant

behaviour in order to have an effect within the real-

world and are tempered further by antecedents like

an organisation, the social structures within an

organisation, and society, Stamper (1996). In order

to model the constancy of affordances that represent

the social parameters of organisations, real-world

effects influence the creation and modification of the

shared vocabulary, evidenced by the communicative

actions of participants. A Multi-responsive

communication framework Benfell and Liu (2009)

based upon communicative act theory by Austin

(1962) and Searle (1969) underpins the

communication segment of affordances.

5 CONCLUSIONS

Referring to figure 2, ‘WebService’ and ‘Code’ is a

mixture of syntactic and semantic web service

description. WebService is the syntactic

representation of the properties of web services and

the semantic meanings of those properties are

structured by codes during encoding. ‘Norm’ and

‘Affordance’ are the pragmatic element (interpretant

signs) of web service description. Not until

argument-symbol-legisigns are created can web

service description be fully achieved. Furthermore,

such signs are time dependent and owned by the

people who represent an organisation. The model in

figure 2 demonstrates the fulfilment of all challenges

initially outlined in this paper.

ICISO 2010 - International Conference on Informatics and Semiotics in Organisations

Figure 2: Affordance Structure.

Proposed in this paper is an alternative approach

to web service description. It promotes the idea of

‘affordance’ as a route to achieve the description and

therefore discovery of web services within the

dynamical conditions of different organisations

whilst enabling service providers and consumers,

through shared semiosis, to forcefully join. The

Peircean triadic viewpoint of a sign as

representamen, object and interpretant and the

different Peircean accounts of semiotic theory prove

in this case the applicability of such theory to

address the challenges outlined in this paper. The

dynamic organisational parameters are captured as

affordances to inform the composition of the final

interpretant as argument- symbol-legisigns, and to

structure joint actions. The work contained in this

paper is a supplement and not a replacement of

syntactic and semantic description files. However,

the limitations of syntactic and semantic description

files were shown. For implementation, a WSDL file

must be present but semantic description files based

upon OWL-S for example could be included within

affordances to describe the data elements and

processing features of web services.

REFERENCES

Austin, J., (1962). How to Do Things with Words.

Cambridge, Massachusetts: Harvard University Press.

Atkin, A. (2006). Peirce’s Theory of Signs, available at:

http://plato.stanford.edu/entries/peirce-semiotics/

Atkinson, C., Bostan, P., Hummel, O., and Stoll, D. A,

(2007) Practical Approach to Web Service Discovery

and Retrieval, IEEE International Conference on Web

Services, 9-13 July, Salt Lake City UT, USA. ISBN:

0-7695-2924-0.

Beller, S., (2003). The flexible use of deontic mental

models. In R. Alterman and D. Kirsh (Eds),

Proceedings of the 25

Annual Conference of the

Cognitive Science Society, pp 127-132.

Beller, S., (2008). Deontic norms, deontic reasoning, and

deontic conditionals. Thinking and Reasoning, 14, 4,

pp 305-341.

Benfell, A., and Liu, K., (2009). Defining a Pragmatic

Based Web-service Discovery Mechanism, I-

SEMANTICS Special Track: 4th AIS SigPrag,

International Pragmatic Web Conference. Graz,

Austria. ISBN 978-3-85125-060-2, pp 763-774.

Berkem, B., (2008) .From The Business Motivation Model

(BMM) To Service Oriented Architecture (SOA).

Journal of Object Technology, Vol. 7, No. 8,

November-December.

Berners-Lee, T. Hendler, J. and Lassila, O., (2001). The

Semantic Web, Scientific American, May, pp34-43.

Boella, G., Torre, L., and Verhagen, H., (2006).

Introduction to Normative Multiagent Systems,

Computational and Mathematical Organizational

Theory, Volume 12, Numbers 2-3, October, pp71-79

(9).

Business Motivation Model (2008). – OMG, Accessed at:

http://www.omg.org/spec/BMM/1.0/.

Carey, M., (2008) SOA What? IEEE Computer, March, pp

92-93.

Chandler, D., (2002). Semiotics – The Basics, Routledge.

Commens Dictionary of Peirce’s Terms, available at:

http://www.helsinki.fi/science/commens/dictionary.ht

Crasso, M., Zunino, A., and Campo, M., (2008). Easy web

service discovery: A query-by-example approach,

Science of Computer Programming, 71, pp 144–164.

De Moor, A., (2005). Patterns for the Pragmatic Web,

Invited paper, Proc. of the 13th International

Conference on Conceptual Structures, Kassel,

Germany, LNAI. Springer Verlag, Berlin, pp 1-18.

Erl, T., (2008). SOA: Principles of Service Design.

Prentice Hall.

Everaert-Desmedt, N., Peirce's Semiotics, available at:

http://www.signosemio.com/peirce/a_semiotique.asp

accessed 15th May 2010.

Finnegan, J, D., and Currie, L,W., (2010). A multi-layered

approach to CRM implementation: An integration

perspective, European Management Journal, Vol 28, 2,

pp 158 – 173,

Gibson, James, J. (1977). The Theory of Affordances. In

Perceiving, Acting, and Knowing, Eds. Robert Shaw

and John Bransford, ISBN 0-470-99014-7.

Huang, S., Chu, Y,. Li, S., and Yen, D., (2008). Enhancing

conflict detecting mechanism for Web Services

composition: A business process flow model

transformation approach, Information and Software

Technology, 50, 11, pp 1069-1087.

Il-Woong, Kim, Kyong-Ho, Lee. (2007). Web Services,

Describing Semantic Web Services: From UML to

OWL-S ICWS. IEEE International Conference on

Web Services, Volume, Issue, 9-13, pp 529 – 536.

«XSDcomplexT ype»

Norm

- normModalType: string [1..unbounded]

- normSpecification: string [1..unbounded]

«XSDcompl exT ype»

Code

- codeID: int

- codeName: string

- codeProcessModel: workflow

- codeSpecification: workflow

«XSDcomplexType»

WebService

- dataType: string [1..unbounded]

- object: class [1..unbounded]

- operation: string [1..unbounded]

«XSDcomplexType»

ValidTime

- finishTime: string [1..unbounded]

- startTime: string [1..unbounded]

«XSDtopLevelElement»

TransactionTime

- finishTime: string [1..unbounded]

- startTime: string [1..unbounded]

«XSDcomplexType»

Communication

- communicati onID: int

- communicati onType: string

«XSDcomplexT ype»

Participant

- email: string

- jobTitle: string

- name: string

- organisation: string

- telephone: string

«XSDcomplexType»

Affordanc e

- affordanceID: int

- affordanceQualifiers: qualifier

Antecedents

1..*

st ru ct ure s

1..*

influence

1..*

must have

1..*

generates

1..*

will belong to

1..*

must have

1..*

must have

1..*

will be involved in

will create

1..*

contributes to

1..*

A SEMIOTIC APPROACH TO WEB SERVICE DESCRIPTION

Li, Y., Liu, Y., Zhang, L., Li, G., Xie, B., Sun, S., (2007).

An Exploratory Study of Web Services on the Internet,

IEEE International Conference on Web Services, 9-13

July, Salt Lake City UT, USA. ISBN: 0-7695-2924-0.

Liu, K., (2000). Semiotics in Information Systems

Engineering, Cambridge University Press.

Liu, K., and Benfell, A., (2009). Pragmatic Web -

Incorporating Semiotics into Web Services,

Proceedings of the 4th International Conference on

Software and Data Technologies, Volume 1, Sofia,

Bulgaria, July 26-29, ISBN 978-989-674-009-2, pp 11

-18.

Morris, C.W., (1938). Foundations of the Theory of Signs,

Chicago: Chicago University Press.

OASIS (2008). Reference Architecture for Service

Oriented Architecture Version 1.0, Public Review

Draft 1, 23 http://docs.oasis-open.org/soa-rm/soa-

ra/v1.0/soa-ra-pr-01.pdf. 2008.

OWL-S (2004). OWL-S: Semantic Markup for Web

Services, Accessed at: http://www.w3.org/

Submission/OWL-S/

Papazoglou, M., Traverso, P., Dustdar, S., and Leymann,

F., (2007). Service-Oriented Computing: State of the

Art and Research Challenges, IEEE Computer, Nov,

pp 38 -45.

Pastore, S., (2008). The service discovery methods issue:

A web services UDDI specification framework

integrated in a grid environment, Journal of Network

and Computer Applications, 31, pp 93–107.

Peirce, C, S., (1931-1958). Collected writings (8 vols),

Editors, Charles Hartshorne, Paul Weiss and Arthur

W. Burks, Cambridge, MA, Harvard University Press.

Redavid, D., Iannone, L. and Payne, T. (2007). OWL-S

Atomic services composition with SWRL rules. In:

Proceedings of the 4th Italian Semantic Web

Workshop, 18-20 December, Dipartimento di

Informatica - Universita' degli Studi di Bari - Italy.

Searle, R., (1969). Speech Acts. Cambridge University

Press, London, UK.

Schoop, M. deMoor, A. and Deitz, J, L, G. (2006). The

Pragmatic Web: A Manifesto, Communications of the

ACM, 49, 5, pp 75-76.

Shadbolt, N. Hall, W. Berners-Lee, T. The Semantic Web

Revisited, (2006). IEEE Intelligent Systems, May-

June, pp 96-101.

Short, T, L., (2007). Peirce’s Theory of Signs, Cambridge

University Press, ISBN 0-521-84320-0.

Singh, M.P. (2002) The Pragmatic Web, IEEE Computing,

May-Jun, pp 4-5.

SOAP (2007). W3C, available at: http://www.w3.org/

TR/2007/REC-soap12-part1-20070427/

Sowa, J. (2000). Knowledge Representation: Logical,

Philosophical and Computational Foundations,

Brookes/Cole, Thomson Learning, ISBN: 0-534-

94965-7, pp 394 – 401.

Stamper, R., (1996). Signs, Norms, and Information

Systems, in Holmqvist, B. et al (Eds.) Signs at Work,

Walter de Gruyter, Berlin, Germany.

Stamper, R., (1985). Knowledge as action: a logic of

social norms and individual affordances. In Gilbert,

G, N., and Heath, C., (eds) Social Action and Artificial

Intelligence. Gower Press, Aldershot, Hampshire. pp

172-191.

SWRL (2004). A Semantic Web Rule Language

Combining OWL and RuleML, available at:

http://www.w3.org/Submission/2004/SUBM-SWRL-

20040521/

UDDI, OASIS, available at: http://uddi.org/pubs/uddi-

v3.0.2-20041019.htm ,2004.

Ulrich, H. 2006, available at: http://msdn.microsoft.com/

en-us/architecture/aa479368.aspx

Young, P., (2008). Social Norms, The New Palgrave

Dictionary of Economics, Second Edition, Steven N.

Durlauf and Lawrence E. Blume, eds. London:

Macmillan.

Welser, T, H., Gleave, E., Barash, V., Smith, M., Meckes,

J., (2009). Whither the Experts? Social affordances

and the cultivation of experts in community Q&A

systems, International Conference on Computational

Science and Engineering, pp 450 – 455.

WSDL, W3C, available at: http://www.w3.org/

TR/2001/NOTE-wsdl-20010315

WSDL-S, W3C, available at: http://www.w3.org/

Submission/2005/SUBM-WSDL-S-20051107/

WSML, WSMO, available at: http://www.wsmo.org/

TR/d16/d16.1/v1.0/

ICISO 2010 - International Conference on Informatics and Semiotics in Organisations