Richard Benjamins1, John Davies2, Elmar Dorner3,

John Domingue4, Dieter Fensel5, Ozelin López1,

Raphael Volz6, Alexander Wahler7, Michal Zaremba5

1
iSOCO, http://www.isoco.com/
2
BT, http://www.bt.com/
3
SAP, http://www.sap.com/
4
Open University Milton Keynes, http://www. kmi.open.ac.uk/
5
University of Innsbruck, http://www.deri.at/
6
FZI, http://www.fzi.de/
7
Hanival, http://www.hanival.com/
About STi2
The mission of STII2 short for Semantic Technology Institutes International is to establish
semantics as a core pillar of modern computer engineering. It is supposed to be the leading
international think tank in this field. STI2 is organised as an association of jointly interested
scientific, industrial and governmental parties. STI2 actively takes the lead in developing new
business models and in improving the way people and businesses communicate and interact.
STI2 provides general services in the areas of
• Research
• Technology
• Realization
STi2 is supported by scientific, industrial and governmental parties of the world wide Semantic
Web and Service community. STi2 is an initiative of the European Semantic Systems Initiatives
(ESSI) 1 , the European FP6 Project Knowledge Web 2 , and DERI International 3 .

Contact: Ilona Zaremba, email: [email protected], http://www.sti2.org/

Vienna, February 20, 2007

1
http://www.essi-cluster.org/
2
http://knowledgeweb.semanticweb.org/
3
http://www.deri.org/

2
Abstract. Service Web 3.0 will help to realize a world where billions of
parties are exposing and consuming services via advanced web technology.
We will provide a comprehensive framework and infrastructure that
integrates four complimentary and revolutionary technical advances into a
coherent and domain interdependent service delivery platform: (1) Service
Oriented Architectures (SOA) as a means to abstract from software to
service ware; (2) Web technology as an infrastructure and underlying
infrastructure for integration of services at a world wide scale. (3) Semantic
Web technology as a means to abstract from syntax to semantics; and (4)
Web 2.0 as a means to structure human-machine cooperation in an efficient
and cost-effective manner. Herby we want to integrate the service world of
large enterprises, SMEs, and end-users and enabling them as peers (i.e.,
service consumers as well as service providers) in a network of equals.

3
Contents
1. Introduction..................................................................................................................... 5
2. Integration Web and SOA............................................................................................... 7
3. Integration Semantics and SOA...................................................................................... 8
4. Integration Web 2.0 and SOA....................................................................................... 10
4.1 Mashing Web and Web Services ............................................................................ 10
4.2 Mediating and Creating Communities .................................................................... 13
4.3 Consensus building mechanisms ............................................................................ 16
5. Scenarios ....................................................................................................................... 18

4
1. Introduction
After four decades of rapid advances in computing, we are embarking on the greatest leap
forward in computing that includes revolutionary changes at all levels of computing from
the hardware through the middleware and infrastructure to applications and more
importantly in intelligence. Service Web 3.0 will provide a comprehensive framework
and infrastructure that integrates four complimentary and revolutionary technical
advances:

• Service Oriented Architectures (SOA) as a means to abstract from software to

service ware;
• Web technology as an infrastructure and underlying infrastructure for integration
of services at a world wide scale.
• Semantic Web technology as a means to abstract from syntax to semantics; and
• Web 2.0 as a means to structure human-machine cooperation in an efficient and
cost-effective manner

Computer science is entering a new generation. The previous generation was based on
abstracting from hardware. The emerging generation comes from abstracting from
software and sees all resources as services in a service-oriented architecture (SOA). In a
world of services, it is the service that counts for a customer and not the software or
hardware components that implement the service. Service-oriented architectures are
rapidly becoming the dominant computing paradigm. 4 However, current SOA solutions
are still restricted in their application context to in-house solution of companies. A
service web will have billions of services. While service orientation is widely
acknowledged for its potential to revolutionize the world of computing by abstracting
form underlying hardware and software layers, that success depends on resolving
fundamental challenges that SOA does not address currently.

SOAs will not scale without: properly incorporating principles that made the web scaling
to a world wide communication infrastructure; signification mechanization of service
discovery, negotiation, adaptation, composition, invocation, and monitoring as well as
service interaction requiring data, protocol, and process mediation; and a balanced
integration of services provided by human and machines. In a services-oriented world,
services must be discovered and selected based on requirements, then orchestrated and
adapted or integrated. Solving these problems is a major pre-requisite to realize a web of
services interconnection billion of services (as the current web does for information
sources) effectively “that enable context-awareness and discovery, advertising,
personalization and dynamic composition of services.” 5

4
M. Brodie, C. Bussler, J. de Brujin, T. Fahringer, D. Fensel, M. Hepp, H. Lausen, D. Roman, T. Strang,
H. Werthner, and M. Zaremba: Semantically Enabled Service Oriented Architectures: A Manifesto and a
Paradigm Shift in Computer Science, DERI Technical Report TR20051226, 26 December 2005.
http://www.deri.at/fileadmin/documents/DERI-TR-2005-12-26.pdf
5
ICT – Information and communication Technologies, Work Programme 2007, Page 13.

5
The mission of Service Web 3.0 is to provide solutions to integration and search that will
enable the Service Oriented Architecture (SOA) revolution on a world-wide scale.
Hereby we must focus on three major areas where we need to extend current approaches
towards service orientation:

• First of all, the web principle and technology are applied to service orientation
providing an open and dynamically changing environment of services open for
third-party usage. Services will appear, disappear, change location; start free then
based on pay-per-use, will be blocked, out of service, inspected for antitrust, etc,
etc. A light-weight integration infrastructure must be provided that provide
openness and easy adoption for service provider and consumer. Hereby, we can
reuse existing solution as much as possible as a means to integrate service-
orientation in open world-wide environment beyond the boundaries of single
organizations.
• Second, we need to provide semantic web technology as a means to implement a
scalable access layer to data and processes. Instead of accessing services through
syntactical means like a name or an address they become usable through a
description of their actual capability. Our objective is to make semantics a pillar
of the software architecture of the next generation of computing. We will provide
a comprehensive framework and architecture with which to augment the
worldwide movement to service-orientation with semantics in the context of
evolving, industrial Service-Oriented Architectures. A semantic service bus will
be developed as a core technology that implements service usage through
semantic descriptions.
• Third, we make usage of Web 2.0 technology as means to generate and access
this semantic service layer. Properly including human interaction and cooperation
will enable us to provide solution to certain tasks such as service ranking or
mediation that remain otherwise unfeasible. 6 Web 2.0 and human computing
approaches together with their underlying social consensus building mechanisms
have proven the potential of proper balancing services provided by humans
and services provided by automated reasoning. In the end, a service is not
necessarily being provided by a computer program and, for example, current
approaches for service discovery and (human) expert finders can be combined.

Based on these achievements, Service Web 3.0 will place computing and programming at
the services layer providing the real goal of computing--problem solving in the hands of
end users through a properly balanced cooperation approach.

In section 2, we will define the implications it has in putting SOA in the context of an
open, distributed, and decentralized world wide solution. Section 3 defines the semantic
access layer that is needed to provide scalable access to this world wide service web.
Section 4 generalizes the concept of a service, introduces the usability aspect, and shows

6
Humans are much better in solving certain tasks (capture recognition, image descriptions, common sense
reasoning, i.e., http://video.google.com/videoplay?docid=-8246463980976635143&hl=en).

6
that only balanced approach of mechanization through semantics and human intervention
can actually work. Section 5 describes some concrete usage scenarios how business
relationships can be implemented through this new technology.

2. Integration Web and SOA

The web can be understood as a collection of principles and highly scalable means for
electronic publication.

Major principles are:

• Openness implies that in principle everybody can contribute to it as a provider or

consumer of information. Openness is a major and essential necessity to ensure the
success of the platform. Usage of this infrastructure as a service provider or user must
be as simple, smooth, and unrestricted as even possible.
• Interoperability is needed and provided through the integration of different
proprietary and legacy solution through a common interface. The solution as such
must be platform and vendor neutral to enable every provider and requester of
information to join in.
• Decentralized changeability and dynamicity implies that content can appear,
becoming modified, or disappearing in a non-controlled fashion. That is, provisioning
and modification of content must be under distributed control of the peers rather
being controlled by a central authority. Central control would hamper access and
therefore scalability, an element of chaos or “messiness” must be tolerated. 7
• Central means to route requests or responses must be automized in order to scale.
Manually generated repositories are inherent non-scalable, costly, and immediately
out-dated. One could argue that web sites like google are actually a centralized
control or access. However, what they really implementing is an abstracting process
on accessing and caching information. In the early days of the web sites were
accessed by magic numbers (and latter by magic names) and list of bookmarked
pages were valuable intellectual property. Through search engines this access is
replaced by key word retrieval and ranking based on relevance factors. Therefore, the
keyword form of Google is just lifting the address bar of a browser at a higher level
of abstraction. 8
• Enabling n:m relationship to maximize interaction. In difference to email, where
the content is target to specific receivers, the web is based on anonymous distribution
through publication. In principle, the information is disseminated to any potential
reader, something mailing tries to achieve trough spam.

7
I will never forget the moment in life where a German Professor was explaining me that the Web is crap
because it can contain broken links.
8
Still, means like google actually have the potential to become misused as central access control when they
are started to be used to manipulate content access as a means to implement censorship or commercial
interests.

7
Providing an infrastructure realizing these principles requires roughly three important
means: world-wide addressing of service, world-wide delivery of service, and platform
independent access of services.
• A web of services requires a world wide addressing schema. At an intermediate
level it may be a unique name and at a more elaborated level it may be a
description of the capability of a service, i.e., the degree it can be used to achieve
a certain goal. In the case of the web, these are URIs.
• A transport layer (a protocol) to transmit request for and results of services. In
the case of the web this is http.
• A platform independent interface to process service request and access. In case of
the web this is html and browsers that interpret html.

Realizing these principles and providing a platform incorporating them is the first and
major necessity to implement Service Web 3.0.

3. Integration Semantics and SOA

There are fundamental requirements for user-centric, service oriented and distributed
environment which all together facilitate seamless provisioning of business services.
These major principles are:

Service Oriented Principle. Service-orientation represents a distinct approach for

analysis, design, and implementation which further introduces particular principles that
govern aspects of communication, architecture, and processing logic. This includes
service reusability, loose coupling, abstraction, composability, autonomy, statelessness,
and discoverability. With respect to the service orientation which enables a service level
view on the organization we further distinguish services from several views.

First, we distinguish two types of services from the point of the functionality they provide
within the architecture, namely (1) Business Services and (2) Middleware Services.
Business services are services provided by various service providers, their back-end
systems – business services are subject of integration and interoperation within the
architecture and can provide a certain value for users (e.g. purchasing a flight). On the
other hand, middleware services are the main facilitators for integration and
interoperation of business services (e.g. discovery, interoperability, etc.).

Second, we distinguish two types of services from the point of their abstraction in the
architecture, namely (1) Web Services, and (2) Services. The Web Service is a general
service which might take several forms when it is instantiated (e.g. purchase a flight)
whereas the Service is actual instance of the Web Service which is consumed by a user
and which provides a concrete value for a user (e.g. purchase of a particular flight from
Innsbruck to Vienna). We use this distinction for Business Services in the architecture.

8
Semantic Principle. Semantics in general is considered as a rich and formal description
of information and behavioral models enabling automation of certain tasks by means of
logical reasoning. Combined with service oriented principle, semantics allows to define
scalable, semantically rich and formal service models and ontologies allowing to promote
total or partial automation of tasks such as service discovery, contracting, negotiation,
mediation, composition, invocation, etc. Semantic service oriented approach to modeling
and implementation of the organization enables scalable and seamless interoperation,
reusability, discovery, composition, etc. of various Business Services.

Distributed Principle. Distributed principle is the process of aggregating the power of

several computing entities to collaboratively run a single computational task in a
transparent and coherent way, so that they can appear as a single and centralized system.
Distributed principle is applied to the architecture middleware system which allows
distributing its components over the network in a transparent way so that the execution
process run in the middleware could be scaled across a number of physical servers over
the network. Distributed principle is also applied to the Business Services allowing
running a process spanning across several Business Services distributed over the network.

User-centric Principle. The user-centric principle puts the user in the center of the
architecture. It refers to concepts like personalizing of Business Services, facilitating
service usability, promoting multi-channel access and service delivery, building trust,
achieving efficiency, accountability and responsiveness according to users’ requirements,
enabling seamless implementation of Business Processes across organizational
boundaries, etc.

Realizing these principles and providing a platform incorporating them is the major
necessity to implement Service Web 3.0. There are following four types of business
services of an infrastructure which are a must for Service Web 3.0 to deliver its promises:

• The problem-solving layer which consists of (1) Ontologies, (2) Applications

(e.g., e-tourism, e-government) and (3) Developer tools (GUI tools such as those
for engineering ontology/web service descriptions; generic developer tools such
as language APIs, parsers/serializers, converters, etc.).
• The broker layer which consists of (4) Discovery, (5) Adaptation (including
selection and negotiation), (6) Composition (web service composition techniques
such as planning), (7) Choreography, (8) Mediation ((a) Ontology mediation:
techniques for combining Ontologies and for overcoming differences between
Ontologies; (b) Process mediation: overcoming differences in message ordering,
etc.), (9) Grounding, (10) Fault Handling (Transactionality, Compensation,
etc.), and (11) Monitoring.
• The base layer that is providing the exchange formalism used by the architecture,
i.e., (12) Formal languages (static ontology and behavioral, i.e.,
capability/choreography/orchestration languages, connection between higher-
level descriptions, e.g., WSML), (13) Reasoning (techniques for reasoning over
formal descriptions; LP, DL, FOL, behavioral languages, etc.) and (14) Storage
and Communication.

9
 • Finally, vertical services such as (15) Execution management and (16) Security
(authentication/authorization, encryption, trust/certification).

The figure 1 presents the set of the business services needed for this.

Figure 1. A semantic service bus

Realizing these principles and providing a platform incorporating them is the second
necessity to implement Service Web 3.0.

4. Integration Web 2.0 and SOA

Web 2.0 has brought a collection of interesting advances to the World Wide Web. First,
we will describe the convergence of web and web service technology. Second, we will
discuss the transformation of the web from a means for individuals to publish into a
platform that organizes communities and finally as a means for social consensus building.
Realizing these principles and providing a platform incorporating them is the third
necessity to implement Service Web 3.0.

4.1 Mashing Web and Web Services

Thanks to the programming paradigm of Asynchronous JavaScript and XML (AJAX),

Web sites become more interactive and user friendly, to an extent that was previously
only achieved by desktop applications. Given the ubiquity and abundance of Internet
band with today, both stationary as well as wireless, we can see a convergence between

10
classical desktop applications and Web applications. The Web is becoming a valid
platform to host all kinds of software services (see Figure 2).

Figure 2: AJAX Architecture

AJAX adds the ability to dynamically update the appearance of a Web page to the static
“request/response” pattern between Web browser and Web application, where only whole
Web pages are displayed. Hence, users experience a more fluid and interactive Web
application.

Web Service standards such as SOAP and REST are already a prominent component of
Web 2.0. Often, external datasets are “mashed” up with a Web application using their
SOAP or REST based interfaces. Thus, Web 2.0 applications often form graphical
interfaces for (enterprise) applications based on the SOA (Service Oriented Architecture)
paradigm, which has established itself as a prominent architecture for modern backend
applications.

Mashup applications integrate data and functionality form several other Web applications
via their exposed Web APIs and provide added value through this integration, e.g.
integrating a mapping application with a phone number reverse lookup application to
display the location of the phone number on a map (see figure 3).

11
 Web APIs and Mashups

1600 4,50

1400 4,00

3,50
1200
3,00

Mashups/API
1000
2,50
Count

800
2,00
600
1,50
400
1,00

200 0,50

0 -
Sep Okt Nov Dez Jan Feb Mrz Apr Mai Jun Jul Aug Sep Okt Nov Dez Jan
05 05 05 05 06 06 06 06 06 06 06 06 06 06 06 06 07
Month
Source: FZI Analysis of programmableweb.com data APIS Mashups Mashups per API

FIGURE 3: Increasing number of Web APIs and Mashups

The dynamics of Mashups has just been unleashed. Figure X indicates the growth of
publicly available Web service APIs. From 39 mashups and APIs that have been listed in
September 2005 on programmableweb.com, we can now find 352 Web APIs that are
used in 1404 mashup Web sites in December 2006 – a growth of more than 600% in
Web APIs and almost 3600% in mashups.

New services like openkapow (http://www.openkapow.com) and Yahoo pipes

(http://pipes.yahoo.com) will further accelerate this growth in the coming years by
providing visual development environments, which allow web developers in easy point
and click interfaces to mashup and connect any service on the Web and republish the
integrated anything on the web. This addresses one of the main problems with mashups,
i.e. that it is not that easy to create a mashup.

For example, the Yahoo!'s new pipes services presents such an user interface for the
integration and republishing of RSS news feed providing a drag and drop editor that
allows people to connect internet data sources, process them, and redirect the output.
Yahoo! describes it as "an interactive feed aggregator and manipulator" that allows to
"create feeds that are more powerful, useful and relevant." While still an early
development, Yahoo! Pipes gives a glimpse on the enormous promise in turning the web
into a programmable environment for everyone.

Semantic mashup sites have to extend the simple means to visual mashup composition
that is just appearing by providing additional significant intelligent processing (A) and

12
data reconciliation (B) mechanisms .Thereby the creation of mashups should become
even simpler. With the recently defined and researched Semantic Web standards, many
mashups can even happen on the fly, and over more than just two data sources.

For example it is easy to imagine a community page that grabs the FOAF file of all
members from their respective social network pages, enrich the data about their addresses
with coordinates from a GIS server, grabs some data on concerts of their favourite bands,
and finally integrates a calendar and a map view of this mixed data to allow them to plan
their common visits to the concerts.

With a rising number of Web Apis and mashups, also a new form of mashup will appear,
namely aggregators, these are similar to mashups superficially, but they are driven by a
different goal. The primary goal of aggregating sites is to collect data from heterogeneous
and multiple sources and republish the cleaned, integrated and aggregated data at a single
point-of-access. Intelligence will be needed to discover and merge instances with the
same identity, by mapping different ontologies based on mappings created by experts and
that can be used by all the users of the aggregating page, by filtering.

4.2 Mediating and Creating Communities

A wave of individualization and democratization of the Internet is connected to the term

“Web 2.0”. So-called “social software” promotes user participation. Examples for this
social phenomenon are social networks such as myspace.com and friendster.com, but also
personal Web blogs and the contribution to common content management platforms such
as Wikis (see figure 4).

In all cases the social phenomenon is based on increased and simplified user
participation:

• In social networks it is easy to discover new friends and easy to express

characteristics about oneself or opinions about others.
• In WebLogs or “Blogs” and the creation of online content has been simplified to
a simple textbox and a submit button, abstracting from the tedious content
creation process behind creating HTML files and uploading data to a Web server.
• In Wikis, a “special” simplified syntax allows to easily interlink several
documents.

13
 Growth of Global Top Level Domains vs. Blogs
70
65
63
61 61
60 59 58
56
54 55
53
51
50
50 49
48
46

40

30
30
26
24
22
20
20 18
16
14
12
10 11
9
10 8 8
6

0
Jan 05 Feb 05 Mrz 05 Apr 05 Mai 05 Jun 05 Jul 05 Aug 05 Sep 05 Okt 05 Nov 05 Dez 05 Jan 06 Feb 06 Mrz 06

Source: FZI analysis of technorati.com and Zooknic Domain Name Count Global Top Level Domains Blogs

Figure 4: Growth of blogs vs. growth of top level domains

Again, with growth come problems that require more intelligence to be solved or elicit
the need for further automatism.

Wikipedia – the World largest encyclopedia and most prominent Wiki – contains many
lists of things to answer the information need of its users. These lists are manually edited
and notoriously outdated, as they are not linked with the content on the individual pages
that are used to form the list (see figure 5).

Figure 5: List of female tennis players (from http://en.wikipedia.org)

14
To avoid such inconsistencies, manually generated lists should be replaced by
automatically generated lists. This would be possible, if Wikis were able to expose their
content in a structured way responding to structured queries instead of text-only queries.

Structured data in Wikipedia, although visually present, cannot be queried in general, a

perceived lack of functionality that among others yields problems such as Inter-language
consistency, an issue highly relevant to multilingual Europe, is also difficult to maintain
in Wikipedia and inconsistencies are easily to spot for structured data. For example, at
time, the population of Edinburgh at the time of this writing is different in English,
German, and French Wikipedia.

Related to the lack of structured data and often a source of confusion to Wikipedia users
is the inflationary use of categories, such as “Rivers in Buckinghamshire”, “Asteroids
named for people”, and “1620s deaths” – all those categories can be easily replaced by a
handful of structured data annotations which could be queried in combination to create
such categories.

The open and flexible data models of the Semantic Web can become key to add
structured data to Wikipedia and thereby not only tackle the aforementioned problems but
also provide further intelligent information retrieval possibilities…

Bloggers are free to write about anything they want in their blogs, but most frequently
bloggers comment on things that exist on the Web. In this sense, they created focused
annotations of Web resources. Unfortunately the fact that Web sites are commented or
annotated is not understandable to anyone but the reader of the blog. For example, it may
be useful to model criticism annotations using a numerical rating and aggregate such
ratings across various blog entries from different authors to come up with a rating of a
Web site.. At the moment, blog entries that embody criticisms may contain a numerical
rating but that rating is rarely recorded in a machine-readable form. Using an ontology to
model such annotations would make it easier to do automated filtering and analyses, such
as finding the average rating for a given resource; such analyses are already done by
specialized data mining sites, but they usually do not track anything more semantic than
just the frequency with which specific sites are hyperlinked to.

Also, because blogs are managed on a per-user basis, users have the flexibility to adopt
such ontologies to mark up their annotations. Additionally, in moving blogging to the
Semantic Web, an obvious extension is allowing blogs to talk about arbitrary Semantic
Web resources. This extension not only broadens the set of available topics but also
allows resources that previously had to go unnamed (e.g., “teddy bear model #22321”) to
be identified more precisely, improving
search capabilities.

Blogs act as logs of messages, usually written to a general audience but at times focused
towards specific parties. This phenomenon is most evident when an event of interest
occurs, such as a product release. A flood of blog posts appear coincidentally, and the
debate that ensures results in blogs containing entries that comment on other blog entries.
This structure is reminiscent of email or newsgroups: a blog entry has a sender, a set of

15
(sometimes targeted) recipients, a subject line, and often one reply-to entry (or more,
although today notating that a blog entry is in reply to another blog entry may need to be
done with human readable prose and a hyperlink, depending on the blogging system in
use).

Thus, Semantic blogging needs to extend the blogging paradigm with the possibility to
add additional metadata to a blog entry. This metadata can cover a variety of aspects,
ranging from information about the entry itself, such as the author or date of publication
(using vocabularies such as RDF Site Summary (RSS 1.0)7), information about the
structure of the blog and relations to other communication sites to metadata describing
the topics mentioned in the entry (a blog entry about a meeting would e.g. include
metadata about that meeting, such as date and location, people attending or details of
project related to the meeting, whereas an entry discussing a book would include
bibliographic metadata for that book). Moreover, this metadata is expressed in a semantic
format such as the Resource Description Framework (RDF), which allows further
inferencing and machine reasoning over data.

4.3 Consensus building mechanisms

One of the gravitational principles of Web 2.0 is that we can see the Web as Platform 9 .
Going from the good old Web (now baptised as Web 1.0) to the current exciting Web 2.0
has changed our ways to perceive and interact with the Web and its content. The Web 1.0
publishing mechanism was very clear; it was conceptually designed for the publication of
documents. There were publishers of content who pushed fixed data for subscribers in a
static “one way” lane. Consequently there were little chances for feedback and even less
for consensus (e-mail contact information when available). With the apparition of Web
2.0, the style of communication changed completely and now, there is a “two ways” lane
of interaction and each resource in the Web is a vehicle for community interaction. There
are some examples in Web 2.0 where this paradigm is tested beyond any foreseen limit:
Wikipedia.

In Wikipedia, the principle is simple: anyone can contribute to the construction of

Wikipedia. “Content creation and editing are highly egalitarian; any registered user can
create a new article, and anyone, registered or not, can make any change to an existing
article, or undo anyone else’s change” 10 . Probably the most controversial and at the same
time amazing thing of Wikipedia is how humans have been able to reach consensus in
certain topics that could be controversial. People with different interests and background
have been working closely to build together contents that have been improved
continuously. In words of Jimmy Wales, founder of Wikipedia: “If you write something

9
What is the Web 2.0? Design Patterns and Business Models for the Next Generation of Software, by Tim
O'Reilly
10
Enterprise 2.0: The Dawn of Emergent Collaboration, by Andrew P. McAfee

16
that annoys other people, it is just going to be deleted. So if you want your writing to
survive, you really have to strive to be cooperative and helpful” 11 .

Wikipedia works by building consensus (see Figure 6). Consensus is an inherent part of
the wiki process. The basic process works like this: someone makes an edit to a page, and
then everyone who reads the page makes a decision to either leave the page as it is or
change it. Over time, every edit that remains on a page, in a sense, has the unanimous
approval of the community (or at least everyone who has looked the page). "Silence
equals consent" is the ultimate measure of consensus — somebody makes an edit and
nobody objects or changes it. Most of the time consensus is reached as a natural product
of the editing process 12 .

When there are disagreements, they are resolved through polite discussion and
negotiation, in an attempt to develop a consensus. If Wikipedia finds that a particular
consensus happens often, it is written down as a guideline, to save people the time having
to discuss the same principles over and over. Normally consensus on conflicts is reached
via discussion on talk pages. In the rare situations where this doesn't work, it is also
possible to use the Wikipedia:Dispute 13 resolution processes, which are designed to assist
consensus-building when normal talk page communication gets stuck.

Figure 6. Wikipedia consensus process flowchart

11
Best of the Web: Extreme Blogging, by M. Rand
12
http://en.wikipedia.org/wiki/Wikipedia:Consensus
13
http://en.wikipedia.org/wiki/Wikipedia:Dispute_resolution

17
If we want to combine billions of services and detect from their behaviour if they are
reliable (just because somebody says that a service is reliable it does not means it is true)
we need to define measurable ways to combine human trust with computer trust. In a
world in which humans still play an important role in services (mouth to mouth and
human feedback are still more accurate that QoS metrics) we are bound to integrate social
skills in service transactions.

In a transactional electronic world when some service is failing it has to undo previous
atomic actions, in order to ensure that nothing has changed since the service was
previously ordered to execute. Compensation, substitution, agreement are terms that
Business Services are introducing to capture the human ability to reach consensus: the
electronic handshake.

Some online marketplaces, like the academic Kashba, the commercials onSale or eBay
introduce collaborative reputation mechanisms to help users to decide on what risk are
decided to give to their transactions based on the reputation of the other party. If some
partner fail to fulfil his agreement, his reputation is damaged and future transactions may
be jeopardized. In the other hand, users that commits to their words and are honest, enjoy
from a better reputation and can make deals more often.

It is a reality that humans trust more those people they know, especially if they know
them for good. The net of contacts around the Web is incredibly dense and with the help
of initiatives like FOAF, people are able to annotate information about them and about
the people they know, and usually, trust. The use of initiatives like FOAF in the field of
Services will allow software agents, and humans, to know something more about the
reliability and reputation of a service. If some service I know and use (e.g. my flight
booking service) has been combined, successfully, with some geographical location
service, I will have more information about its reliability than before. At the end, this is
all about trust and reputation, from human to computers, computers to computers and
human to humans.

5. Scenarios
Web Services are the major building block in the developing evolution of the Internet
from a network of information to a network of services. The amount of services which
will be offered on the Internet is expected to rise dramatically in the next few years,
necessitating the emergence of a new kind of search engine: a Service Broker, which can
offer suitable Web Services to interested Customers, who will then use these Services in
their own integration solutions (thus becoming the Service Requestors in the terms of the
SOA paradigm).

The scenario described in this section shows the results of applying these technologies in
a specific area: full scale e-Commerce solutions usable for everybody. Due to the

18
envisioned availability of a large set of e-Commerce and content business services, which
can be combined to cover all needed functionalities of such an integrated e-Commerce
solution, the emergence of two new market areas can be expected: a B2C and a C2C
market.

The B2C area will cover Service Providers offering a collection of business services
which can be used by a Customer to build his or her own e-Commerce solution. The
Service Provider does not directly offers these services, but rather requests a Service
Broker to provide the infrastructure for a customer to find, purchase and consume the
offered services. The Broker can also offer the possibility to combine services from
several providers and offer suitable service packages, covering a wide range of
functionalities (e.g. including web shop solutions, payment services, monitoring and
marketing services etc.).

Additionally a C2C market will evolve: Customers use their business services bought via
a Service Broker to create their own e-Commerce solutions which they can use to
perform e-Commerce with other consumers in a flexible and dynamical manner, while
simultaneously lowering the acceptance threshold of consumers to peruse such solutions.
Typical customer needs to set up their own e-Commerce solution include the sale of used
cars, selling or trading collectibles and offering photo and multimedia content to other
users. Thus the vision of flexible C2C e-Commerce, where every interested party can
easily make offers and buy products is fulfilled.

These trends will radically change the existing networked solutions of large service
providers. For example, BT is undergoing a complete network transformation. At the
heart of this transformation is the £10 billion 21st century network (21CN) initiative.
21CN is an IP-based, multi-service network, and is the driver for a radical change. As an
end-to-end Internet Protocol (IP)-based network, 21CN will consolidate BT's 16 separate
network platforms into one. It will replace the complex network and systems
infrastructure with a physically simpler and more reliable network, to ensure the delivery
of the next generation of converged services faster, more efficiently and more cost-
effectively. As part of the massive 21CN programme, BT have launched a project named
Web21C.

Web21C is the name of the programme to launch new software based services from BT
and 3rd parties, and it builds on BTs investments in 21CN. Web21C allows 3rd parties to
use BTs network as a platform for delivery of their services, for which we get revenue.
These are not typically other network competitors, but a new breed of partner - software
companies; developers and content providers. These partners can use tools we provide to
take advantage of network and other assets we have, for a share of the revenue, or other
source.

Critical to the success of Web21C will be the use of Common Capabilities. Common
Capabilities refers to a set of common re-usable components designed to underpin BTs
product set. An example of a Common Capability would be 'Authentication' which is
used for validation of customers. Third parties will now be able to use this set of re-

19
usable ‘off-the-shelf’ components to build their own solutions for customers. This service
will not just be limited to consumers. Increasingly, small and large business, as well as
government, is going towards a "shared services" model where applications and content
are provided to them on a subscription basis from a supplier.

Benefits for customers are that they get a company dedicated to helping them thrive in
this changing world with a much wider range of services to choose from. Complementary
benefits for BT are that it can serve its customers in new and innovative ways; extend its
markets, and grow its business.

20