From CBT to eLearning

From CBT To eLEARNING S. Stoyanov, I. Ganchev,, I. Popchev, M. O’Droma 

eCommerce Lab, Department of Computer Systems, University of Plovdiv, Plovdiv, BULGARIA {stani,i.popchev}@ecl.pu.acad.bg, phone: + 359 32 277292, fax: +359 32 945736 

Department of Electronic and Computer Engineering, University of Limerick, National Technological Park, Limerick, IRELAND {ivan.ganchev,mairtin.odroma}@ul.ie, phone: + 353 61 202248, fax: + 353 61 338176

Abstract

Two basic concepts, significant for the development of software in support of the learning process, are formally presented, distinguished and compared. The first, Computer Based Training (CBT), is used as a starting point for the development of a means for learning support. The second,eLearning, is used as a reference point for long-term research and development of education and learning using the full potential of electronic media and technology. The paper includes a demonstration of an approach for the development of information systems for education, based on clear differentiation between these two concepts. Key words: Computer based training (CBT), eLearning, eServices, Distributed eLearning Center (DeLC), object-oriented, agent-oriented, concept model, infrastructural model, service model, Consumer-based Business Model (CBM), InfoStations, Semantic Web, Grid. 1. INTRODUCTION In recent years the use of Information and Communication Technologies (ICT) in education has become an area of ever growing research and development (R&D) interest as well as a topical application area. Different terms and notions used in the specialized bibliography sources are confusing in many cases and do not fully express the essence of the problems and the complexity of the tasks that must be solved when creating automated means for the support of eLearning process. In many cases the challenges and the problems are simplified, which hampers significantly the development of effective, added-value, eLearning systems. The quick solutions offered often are without real benefit for the learning process itself. Aiming at the development of adequate automated ICT means for the effective support of eLearning as well as seeking new approaches, models, and architectures that could facilitate it, in this paper we consider two basic concepts. The first concept ‘Computer Based Training (CBT)’ can be used as a starting point for the development of means for eLearning support. The second concept ‘eLearning’ can be used as a target serving as a reference point for long-term R&D. In addition we demonstrate our approach for the development of information systems for education, based on clear differentiation between the above two concepts. 2. CONCEPT MODEL FOR ELECTRONIC MEDIA AND TECHNOLOGY EDUCATION SUPPORT The concept model depicted in Figure 1 is the basis of our research approach to elaborate a suitable infrastructure for the support of eLearning. In general the two main processes in each

1

automated environment supporting eLearning is the creation (or generation) and interpretation of an electronic content (eContent). These two processes must be managed in the context of mainly three models – domain model, student model, and pedagogical model (includes the educator model as well).

Domain Model

Student Model

Pedagogical Model

Generation

eContent

Interpretation

Fig.1. Concept model for electronic media and technology education support 3. CBT vs. eLEARNING In this section we present these two basic concepts and their comparative characteristics, which in our opinion are of significant importance for the development of software for the support of learning process. CBT is an attempt to automate education, replace an educator, and develop self-paced learning. The CBT focus is primary on electronic recorded education. So this kind of learning is time/place/content predetermined learning. eLearning has its origins form CBT. The focus of eLearning is not only on education, but also on education without barriers of time and distance, and customized to user’s and business’ needs [6]. The eLearning is just-in-time/at-work-place/customized/on-demand process of learning [27]. It is essential to understand that the differences between CBT and eLearning are not just semantic. There are eight distinct ways (Table 1) in which computer-based training and eLearning differ [12,37].

Delivery mode Responsiveness

CBT Push (Instructor determines agenda)

Table 1. CBT vs. eLearning eLearning Pull (Student determines agenda)

Anticipatory

Reactionary

(Assumes to know the problem)

(Responds to problem at hand)

Linear (Has defined progression of knowledge)

(Allows direct access to knowledge in whatever sequenced makes sense to the situation at hand)

Asymmetric

Symmetric

Non-linear Access Symmetry

2

(Training occurs as a separate activity)

Modality

(Learning occurs as an integrated activity)

Discrete

Continuous

(Training takes place in dedicated chunks with defined starts end stops)

(Learning runs in parallel and may never stop)

Centralized

Distributed (Content comes from the interaction of the participants as well as the educators)

Authority

(Content is selected from a library of materials developed by the educators)

Mass produced

Personalized

Personalization

(Content must satisfy the needs of many users)

(Content is tailored to the individual users needs)

Static Adaptability

(Content and organization/taxonomy remains in their original authored form without regard to environmental changes)

Dynamic (Content changes continuously through user input, experiences, new practices, business rules and heuristics)

A classification of the main types of an education (as regards the content and the context) supported by electronic means is shown in Figure 2. 4. eLEARNING CHALLENGES In order to make the eLearning a reality, a number of research challenges need to be addressed. These include: ▪ New architectures – providing better flexibility especially on server side, i.e. serviceoriented architectures such as Semantic Web, GRID (Semantic GRID); ▪ Multi-communication access to information sources – using different types of end user devices for access to learning resources, e.g. mobile terminals with added intelligence for anywhere-anytime-anyhow access; ▪ Agent-based approaches – providing additional flexibility and opportunities for machine interpretation of semantic and context-dependent information, as well as providing support for the process of personalization and customization; ▪ Knowledge technologies – using different intelligent models for knowledge processing and control (both for client and server sides). In particular the use of ontologies and combinations with intelligent agents give additional flexibility and new opportunities for semantic- and context-oriented data interpretation; ▪ Application integration – inclusion of inherited systems especially for the integration of electronic services (within architectures oriented towards electronic services, eServices); also inclusion of different external models such as pedagogical models, user models etc; ▪ Profiling – flexible processing and management of different profiles – users’, services’, educational units’, etc.; ▪ Use of existing standards – e.g. for open architectures, communication protocols, eContent (e.g. SCORM ), etc.

3

Just-in-Time eLearning

Self-paced CBT Personalized

Generalized

Content-axe Distance Learning

Academic Education

Context-axe Just-in-Case

Fig. 2. A classification of the main types of an education (as regards the content and the context) supported by electronic means The creation of real-functioning eLearning systems used for distance/remote education is a complex problem for the solution of which besides the variety of technological aspects the following factors too have to be taken into account: ▪ Education must follow certain pedagogical approach and model; ▪ Education must be personalized, i.e. tailored to the individual needs of each student; ▪ National traditions and particularities may influent significantly the educational process particularly in certain courses/modules like national history, national geography, folklore etc. Another difficulty is related to the different nature of influential factors, which predetermines their possible level of formalization and degree of technological and software-technical support. 5. EMERGING ARCHITECTURES It is envisages that two types of emerging eServices-oriented architectures can provide the needed flexibility: Semantic Web architecture and Grid architecture. 5.1. Semantic Web Architectures The main challenge in transition from today’s Web to the Semantic Web is in relation to the engineering and technology adoption nature of the problem rather than to the scientific one [4, 38]. Partial solutions to all-important building blocks of the Semantic Web exist already, e.g. knowledge management systems, business-to-business eCommerce and business-to-consumer applications, intelligent and personal agents. At present, the main problems are: the integration, the standardization, the development of supporting tools, and the adoption by users of all these rational solutions.

4

A layered approach is adopted for development of the Semantic Web, which allows step-by-step building of one layer on the top of another. This way it is easier to achieve consensus on small steps during the integration process. In doing this two principles should be followed: ▪ Downward compatibility – if an agent is aware of a layer, it should be able to interpret and use information from the lower layers; ▪ Upward partial understanding - if an agent is aware of a layer it should take at least partial advantage of information at higher levels. The standard Semantic Web architecture includes six basic layers. The XML layer, which provides a language for structured Web documents with a user-defined vocabulary, lies at the bottom. The next RDF and RDF Schema layer provides a basic data model for writing simple statements about Web resources and organization of hierarchies. The Ontology layer provides powerful languages and tools for representation of more complex relationships between Web objects and resources. The Logic layer is used to enhance the ontologies by rule-based languages and inference machines in order to process application-specific declarative knowledge. The Proof layer involves the actual deductive process as well as the representation of proofs in Web languages and proof validation. The most upper layer in the Semantic Web architecture is the Trust layer. This emerging layer uses digital signatures and other kind of knowledge, based on recommendations by trusted agents or rating of certification agencies and consumer bodies. 5.2. GRID Architectures GRID has emerged as a distributed computing infrastructure for advanced science and engineering [15]. The real and specific problem that underlies the Grid concept is coordinated resource sharing and problem solving in dynamic, multi-institutional virtual organizations [16]. The establishment, management, and exploitation of dynamic, cross-organizational virtual organization relationships require new technologies. The open Grid architecture organizes components into different layers where components within each layer share common features but can build on capabilities and behaviors exposed by any lower layer. The specification of the various layers follows the principles of the “hourglass model” [32]. The narrow neck of the hourglass defines a small set of core abstractions and protocols (e.g. TCP and HTTP in the Internet), onto which many different high-level behaviors can be mapped (the top of the hourglass), and which themselves can be mapped onto many different underlying technologies (the base of the hourglass). By definition, the number of protocols defined at the neck must be small. In the Grid architecture, the neck of the hourglass consists of Resource and Connectivity protocols, which facilitate the sharing of individual resources. Protocols at these layers are designed so that they can be implemented on top of a diverse range of resource types, defined at the Fabric layer, and can in turn be used to construct a wide range of global services and application-specific behaviors at the Collective layer – so called because they involve the coordinated (“collective”) use of multiple resources. The Semantic GRID [10] is an extended architecture in which additional layers are integrated. The new layers enhance the GRID for powerful processing of the information semantics. In respect to computation and semantics there are close relationships shown on the Fig.3. [11].

5

Semantic Web

Semantic Grid

Web

Grid

Richer Semantics

Greater Computation

Fig.3. The Semantic Grid

6. DISTRIBUTED eLEARNING CENTER (DeLC) Established as a collaborated project between the University of Limerick and the University of Plovdiv, the Distributed eLearning Center (DeLC) aims to provide a distance eLearning / eTeaching facility available at any place and at any time to individuals and groups of students / educators both in synchronous mode (on-line) and asynchronous mode (off-line). The DeLC project focuses at the development of a common concept for the creation of eLearning information systems and a theoretical and conceptual base of service-oriented eLearning infrastructure for the integration of eServices. A significant part of this project is dedicated also to the development of a suitable technological environment and architecture that is independent from the embedded eServices. The DeLC is a complex and complicated project for realization of which we adopted a decomposition approach, which includes the following main steps: ▪ Abstract specification of DeLC – the DeLC specification includes the description of following three models: Infrastructural Model, Services Model, and eLearning Node Model; ▪ Object-oriented DeLC – the aim of this step is to realize object-oriented and eServices oriented architecture, suitable for the CBT support and used as a basis of the future eLearning-oriented DeLC; ▪ Agent-oriented DeLC - this is the last phase of the DeLC implementation, which will satisfy to a great extend the requirements for an eLearning-oriented system. This version will be developed by means of re-engineering of the object-oriented DeLC. 6.1. DeLC Infrastructural Model The DeLC Infrastructural Model (Figure 4) specifies the basic building blocks of the DeLC. Furthermore it characterizes all the possibilities for integration and management of eServices within the defined clusters. An initial DeLC infrastructural model was proposed in [35,36] consisting of DeLC Nodes established and supported by real administrative units offering a complete educational cycle (e.g. laboratories, departments, faculties, colleges, universities). The enhancement of the DeLC for the provision of mobile services improves organizing and functioning of the entire eLearning/eTeaching process within a University Campus. An initial

6

outline proposal for this was given in [17, 18]. We plan to develop the DeLC architecture as a consumer-oriented one in line with the trend in mobile communication services based on the Consumer-based Business Model (CBM) [28, 29]. This model will be a much more pro-active business driver for the evolution of next fourth generation (4G) wireless world vision [30], which will provide for users access network choices, price-performance choices etc. to enable an always best connected and an always best served (ABC&S) wireless networking. When applied to eLearning/eTeaching, the CBM will provide flexible opportunities for reach m-Learning/mTeaching environment. The CBM eliminates the disadvantages of the Subscriber-based Business Model (SBM), which is widely used today but seems not suitable for future 4G mobile communications where users will want to act more as consumers seeking better value for their money and not as restricted (as regards the access and services) subscribers. In CBM the users get services much like shoppers entering this or that shop, buying goods and paying by their credit cards. A key element is that the CBM model separates out the administration and management of consumers’ one-stop-shop authentication and accounting system from the business of supplying services, and locates it with a third-party AAA SP1 (institutions, such as present day credit card companies). A ready-made application is mobile wireless access to full eLearning education. Our pilot system development would start with InfoStations (free & wideband on universities campuses!), then expand into 2G, 2.5G and 3G systems, then through an integrative and a re-structuring process described herein effect a transition to “consumer-driven” environment of integrated heterogeneous networks. For this the DeLC network model first has to be extended to a 3-tier structure by an introduction of additional entities (e.g. InfoStations, Intelligent Redirectors, Profile Managers, Intelligent Agents acting as personal helpers for users, etc) needed for the provision of intelligent mobile services. This will effectively be a pilot infrastructure providing to the user efficient seamless and ABC&S wireless eLearning services delivery. We suggest that research and development should be made on the basis of intelligent (especially from a communications viewpoint) mobile terminal, which is matched by intelligent service provider (e.g. an intelligent DeLC in the pilot scheme proposed here). In order to support also mobile eServices a so-called Expanded DeLC Infrastructural Model is developed [19, 20], which has a 3-tier structure (shown in Figure 4), consisting of: ▪ Mobile devices - such as: cellular phones (e.g. GSM2 [22], GPRS3 [21], UMTS4 [39]); Personal Digital Assistants (PDAs); laptops/notebooks; vehicle communication terminal systems (VCTS). ▪ Information Stations (InfoStations), [42] - deployed on the University key points and providing network access (e.g. WLAN5 [41], Bluetooth6 [7], UWB7 [40]) for mobile users with wireless devices. The InfoStations accept requests from mobile devices and forward them to an InfoStations’ center for further processing. The InfoStations are used for urgent messages downloading, Internet caching, synchronization of off-line eLearning process with on-line eLearning system (e.g. tracking of a learner's progress, sending of asked off-line questions to educators, receiving of answers, sending of other relevant off-line information, such as test scores, time spent on task etc.). The InfoStations obtain all updating information from an InfoStations’ center. ▪ InfoStations’ center - implemented as a server module in one of the DeLC nodes. This 1

Service Provider of Authentication, Authorization and Accounting services. GSM - Global System for Mobile communications (2G – 2nd generation of mobile communications) 3 GPRS - General Packet Radio Service (2.5G – 2.5 generation of mobile communications) 4 UMTS - Universal Mobile Telecommunications System (3G – 3rd generation of mobile communications) 5 WLAN – Wireless Local Area Network 6 Bluetooth - an industrial specification for wireless Personal Area Networks (PANs). 7 UWB – Ultra-Wide Band 2

7

center controls all InfoStations and provides updating and synchronizing information. It contains also Intelligent Redirectors (e.g. needed to redirect a message or incoming phone call to the current user location and most appropriate user terminal used at the moment as specified in the user profile) and corresponding Profile Managers. Virtual DeLC Unit (at University Campus)

DeLC Node (DeLCN)

InfoStations' Center with Intelligent 3rd tier: Redirectors & Profile Managers 2nd tier:

InfoStations

1st tier:

Mobile Devices with Intelligent Agents

Fig. 4. The 3-tier DeLC Infrastructural Model The DeLC communication infrastructure corresponding to the Expanded DeLC Infrastructural Model is depicted in Figure 5.

Fig. 5. The DeLC Communication Infrastructure 5.2. DeLC Service Model The DeLC Service Model serves the development of the system managing the e-Service invocation, which will be a constituent part of the DeLC node’s run-time module. The Service

8

Model is open for extensions including new values of already specified dimensions or entirely new dimensions. In order to separate the DeLC architecture from the offered eServices, the Service Model is developed as an infrastructure-independent one. The Service Model consists of two main components: eServices characteristic space (meta-model) and subject models. The characteristic space classifies the DeLC eServices within an n-dimensional discrete hyperspace. For the moment four characteristic dimensions are defined: ▪ Content/nature – a classifier of services according to their content/nature, i.e. eLearning, eEducator, Administration, and System services; ▪ Invocation - a classifier of services according to the activation option provided in the infrastructure, i.e. Local, Remote, and Back-End services; ▪ Mobility – specifies whether the service is a mobile one or a stationary one; ▪ Standard-conformance – whether the service is conformant (or not) to some eLearning standard, e.g. SCORM [1], ARIADNE [2], AICC [5], IEEE LTSC [24,31], IMS [25,26], and CEN/ISSS [13]. For instance, according to this service model the DeLC service Intelligent Phone Call (presented in section 4.2) is an eEducator, local, mobile service and not necessarily conformant to any eLearning standard (Figure 6). Invocation Local

Intelligent Phone Call

Remote BackEnd

ARIADNE IEEE CEN/ LTSC IMS StandardISSS conformance

AICC SCORM

e-Learning e-Educator Administration System Content/nature

Mobile Stationary

Mobility

Fig. 6. The discrete hyperspace of DeLC services

5.3. DeLC Node Model In the DeLC concept the DeLC nodes are the main components needed for the configuration of any eLearning system. These nodes are models of real organizations/institutions certified to provide education. The main task of the DeLC nodes is to provide different services to users/consumers of a DeLC system. The DeLC Node has to support the integrated eServices and solves the following problems: ▪ Analysis and management of the users requests; ▪ Localization of the required services and their mapping to the actual user request; ▪ Service customization and personalization; ▪ Service activation; ▪ Control over service processing and execution – single services and transactions; ▪ Sending back the results. The main part of the functionality of the DeLC Node is developed as a set of system services (in relation to the classification used by the Service Model). This group of services is integrated into the architecture of the DeLC Node as a constituent of the run-time module. In order to complete

9

the above tasks we propose a DeLC Node Architecture, which includes the main components shown in Figure 7. Currently we develop a concept for the creation and support of clusters, within which an integration and remote invocation of eServices will be possible. 7. OBJECT-ORIENTED DeLC The first DeLC version (Figure 7), which is object-oriented and client-server oriented, placed special emphasis upon the solution of infrastructural and architectural problems. This version aimed at the creation of a basic DeLC infrastructure and realization of the main eServices provided by the distributed center. The functionality was decomposed and represented by a set of electronic services. The support of the eLearning standard SCORM was of first priority. Another task was the development of initial pedagogical and consumer models and their integration in the infrastructure. Very important here was to find a suitable mappings of these models into the SCORM specification. In an addition the first version of eLearning authoring tool is developed based on an adapted version of the Reload editor [33]. The object-oriented DeLC can be defined as eLearning-oriented CBT education system.

Portal Server

Reload Editor

Presentation Layer

Users

6

1

eServices control Publ. eLServices

eL-Services (SCORM)

3 2

Integration

5 4

DB

Back-end systems

Remote eLn Node

Fig. 7. The DeLC object-oriented and client-server oriented architecture

8. AGENT-ORIENTED DeLC

10

The second DeLC version (an agent-oriented one) aims at the reinforcement of the infrastructure in order to transform it in an effective technological framework for the realization of sophisticated eLearning scenarios. For this the infrastructure will be extended with a system for the control and support of mobile eServices. Special emphasis is laid upon ensuring semantic compatibility (through the use of ontologies) and needed flexibility (through the use of intelligent agents). In this version we plan to develop an authoring intuitive and user-friendly tool for the creation of eLearning content. For the supporting of real eLearning and eTraining a significant enhancement of the flexibility of the existing DeLC architecture is required along with introduction of possibilities for intelligent interaction and interpretation of the data and contents exchanged between the different parties involved in the process of execution of eServices. To achieve this goal, we are developing the second version of the DeLC system architecture on the basis of the first version (in which services are implemented as Web-services) by using a re-engineering process, described below, which will lead to more open environment, supporting context-based discovery and access to user’s personal information. The flexibility and intelligence of the system will be enhanced through an introduction of intelligent agents, which will communicate with the functional modules, implemented as Web-services. For the interaction between the agents and the Webservices we use the DAML-S (OWL-S) specification [3,8,9] because it offers a good opportunity for the realization of software architecture with sufficient flexibility and also provides a suitable environment for the support of a variety of mobile services. Portal Server Presentation Layer

Users

Personalization (user, pedagogical model)

eServices control Ontologies

Publishing and distribution of eLn Services

Domainoriented Development Environment

Integration Local Services

InfoStation

IS Center

Collaboration Search

Backend systems

eLn Services

DB

Categorization

Remote DeLC Node

Fig. 8. The DeLC agent-oriented architecture According to the DAML-S specification, each service could be described in three abstract levels:

11

  

Service profile, which describes what the service performs (including information about the service’s inputs, outputs, preconditions, and other features, that can be used for advertising, discovery, and matchmaking of the appropriate service); Service model, which shows how the service works. This is an expansion and more detailed specification of the service profile; Grounding, which describes how the service can be used and also specifies a communication protocol that can be used for the direct activation of the service. Possible groundings include: SOAP [43, 44], Java remote call, KQML [14, 45], CORBA IDL [34, 46].

The development of the agent-oriented version is supported by the re-engineering process, which aims to transform the DeLC architecture into a Semantic Web (Figure 8) for education. In order to reach this ambitious goal we propose a transformation scheme, which consists of the next six activities marked in Figure 7: 1. Extension of the presentation layer with personalization layer which includes the implementations of the pedagogical and user model 2. Implementation of new service layers which fit as containers for intelligent services 3. Development of DeLC ontologies 4. Implementation of the InfoStation Center which will support a DAML-S (OWL-S)–based communication between the intelligent agents and available eServices on the server site as well as dynamic profile processing and monitoring of the activated eServices. 5. Implementation of InfoStation and appropriate personal mobile agents 6. Build up a development environment on base of Reload editor. Two approaches will be investigated for the development of the transition to the agent-oriented DeLC version: ▪ ‘Thin agent’ approach - server parts of eServices (functionality) interpreted as standalone autonomous agents. In this case they can autonomously process the user requests coming through the personal assistants. Before finalizing the request for a particular eService, a personal assistant must first negotiate arrangements with the appropriate server for all aspects of the rest of service execution. The personal assistants will mainly process the users’ profiles, whereas the server agents will process the eServices’ profiles and their models (if needed), which are DAML-S (OWL-S) records; ▪ ‘Thick agent’ approach - server parts of eServices considered as a dynamic extension/enhancement of the functionality of the personal assistants. The personal assistantscan activate the execution of the server functionality (‘adopt’ the functionality) by means of the server part of respective e-Service. These will make a decision about what e-Service has to be activated. In this case the personal assistants have to process not only the users’ profiles but also the eServices’ profiles and eServices’ models (DAML-S records). The advantage of this approach is that the personal assistantsare in possession of finer granularity of control. For instance if there are different ways for processing of a particular request, a personal assistant (i.e. an intelligent agent) can pick up the one which better suits the user needs (specified in the user profile). 9. CONCLUSION AND FUTURE WORK We intend to develop the proposed approach in a three-year period. The current developmental phase is focused on the basic modules of the DeLC object-oriented version. The portal is implemented employing the JetSpeed framework [47], which uses XML based meta-structures (e.g. users’ profiles and services’ profiles). For realization of eLearning services a computational model, compliant with Web Services [23, 48], has been specified.

12

Additionally planning of the re-engineering process for transition to the agent-oriented version has commenced. A fundamental point is a possible distributed treatment of the DAML-S specification, where the exact scheme of distribution will depend on the chosen approach (thin or thick agents). In both cases however the third abstract level of the services will be supported and processed in the DeLC nodes / InfoStations’ Center, because the run-time module of the nodes is aware of the physical location of the services, and activates and controls their actual processing. 10. ACKNOWLEDGEMENTS

This publication has emanated partly from the research conducted with the financial support of the Bulgarian Ministry of Education and Science under Research Projects “A model and an architecture for mobile user-oriented eLearning services” Ref. No. ВУМИ-101/2005 and “ Intelligent Portal for the Second School in Bulgaria” Ref. No. МИ1508/2005 and partly from the research conducted with the financial support of the Ireland’s HEA Targeted Funding Program ‘Technology in Education’. 11. REFERENCES 1. ADVANCE DISTRIBUTED LEARNING INITIATIVE (ADL) (2001), Sharable Content Object Reference Model (SCORM) 2004. 2nd Edition, http://www.adlnet.org/scorm/index.cfm 2. ALLIANCE OF REMOTE INSTRUCTIONAL AUTHORING & DISTRIBUTION NETWORKS FOR EUROPE (ARIADNE), ARIADNE Educational Metadata Recommendation, http://www.ariadne-eu.org/ 3. Ankolekar, A., M. Burstein, J. Hobbs, O. Lassila, D. Martin, S. McIlraith, S. Narayanan, M. Paolucci, T. Payne, K. Sycara, H. Zeng, “DAML-S: semantic markup for web services,” In I. Cruz, S. Decker, J. Euzenat, D. McGuiness (eds.) Proc. of the First Semantic Web Working Symposium (SWWS’01), Stanford, pp. 411-430, July 2001. 4. Antoniou, G., F. Harmelen, A Semantic Web Primer, The MIT Press, Cambridge, 2004. 5. AVIATION INDUSTRY CBT (COMPUTER BASED TRAINING) COMMITTEE (AICC), AICC Guidelines and Recommendations (AGR's), http://www.aicc.org/pages/down-docsindex.htm 6. Barker, P. Designing Teaching Webs: Advantages, Problems and Pitfalls, Educational Multimedia, Hypermedia & Telecommunication, Association for the Advancement of Computing in Education, Charlottesville, VA (2000), pp. 54-59. 7. Bluetooth: http://www.bluetooth.com/ 8. DAML Joint Committee: DAML + OIL Language, 27 March 2001, http://www.daml.org/2001/03/daml+oil-index.html. 9. DAML Services Coalition, “DAML-S: web service description for the semantic web”, First International Semantic Web Conference, ISWC’02, Sardinia, Italy, LNCS 2342, pp. 348-363, 2002. 10. De Roure, D., N. Jennings, N. Shadbolt, Research Agenda for the Semantic Grid: A Future e-Science Infrastructure, Report commissioned for EPSRC/DTI Core e-Science Programme, e-Science.dti, December 2001 11. De Roure, D., M. A. Backer, N. Jennings, N. Shadbolt, The Evolution of the Grid, www.semanticgrid.org/documents/evolution/ 12. Drucker, P., Need to Know: Integrating eLearning with High Velocity Chains, A Delphi Group White Paper, http://www.delphigroup.com/pubs/whitepapers/

13

13. European Committee for Standardization / Information Society Standardization System, http://www.cenorm.be/cenorm/businessdomains/businessdomains/isss/index.asp?pClose= 2 14. Finin, T., J. Weber, G. Wiederhold, M. Genesereth, “Specification of the KQML agentcommunication language.” The DARPA Knowledge Sharing Initiative External Interfaces Working Group, University of Toronto, 1994. 15. Foster, I., Kesselmann, C. (eds). The Grid: Blueprint for a New Computing Infrastructure, Morgan Kaufmann, 1999. 16. Foster, I., Kesselmann, C., Tuecke, S. The Anatomy of the Grid: Enabling Scalable Virtual Organizations, International Journal of Supercomputer Applications and High Performance Computing, 2001. 17. Ganchev, I., S. Stojanov, M. O’Droma, I. Popchev. “Enhancement of DeLC for the Provision of Intelligent Mobile Services”, Proc. of the 2nd International IEEE Conference on Intelligent Systems (IS’2004), vol. 1, Pp. 359-364, 22-24 June 2004, Varna, Bulgaria. IEEE Cat. No. 04EX791. ISBN 0-7803-8278-1. Library of Congress 2003115853. 18. Ganchev, I., M. O’Droma, S. Stojanov, I. Popchev, “Provision of mobile services in a distributed eLearning center”, Proc. of the International Conference on Automatics and Informatics, Pp. 79-82, Sofia, 6-7 October, 2003. 19. Ganchev I., S. Stojanov, M. O’Droma. “Consumer-oriented DeLC Service Architecture”. In Proc. of the 3rd International Conference on Education and Information Systems, Technologies and Applications (EISTA 2005), vol. 2, Pp. 213-218, 14-17 July 2005, Orlando, Florida, USA. ISBN 980-6560-34-5. 20. Ganchev I., S. Stojanov, M. O’Droma. “Mobile Distributed eLearning Center”. In Proc. of the 5th IEEE International Conference on Advanced Learning Technologies (IEEE ICALT’05), Pp. 593-594, 5-8 July 2005, Kaohsiung, Taiwan. DOI 10.1109/ICALT.2005.199. 21. GPRS: http://www.gsmworld.com/technology/gprs/intro.shtml 22. GSM World: http://www.gsmworld.com/index.shtml 23. Hendricks, M., B. Galbraith, J. Milbery, T. Modi, A. Tost, A. Toussaint, S. J. Basha, S. Cable, professional Java Web Services, Wrox, 2001 24. IEEE LEARNING TECHNOLOGY STANDARDS COMMITTEE (IEEE LTSC) (2002), IEEE Standard for Learning Object Metadata (LOM), http://ltsc.ieee.org/wg12/par1484-121.html 25. IMS GLOBAL LEARNING CONSORTIUM, INC. (2001), IMS Learning Resource Meta-Data XML Binding, http://www.imsglobal.org/metadata/imsmdv1p2p1/imsmd_bind v1p2p1.html 26. IMS GLOBAL LEARNING CONSORTIUM, INC. (2003), IMS Content Packaging, http://www.imsglobal.org/content/packaging/index.cfm 27. Maurer, H., Sapper M.E-Learning Has to be Seen as Part of General Knowledge Management, In Proceedings of ED-MEDIA 2001 World Conference on Educational Multimedia, Hypermedia & Telecommunications, Tampere, AACE, Chalottesville, VA (2001), pp 1249-1253. 28. O'Droma, M. S. and I. Ganchev. "Enabling an Always Best-Connected Defined 4G Wireless World", Annual Review of Communications, Volume 57 (Chicago, Ill.: International Engineering Consortium), ISBN: 1-931695-28-8, pp. 1157-1170. 2004. 29. O’Droma, M. and I. Ganchev. “Techno-Business Models for 4G” (invited paper), Proc. of the International Forum on 4th Generation Mobile Communications, Pp. 3.5.1-30, 2021 May 2004, King’s College London, London.

14

30. O’Droma, M. S. “Wireless, Mobile and Always Best Connected.” IEEE Communications Magazine, GCN, Vol. 42 No.1. Jan 2004. pp. 29-32. 31. O’Droma M., I. Ganchev, F. McDonnell. 2003. “Architectural and Functional Design and Evaluation of e-Learning VUIS based on the proposed IEEE LTSA Reference Model”. The Internet and Higher Education, Volume 6, Number 3, Pp. 263-276, An Elsevier Publication, Pergamon Press, September. ISSN: 1096-7516. 32. Realizing the Information Future: The Internet and Beyond, National Academy Press, 1994, http://www.nap.edu/readingroom/books/rtif/ 33. Reusable eLearning Object Authoring & Delivery RELOAD Project. http://www.reload.ac.uk) 34. Siegel, J., CORBA 3. Fundamentals and Programming, Wiley: 2000.

35. Stojanov, S., I. Ganchev, I. Popchev, M. O’Droma, R. Venkov, “DeLC – distributed eLearning center”, Proc. of the 1st Balkan Conference on Informatics BCI’2003, Pp. 327-336, Thessaloniki, Greece. 21-23 Nov. 2003. 36. Stojanov, S., M. O’Droma, I. Ganchev, “A model for integration of electronic services into a distributed eLearning center”, Proc. of 14th EAEEIE International Conference, A17, 7 pp., Gdansk, Poland. June 2003. ISBN 83-918622-0-8. 37. Stojanovic, L., eLearning based on the Semantic Web, University of Karlsruhe, Germany 38. Stuckenschmidt, H., F. Harmelen, Information Sharing on the Semantic Web, Springer Verlag, 2005. 39. UMTS Forum: http://www.umts-forum.org/servlet/dycon/ztumts/umts/Live/en/umts/Home 40. UWB: http://www.uwbforum.org/ 41. WLAN IEEE 802.11 standards: http://grouper.ieee.org/groups/802/11/ 42. Wu, G., C.-W. Chu, K. Wine, J. Evans, R. Frenkiel. “WINMAC: A Novel Transmission Protocol for Infostations”, IEEE VTC’99, Houston, May 1999. http://www.cs.rutgers.edu/dataman/nimble/papers/winmac.pdf 43. http://www.w3.org/TR/soap12-part1/(to date) 44. http://www.w3.org/TR/soap12-part2/(to date) 45. http://www.cs.umbc.edu/kqml/ (to date) 46. http://www.omg.org/techprocess/meetings/schedule/adopt.html (to date). 47. http://jakarta.apache.org/jetspeed/ (to date) 48. http://www.w3.org/2002/ws/(to date)

15