The deployment of an academic middleware to support advanced learning services
The rapid development of the significance of digital and human networks in academic institutions, the development of vocational and distance learning methods and systems and their associated services delivery have modified the organizational and informational structures of academic institutions. In this new scenario, management information systems, various learning resources in printed or digital format and network related user services are usually competing for acceptance in a rapidly changing system. There is an emerging need to provide new services in order to function in a changing international market, to prepare for the expected student mobility and the increasing diversity of educational needs. There are two main research areas that are looking at the problem each from their own perspective: the learning technology area and the network technology area. The first is working mainly on standards and best practices for the delivery of on-line and distance education. Efforts include the description of learning objects with metadata, the description of learners with profiles and the generic depiction of learning technology systems in terms of information systems architecture. Recently deployment issues are increasingly of concern to this research community. Inevitably the research is moving towards an area of work where already major effort is under way. This is the part of the networking community that is working on the definition and leverage of middleware services. This paper argues for the introduction of academic middleware to speed up the delivery process of the new services and reduce the complexity of the managed systems.
Keywords: Middleware, Learning Technology, User profiles, Learning Objects, Directory services.
Technology and economy are the two main driving forces behind the acceleration in the transformation of present day on-line education delivery. Technology is enables the delivery of new types of educational services. Computer networks are the main tool for the introduction of these new services. The most basic effect of networks, is that education is rapidly transformed from location based to distance online based . This affects both the roles the participants play, and the organisations involved. New entities enter the education arena such as information technology experts and commercial producers of educational software. Economically, organisations are investing on human intellectual capital and training organisations are discovering the “corporate learning industry” . The need for life-long vocational training, the growing trend for student exchange, the on growing demand for interdisciplinary studies, are all parts of the move away from traditional education based on either the acquisition of a degree or the student being tied to a specific institution. The promotion of autonomy has become accepted as a goal in the education of children and adolescents  and the network technology with its benefits and disadvantages  promotes this tendency.
Organisations are rethinking their modus operandi based on these facts. Popular courses are offered under franchise to others , a growing number of institutions are participating to student exchange programs [10,27], and the introduction of distance learning programs is a common policy both for academic institutions and other learning organizations. A number of strategies are being called into play within this online environment, where geographical constraints are lifted, and where increasingly, the emphasis is upon individualised courses of learning rather than standardised offerings from institutions. This is particularly true in training organisations . For instance, programmes of study are being offered which are modular in construction with increasing flexibility as to how students combine modules within a discipline, or select modules across disciplines to gain a combined qualification e.g. a vocational/academic qualification.
Clearly the most important new element that has arisen is inter-institutional communication. Student exchange, learning modules franchise and inter-institutional degrees necessitate systems that will facilitate the communication . Apart from the alleged need for integrated information systems to support the institutional organisational and educational components (user network services, learning technology systems, institutional management information systems), there is an augmenting need for any such system to be scalable and adaptable to a complex globally expanding environment .
There are two main research areas that are looking at the problem each from their own perspective: the learning technology area and the network technology area. The first is working basically on standards and best practices for the delivery of on-line and distance education. Efforts include the description of learning objects with metadata [3,19,16], the description of learners with profiles  and the generic description of learning technology systems in terms of information systems architecture . Recently deployment issues are increasingly of concern to this research community [14,24,25]. Inevitably the research is moving towards an area of work where already major effort is under way. This is the part of the networking community that is working on the definition and leverage of middleware services.
The middleware term was initially used in a software engineering context , but currently its major use can be considered to relate to the network services deployment area tying together networks and services. Moreover the setup of middleware communities [http://www.internet2.edu/middleware, http://www.terena.nl/middleware] dealing with deployment issues of such services indicates that middleware services are more complex than protocol definitions. While still no consensus can be reached on the services, which must be considered “middleware”, there are some that are the common denominator. Such services are common place in present day organizational information systems but more work is required on the definition of a common inter-organizational framework. While both these areas of research are moving quickly to standardize and implement their work, there seems to exist a gap in the establishment of a surrounding environment for the initiation and acceptance of these standardization efforts in the working place, in this case educational institutions and those dealing with education delivery.
This paper argues that need for the specification of middleware to support the new functions of the educational community. The next section presents the problem area and summarises the requirements for the emerging educational services of the future. Section 3 gives an overview of the technology advancements regarding the areas of the learning technology and networking middleware services. Section 4 sketches a future implementation based on the results of previous works on the areas. Finally, conclusions and future directions are given in section 5.
The educational sector is directly connected with the developments on the societal environment. The provision of education cannot be considered outside this context. Summarizing the essential changes in the educational sector the following facts are beyond doubt: the economical landscape is rapidly changing. Apart from the privatisation of a large portion of the educational sector, there is a growing need for life-long learning. The “new economy” demands a highly educated working force with up to date knowledge . This fact is changing the clientele –and consequently the programme - of educational institutions. The network reality is offering new tools for the delivery of life-long learning programs with substantially reduced costs . The provision of an integrated service for both the traditional and distance learners with equally attractive programs of studies and the possibility to swap from location based to distance on-line based or modularise (personalise) the programme of study, is been requested from increasingly competing educational institutions. Finally the instructional materials are steadily more reliant on the use of learning technology. Computer labs, networks, digital learning environments and systems are all considered indispensable tools for the delivery of knowledge.
The main application areas of the near future are believed to be:
· The introduction of distance learning programs as part of the regular studies in an institution.
· The need for further modularisation of the study programme in order to offer customised degrees and enhance student autonomy.
· The development of inter-institutional programs
The main technological infrastructures that exist in a typical institution presently are:
· One or more custom management information systems that contain student records, financial data etc.
· One or more distance learning programs
· A campus network and associated user services
Recent researches indicate that of the competing learning institutions in this area, the successful ones will be those that manage to offer new services, while satisfying quality criteria applicable to a number of processes and entities like course development, faculty training, student services, learning resources, infrastructure, and outcomes assessment. A set of benchmarks  to evaluate the quality of the distance-learning programs in the US were better satisfied by an integrated information system. For example the benchmarks on “Student Support” and “Faculty Support” presupposed that the institution as an organisation – with its infrastructure - is directly related with the delivery of the courses (e.g. “Electronic security measures are in place to ensure the integrity and validity of information” or “Support for building and maintaining the distance education infrastructure is addressed by a centralized system”)
In this situation where the complexity of the system steadily increases and where the network based, global, inter-institutional services are continuously gaining importance the introduction of an institutional middleware would:
In the learning technology area there are efforts oriented towards the promotion of interoperability between the different learning systems and the management of pedagogic elements. Therefore standardisation is directed to those components that can be considered reusable and therefore possessing commercial value (like the learning objects) or those components that are essential for the provision of the new services (like the profiles of users / learners). While there is mobility in other areas too, like the generic learning technology architecture, the student identifiers, and the learning environments, the above mentioned information (learner profiles and learning material) is considered central to the educational procedure .
The pedagogic elements (or learning materials) are the digital informational content to be provided to the students. In this field, there are some very advanced efforts on the design of learning materials metadata. The most widespread of these are the IEEE’s Learning Resource Metadata , Ariadne  and IMS [18,19]. These efforts have led the technology in the area and have produced metadata object schemas. The design of these specifications was based on the possible uses of the learning resources as part of learning systems. In this way noteworthy categories of the LOM schema are the Educational category that “groups the educational and pedagogic features of the resource”, and the Rights category that “groups the features that deal with the conditions of use for the resource” . On the other hand the learner profiles in IEEE are described by PAPI . The PAPI specification divides learner information into four types: personal information, such as names, phone numbers, addresses, etc, preference information, such as whether the learner prefers video over just plain text, performance information: referring to the learners’ achievements in the context of learning environments (digitised or not) and finally, portfolio information, referring to the record of student achievements in terms of skills acquired, and abilities.
On the networking side, the main research or areas include the provision of an intermediate service between the network and the application level, basically for the deployment assistance of distributed applications. In this context the main effort has been given to the development of infrastructure on directories and PKI (Public Key Infrastructure) in order to support functionalities (Authentication, Authorization, Accounting) and application on QoS (Quality of Service), Active networks, Distributed data management, information sharing, or GRID . While much emphasis has been given in this research to the support of academic research, little effort has been paid to the provision of education especially in the new context earlier described. However the development of a wide scale –global- directory service as a crucial component for the provision of any user related distributed service affects the provision of education also. This is especially true now, that technology is unifying operating systems, mail servers and web server user records, and the trend is to expand these directories in order to store profiles and define access rights for as many applications as possible running in a digitised system [5,26]. The speed, with which technology is moving, points to a situation where student records would be some kind of operating system user records . This appears the most likely since the directory services protocol and corresponding implementations of it are moving in this direction, not just for educational organizations but for any kind of organisation, and especially those that are multi-national and globally distributed. It is now accepted that a well-established directory service offers a stable and tested environment for the purposes of storing and manipulating, among other things, education provision related records.
Middleware is not a simple collection of protocols. The notion of a middleware contains also the provision of a service. Therefore middleware is not considered as a set of interfaces but system components with functionality. Thanks to the middleware technologies, these services have the ability to scale globally . A definition states  “Middleware can be viewed as a reusable, expandable set of services and functions that are commonly needed by many applications to function well in a networked environment". A typical application can be considered the provision of educational services over or using the network.
Research is ongoing for the integration of learning environments with organizational middleware services. In this context the GESTALT project produced a demonstrator of a resource discovery service (RDS) with retrieval mechanisms on metadata  combined with portable student records using a Directory Services model .
The objective of GESTALT was to design and implement an open online learning environment by bringing together and enhancing the results of previous projects and creating an integrated trial system, which encompasses the whole of the delivery of online learning from searching for learning courses and resources, enrolling on and following programs of study delivered over networks, and providing mechanisms for passing information between various parts of the system in a secure and trusted manner.
Figure 1 The overall GESTALT Functional Architecture
Figure1 summarizes the GESTALT functional architecture. From one point of view, central to such a system are considered the user profiles and the asset management system. These are the two main stores for the learner and educational content data. User Profiles hold user preferences for interaction (e.g. preferred language) with the Learning Environment (LE), with the Resource Discovery Services (RDS) and other information such as personal portfolio and performance data. The Asset Management System controls access to added value resources, for example course/module data, which could be accessed at an extra cost program of study. This subsystem also publicizes learning objects (i.e. course module data) to the Resource Discovery Services.
Although not shown in the figure, other components like the RDS, the LE and the administration component have or can have their own directories holding information specific for their functions but not of direct interest to the other components. GESTALT indicated that in order for the various learning environments and brokerage services (like the RDS) to be capable to interwork - and to cooperate on the data exchange -which is crucial in order to add value to their services - the definition of common descriptions of the exchanged data is required. Moreover such descriptions must be integrated in the information systems of the related parties (e.g. academic institutions) and should be incorporated in a wider context managing related chunks of information. This work is now been undertaken by a new project GUARDIANS (Gateway for User Access to Remote Distributed Information and Network Services).
The problematic areas of the researches presented in the previous section can be summarised in the following:
An answer to go some way to resolve the problem and give boost to both areas is the development of an academic middleware that will be based on existing middleware efforts of the “networkers” and has as objective the learning technology. The vital questions are: What should such a middleware contain? And how should the middleware be constructed?
As Aiken et al.  noted, “the better approach would be to consider middleware as an unstructured, often orthogonal, collection of components (such as resources and services) that could be utilized either individually or in various subsets”. Stretching this hypothesis, middleware is considered to be a set of components placed horizontally in the academic information infrastructure. Middleware is not a new product but rather a new service. The academic middleware, like all others, should contain the most common data, components or functions of the systems. This way this new service can become central to the intra-institutional information system. As was previously analysed the main information to be manipulated by such a component must be the learner (student, client) information and the module and learning content information from which the learning programs or qualifications or degrees are constructed (figure 2).
Figure 2 Systems and services in connected with middleware
The problem of how the middleware should be constructed is directly related with the direction of the deployment of network middleware services. It is highly probable that directory services will play a central role. It is also required that such middleware will inherit implemented best practices by other middleware initiatives like the student identifiers, AAA (Authentication, Authorization, Accounting) and QoS schemes. The degree of interdependence between the two areas of research will depend by the applications that will exploit this new framework.
The possible scenarios of use of the academic middleware are perhaps the best indication of the level at which academic middleware will assist the deployment of new academic services:
· Provision of common middleware services for the educational procedure. For example, distance based programs connected with learning technology environments with QoS and AAA directly applied.
Figure 3 Users and the Inter-institutional Middleware
As these new services show the development of a middleware will affect both the communication of components in the intra-institutional and inter-institutional environment. This dual nature of the middleware services require their deployment to take into account the international standards and practice.
The ideas discussed in this paper concern the implementation of middleware infrastructure with learning technology. Middleware as a notion is tied with the network technology. It is expected that the significance of middleware services will increase correspondingly with the development of distributed, network based services. Widely implemented middleware will greatly reduce the complexity of developing worldwide services.
Education, and especially academic education is an application area where new technologies are usually originally introduced. Future development of services based on networks, using these either on the inter-institution relations or the institution to learners relations is a need deriving from the economical and social environment. While the deployment of these services will require a wider consensus between the related parts, the definition of a technical context in the conceptual level is the idea of this paper.
Developing a middleware enclosing the learners and the learning materials, may be the first step in an effort to release the learners from the restricting frontiers of single institutions and for reusability of the learning materials on a global scale. Of course the significance of such a middleware, like all network related technologies will increase in parallel with its adoption by the related parties. However, efforts such as IMS in US and Prometheus in Europe show wide commitment from both educationalists and educational service suppliers to implement commercially viable, interoperable systems.
This work has been partly supported by the European Union through the IST Programme (ACTS Project GUARDIANS Gateway for User Access to Remote Distributed Information and Network Services) and the GRNET (Greek Research Network) S.A..
 Achievements in Higher Education at East Birmingham College, (September 1997)Retrieved February 2, 2000 from the World Wide Web: http://www.ebham.ac.uk/college_info/achievements_profile/achieve_main.html
 B. Aiken, J. Strassner, B. Carpenter, I. Foster, C. Lynch, J. Mambretti, R. Moore, B. Teitelbaum, (2000). Network Policy and Services: A Report of a Workshop on Middleware. Request for Comments 2768. February 2000.
 ARIADNE (1999). Educational Metadata Recommendation, Version 3.0. © EPFL (Lausanne, Switzerland), KUL (Leuven, Belgium) and the ARIADNE project. Retrieved March 27, 2000 from the World Wide Web: http://ariadne.unil.ch/Metadata/ariadne_metadata_v3final1.htm
 Bank of America Securities, (1999). The e-Bang Theory. Education Industry Overview. Bank of America Securities. Montgomery Division. September 1999.
 BED Web Team (1999). Active Directory Architecture. White Paper. Retrieved 10 March 2000 from the World Wide Web: http://www.microsoft.com/WINDOWS2000/library/howitworks/activedirectory/adarch.asp
 P. Bernstain, (1996). Middleware: A model for Distributed System Services, Communications of the ACM, Vol 39, No. 2, pp. 86-98
 R. Close, R. Humphreys, B. Ruttenbur, (2000). e-Learning & Knowledge Technologies, SunTrust Equitable Securities, March 2000.
 Mc Crea et al., (2000). Riding The Big Wave. A White Paper on the B2B e*Learning Industry. Fred McCrea, R. Keith Gay, Rusty Bacon. January 18, 2000.
 Henk Eertink, Yuri Demchenko, Notes from European Middleware Workshop (EMW2000), Retrieved 10 October 2000 from the World Wide Web: http://www.terena.nl/middleware/emw2000notes01.html
 European Parliament and Council, (20 April 95). Decision no 819/95/ec of the 14 March 1995, Establishing the Community action programme 'Socrates'. Official Journal of the European Communities No L 87/10.
 European Commission, ECTS User Guide, 31.03.1998. Retrieved September 15, 2000 from the World Wide Web: http://europa.eu.int/comm/education/socrates/guide-en.doc
 F. Farance, (1999). Learning Technology Systems Architecture (LTSA) – Specification Draft. Retrieved February 2, 2000 from the World Wide Web: http://edutool.com/ltsa/
 E. Forte, F. Haenni, K. Warkentyne, E. Duval, K. Cardinaels, E. Vervaet, K. Hendrikx, M. Wentland-Forte, and F. Simillion, Semantic and Pedagogic Interoperability Mechanisms in the Ariadne Educational Repository, ACM SIGMOD Record 28 (1999), no. 1, 20-25.
 P. Foster, M. Kraner, A. Graziano, (1999). D0401 Courseware Metadata design. Retrieved 4 April 3, 2000 from the World Wide Web: http://www.fdgroup.co.uk/gestalt/D0401.pdf
 GUARDIANS (2000). GUARDIANS (Gateway for User Access to Remote Distributed Information and Network Services) Technical Annex. Information Societies Technology (IST) Programme
 IEEE LTSC (2000:1). Draft Standard for Learning Object Metadata. P1484.12/D4.0. Retrieved March 27, 2000 from the World Wide Web: http://ltsc.ieee.org/doc/wg12/LOM_WD4.htm
 IMS (1999:1). Learning Resource Meta-data Information Model, Version 1.0. Retrieved February 2, 2000 from the World Wide Web: http://www.imsproject.org/metadata/mdinfo01.html
 IMS (1999:2). Meta-data Best Practice and Implementation Guide, Version 1. Educause.
 M. Konstantopoulos, T. Spyrou, P. Koutsabasis, J.S. Darzentas, C. Lambrinoudakis, J. Darzentas, (2000). The Role of Directory Services in Online Education Delivery: The GESTALT Example. International Journal of Continuing Engineering Education and Life-Long Learning. To Appear.
 M. Konstantopoulos, J.S. Darzentas, T. Spyrou, P. Koutsabasis, C. Lambrinoudakis, J. Darzentas, (2000). Towards Integration of Learning Objects Metadata and Learner Profiles Design: Lessons Learnt from GESTALT. Interactive Learning Environments. Accepted.
 Microsoft (1999). Microsoft Windows Active Directory: An Introduction to the Next Generation Directory Services. White Paper.
 Moore M.G., (1972). Learner Autonomy: The second dimension of independent learning. Convergence. 5, 76-88
 Rehak D. (1997). Carnegie Mellon Online: System Architecture. Retrieved September 27, 2000 from the World Wide Web: http://online.web.cmu.edu/public/information/publications/architecture/index.html
 Smith B., (2000). Higher Education & Enterprise Learning Management Systems. Retrieved September 27, 2000 from the World Wide Web: http://www.convergemag.com/Publications/CNVGNov98/highered/highered.shtm
 SUN™ (2000). The Solaris™ Operating Environment. Data Sheet. Retrieved March 30, 2000 from the World Wide Web: http://www.sun.com/software/solaris/ds/ds-sol8oe/
 U. Teichler, F. Maiworm, (1996). The ERASMUS Experience Major Findings of the ERASMUS Evaluation Research Project. Office for Official Publications of the European Communities, Retrieved February 2, 2000 from the World Wide Web:: http://europa.eu.int/en/comm/dg22/socrates/erasmus/erexpe.doc
 The Institute For Higher Education Policy (2000). Quality On The Line, benchmarks For Success In Internet-Based Distance Education.. April 2000.
 Chih-Hsiung Tu, On-line learning migration: from social learning theory to social presence theory in a CMC environment, Journal of Network and Computer Application (2000) 23, 27-37
 V. Wade, K. Riley, B. Banks, P. Foster, N. Evans-Mudie, Y. Nicol, P. Doherty (1999). Object (Interface) Specification. Work Package 5, The GESTALT project, final draft.
 J. Waldo, (2000). The End of Protocols or, Middleware for the New Millennium. Middleware 2000 Keynotes. Retrieved September 9, 2000 from the World Wide Web: http://www.research.ibm.com/Middleware2000/Keynotes/keynotes.html
Michalis Konstantopoulos is the technical director of the GRNET directory service. He is a PhD candidate in the department of Information and Communication Systems, University of the Aegean, Greece. He has an MSc and a Degree in Information Systems from the Department of Informatics, Athens University of Economics and Business, Greece. He has worked in several national and European projects and participated in the authoring of proposals, technical reports and research papers published in international conferences and journals and he has been technical director of the Greek national directory services program. He is currently lecturing at the Department of Product and Systems Design, University of the Aegean, Syros (PD407/80) and also working as a research associate for the “Aegean-Net” network service and the GUARDIANS project.
Thomas Spyrou is a Lecturer at the Department of Information and Communication Systems, University of the Aegean, Samos, Greece. He has participated to many research projects, funded nationally and by the European Union. He is a member of the Greek National Research Network (GRnet), the Greek Universities Network (GUnet) and Greek National Educational Network (EDUnet) technical committees. He is a member of the Organising Committee of the Dept of Product and Systems Design, University of the Aegean, Syros, Greece. His current research interests include Intelligent Systems, Information Systems, Systems Thinking, Simulation modelling, Information Systems Security, Knowledge Management in Decision Aiding and Support Systems, Human - Computer Interaction Modelling, Intelligent agents and Distance and Life long Learning.
Professor John Darzentas (MSc Sussex, PhD London UK, Docent of the Abo Akademi, Finland) is the Acting Head of the Department of Product and Systems Design, University of the Aegean, Syros, Greece and Visiting Professor, University of Athens. He has collaborated in and led many research projects, both in the UK and Greece as well as projects funded by the European Community, on a range of subjects, including Systems Thinking; Decision Support; Simulation; Knowledge Management; Human Computer Interaction; Design; and Learning Technologies. He is author of a substantial number of papers in scientific journals and books.