The InterMed Network: Changing Cultural Patterns with a Large-Scale Cooperative Internet Network in the Mediterranean

Leandro Navarro <>
Associate Professor, Universitat Politècnica de Catalunya (UPC), Spain
Technical Director of Pangea, a local APC network

Gerard Rodríguez <>
UPC doctoral student
Aleph designer

Artur Serra <> (contact)
UPC research scientist, coordinator of Catalan chapter of Internet Society
InterMed director
Tel: +34-3-4017182
Fax: +34-3-4017055


InterMed [1] is a cooperative network among research institutions, companies, administrations, and civic organizations engaged in the establishment of new patterns of cooperation and accelerating the development of the global information society, based on the Internet technology, in the Mediterranean region.

Last 19-20 January, the InterMed network began with an international conference in Barcelona. One hundred participants, from 12 different Mediterranean countries (Morocco, Argelia, Tunisia, Israel, ANP, Greece, Spain, France, Italy, Bosnia-Herzegovina, Slovenia, and Malta), as well as representatives of international organizations such as the EU Commission, cooperation networks such as the local chapters of the Internet Society (ISOC) and local nodes of APC, and the Association for Progressive Organization, gathered to debate a framework of cooperation in the region. The result of the meeting was to establish a network with several common services: mailing lists, a Network Information Center for the region, and a task force on the InterMed future, to study how to develop the network as an association.

The rest of this paper is structured as follows. The second section deals briefly with the state of the art of the networked communities. The third one focuses on the InterMed network developing its projects. Finally, the fourth and fifth sections are related to the Aleph environment, a distributed system supporting the InterMed network.

Networked communities: State of the art

The InterMed network is a prototype that develops some ideas generated by the GAIA UPC research group. Those ideas are taken from different fields:

Between 1992 and 1995, our team has been an active partner in the EU ESPRIT COMIC project on CSCW (1992-1995) [3]. In the framework of this project, the UPC team developed a distributed architecture called Aleph [4], that we are negotiating as the InterMed architecture. At the same time, we have conducted some analysis of large-scale networked communities [5]. Finally, we included some lessons from other studies we did, from an ethnographic point of view, at the U.S. about this new kind of organization.[6]

Several services are now in operation as a result of these ideas:

A deeper understanding of the so-called "networked communities," and their implications for the new international model of cooperation is necessary. The development of the Information Society (IS) is closely related to the evolution of this new research field. After the establishment of TCP/IP protocol as the worldwide, de facto standard in the interoperability of digital networks and the success of the World Wide Web (WWW) technology in the field of information services, we consider the design of new technology for cooperative systems and networked communities to be a promising next step.

The "virtual community" field is a heterogeneous and complex area.

We can consider at least two big, different groups. The traditional organizations that decide to establish an IP computer network, and a new type of organization born directly from the Internet and other computer networks. This is not the same as a university that establishes a distributed system, as in the case of the Andrew System at Carnegie Mellon University (CMU) [6], nor an international electronic network, that organizes a cooperative activity, as in the case of I*EARN, the International Education and Resource Network [5], or the APC. In the first case, we have a "networked community" [11,12], and in the second case, we have an "organizational network" [5]. The first group represents the bulk of Internet networked communities, the second is the avant-garde.

The dynamic between these two big types of organizations differs too. The development of the "computer-intensive campus" in the U.S. in the 1980s (Andrew-CMU, Athena-MIT), was prepared by the existence of an ARPA computer research community connected through Internet. On the other side, the extension of Internet network communities in the Third World has been preceded by the APC organizational network. These networks seem to be the pioneers in the introduction of the innovations in a country or community. Cross-sectorial and international cooperation activities are the natural ground field of these networks.

Third, we can see this phenomena from different approaches. The literature is divided between a radical and artistic vision of the "cyberspace" and "virtual communities" as communities totally embedded in the virtual space, as in the Neuromancer novel, and a more conservative, scientific vision of the so-called "networked communities," describing the changes that traditional organizations (universities, companies, public administration, NGOs) are experiencing when they discover electronic communications, or more recently the Internet and the WWW universe [11,12]. We consider a third approach necessary, an engineering approach, based on an interdisciplinary design methodology, for the understanding and the building of these communities.

The InterMed network

The InterMed network is an organizational network, or better, a federation of them.

Its goals will be to promote new models of cooperation to facilitate the development of the Global Information Society in the Mediterranean, based on Internet technology and applications. It supports a clear orientation through international cooperation and stresses the leadership of the civic society in this process, with respect for cultural diversity.

The InterMed has four key components:

The projects we are working on include:

MIP (Mediterranean Inventory Project)

The goal of MIP is to build a distributed inventory of the Mediterranean projects related to the development of the Information Society in the region, country by country. This inventory will be part of the G7 project called GIP (Global Inventory Project). More information is available at

This project implies that all of the InterMed members have to build a Web page with all the national or regional projects related to the IS (Argelian Inventory Project, Israel Inventory Project, etc.) There is a good example in the U.K. at the same server. We are currenlty building the Catalan and Spanish part of the MIP-GIP.

The MIP is the first step toward the building of a InterMed Network Information Center (NIC). This service will keep updated information on the Internet resources of the Mediterranean region, including a directory of people interested in InterMed purposes.

Med Agrinet [13]

Dov Winer of the Israel Chapter of ISOC is starting a Mediterranean Agricultural Extension Services Network project to share common knowledge about agriculture matters in this region. We would like to stress the importance of this project. First, the Mediterranean, in particular its Southern rim, is now undergoing a demographic transition. That will result in a large, young population segment over the next decades, with a great potential for learning. On the other side, the Mediterranean societies have to be able to feed and develop this potential.

Second, in the developed countries, there is increasing talk about the benefits of the Mediterranean diet. New cultural habits based a more healthy diet, with less fat and cholesterol, etc., can benefit the development of the products of the Mediterranean agriculture and its rich cuisine.

MEDnet (Mediterranean Education and Research Network)

Oliver Popov, the coordinator of MARnet (Macedonian Academic Network), has opened our eyes to the possibility of building a Mediterranean association of academic and educational networks. He has a leading role in the development of the CEENet, the central and eastern European association of research networks. CEENet was founded in 1993, based on an initiative from ACONet (the Academic and Research Network of Austria). Today CEENet has 19 member countries and two more have recently applied for membership.

Salem Zougbi, director of PLANET, the Palestinian research network, and Dov Winer, the director of MACAM 98 [14], the National Teachers' Colleges Network of Israel, and others in the InterMed in charge of academic and educational networks, are studying the possibility of building a Mediterranean Educational and Research Network association (MEDNet), following the CEENet and the RARE experiences. That could be of great help in building academic relationships on research projects, particularly in the field of Internet technology. At the same time, it could be useful to share educational projects between universities and secondary schools.


We started the EPITELIO project at the UPC last month [15]. It is a European telematics project promoting teleworking against social exclusion (unemployment, etc.). It will be a two-year project, ending in 1998. More information is available at Individuals from Madrid, Milano (ENAIP), Manchester (MMU), TERN, and Barcelona are involved.

We are establishing what we call "teleworking centers" in the UPC, in Barcelona neighborhoods, testing the use of telematics to create self-employment at different levels.

We think the idea of teleworking can be very promising in the Mediterranean. InterMed is exploring the possibility of extending the EPITELIO project in the Med. We particularly want to do some pilot projects of teleworking with young people from the Southern rim of the Med (Argelia, Morocco), as well as from Bosnia or other Eastern Med countries.


We consider EUMEDNET [16] a brother network of InterMed, specialized in SME and electronic commerce in the Mediterranean. The director is Juan Carlos Coll, professor in the Economic Department at the Universidad de Malaga. (

EUMEDNET tries, from an academic and business point of view, to promote the commercial use of the Internet for trade in Europe and the Mediterranean area, focusing on SMEs. EUMEDNET is closely related to some InterMed areas; mainly (but not only) telematics for SMEs, teleworking, and telematics for research.


The CERT (Computer Emergency Response Team) is an institution in the Internet community dedicated to the security of projects. In Barcelona, we have the Spanish CERT, which is a member of the European network of CERTs. This center is interested in establishing a Mediterranean network of CERTs, sponsored by the InterMed association. Manel Medina ( and J. Buch ( are in charge of this project .

The basic idea of all these projects is that developing the Internet as an open network in the region goes together with the designing and building of open social networks of cooperation between North, South, West, and East. Open systems need open societies to develop themselves. This is InterMed: an open system-open society network for project cooperation in the region.

The Aleph environment and InterMed

A distributed architecture

From the computing, as well as from the social point of view, InterMed is based in the concept of federation [17]. We conceive this new system as a federation of networked associations working on IP networks, sharing Internet basic protocols, as well as general WWW technology and specific services designed for a region which combines enormous economic, social, and cultural differences.

We are currently developing InterMed as a network based in the Aleph environment. This new environment is decentralized, allowing local control of the information, but open at the same time to the global Internet. Each country participant can index the information about their projects, Web pages, users, etc. The idea is not only surfing on the Net, but participating actively in the creation of new information and contents on the Net.

The Aleph environment in the InterMed Network

The InterMed network is formed by a set of sources of information supported by Internet servers. The purpose of these sources of information is to support and encourage the development of Internet and the Information Society in the Mediterranean area through the InterMed projects. We think that the key to achieving our purposes is to provide an easy and efficient access to the information related to the Information Society and its development. The kind of information we think might be on those servers is:

All this information must be easy to access for each region or country if we are to efficiently achieve our purposes. It means that this information must be written in the language of each region, not only in English; that information should be written according to the social context of each region as well. If we take these measures into account, we will achieve a deeper penetration of our purposes and a higher dissemination of the information society and its terms in each region, and in the whole Mediterranean region.

To fulfill our objectives, we propose to place the InterMed information servers in each region which will be responsible for gathering information concerning its own region, building a set of databases and Internet-based services adapted to the social context of each area. Figure 1 shows a first approach of how the InterMed architecture might be structured.

Figure 1: The distributed structure of the InterMed network based in cooperative local servers.

The InterMed architecture is divided into sites or local servers, each will be placed in a geographical region or specialize in gathering information about a specific topic, such as a database of NGOs (Pangea NGO database [7]). The architecture of all these sites will be based on similar tools, but each one will be adapted to the requirements of its own region, such as low-speed connections or multilingual services. The main services of each site are:

  1. A Web server that will keep information of the assigned topic of each site, either a region or topic, such as the MED Agrinet project [13]. The Web pages should be written in the local language and in English to be easily accessible by foreign visitors, and to enable interregional cooperation. Furthermore, the Web server will play the role of the main entrance to the rest of planned services, since the Web technology is worldwide and well-known.
  2. Discussion forums, mailing lists, and newsgroups to encourage direct communication among people in the region.
  3. File servers (File Transfer Protocol [FTP]) to share important documents.
  4. Longhaul accessible databases by means of Web interfaces, with e-mail interfaces for low-speed connections.

Apart from the services offered by each InterMed site, such as newsgroups or Web servers, the proposed InterMed architecture has mechanisms to share the information generated by each site with the rest of members. All the sites cooperate with the rest of the sites, offering information, and, all sites will be able to get information from the rest of the sites. Therefore, although the information is generated and stored locally, the users accessing the InterMed network at any site will see all the information gathered by all the sites in a unique database.

By applying this method of "think globally, act locally," we have access to global information, but we will have the advantage of information being collected locally, taking into account the social context of each region. As a result of this policy, the information contained in each site will have higher semantic quality and will be more realistic. In other words, what we propose is to built a distributed architecture, providing tools to allow for cooperation among sites.

The proposed mechanisms to implement our distributed cooperative policy are:

The InterMed network would have a main site which would keep a copy of all the information. This site would act as a backup and a central reference to all the distributed sites. This site is already working in the Universitat Politècnica de Catalunya (

However, the proposed InterMed architecture takes a step forward. The previous proposed architecture has an important lack: How to find the information users need inside the group of InterMed sites? The task of finding out the information that users require is more difficult in the kind of distributed architecture we have proposed for the InterMed network, because there is no central point where the information is stored. To be exact, there is a central point which is the backup site, but this site will not necessarily have the latest data, so the problem is still there To solve this problem the InterMed architecture adds the Aleph Package which is explained in more detail in the next section.

The most relevant feature of the Aleph Package is that it gives the control of how the information is indexed and searched to the owners of the information. It is achieved by installing an Aleph Package at each InterMed site.

Figure 2: The final architecture of the InterMed network based on the federation mechanisms of the Aleph prototype

The Aleph Package

The Aleph environment has been developed in the UPC as a result of three years' study in the field of CSCW in the COMIC Project [17]. The Aleph Package is a set of tools to find and share information in a large-scale, heterogeneous environment. This large-scale environment is composed of InterMed sites that share information by means of the Aleph federation mechanisms [4][18]. The set of tools we have developed are clustered under the generic name of the Aleph Package. They are the AlephWeb, a distributed search engine; AlephCal, an electronic and distributed agenda; and AlephMarks, a tool to share bookmarks on the Web among a trusted small group of cooperative people, such as those in the InterMed project.

These three tools have been designed to use the federation mechanisms of the Aleph computational model[19] to allow interaction among servers. Each server, AlephWeb, AlephCal, or AlephMarks, has a federation manager attached. This federation manager deals with the interactions with the rest of the Aleph-based applications, no matter which kind of server, following the federation mechanisms. By doing so, we have a set of heterogeneous tools which are able to cooperate, providing a generic information-provider service to the whole community of users, no matter where the user is placed or his contextual knowledge.

Figure 3: This figure shows the components of the Aleph Computational Model: the Aleph based applications that use the Aleph federation tools, the federation managers, the set of interceptors, and the databases with the contextual information of each entity.

To summarize, the main features of the Aleph Package are as follows:

  1. It is based on the Web as an underlying infrastructure, using the CGI to build a parallel protocol to support the federation mechanisms. The Web was chosen as an underlying infrastructure to achieve a real, large-scale service.
  2. AlephWeb [20] provides tools to build a distributed search engine where each server builds catalogues of selected sources of information stored in each entity. The AlephWeb architecture is organized by contexts or environments. Each environment represents an InterMed site. The information of each environment is stored in its own database, and the elements of the AlephWeb architecture makes it available to the whole community. The AlephWeb architecture in each environment has three main components: the AlephWeb servers, the federation managers, and the databases. These elements can be grouped according to their functionality in search-oriented objects and federation-oriented objects. The AlephWeb servers and the databases form the search oriented objects. These objects keep track of all the links of each information provider in their environment, and perform the search in their local database. Meanwhile, the federation oriented objects, which are the federation managers, interconnect several AlephWeb servers from different environments.
  3. AlephCal provides mechanisms to share and announce important events inside a cooperative community, offering tools to perform searches by keywords. AlephCal is a distributed calendar of events such as conferences, presentations, expositions, and courses. This application is distributed by nature. We are not interested in any kind of event, organized by any entity or located anywhere in the world. Before we ask for information, we have a general idea of the location, the topic, and the organizers of the events we are interested in. In addition, this information comes from many different sources (e.g. our university has an agenda of events organized around the university, while the city council has events related to the city, organized by them or affiliated entities). Someone asking for information may want information concerning what's going on in Barcelona (this includes the university), or what's going on at the university (this may exclude the local council). Therefore, one request may be addressed to a particular AlephCal server or to a cluster of AlephCal servers. This means that the AlephCal service must be provided by a federation of many AlephCal servers, one for each source of events, in a similar way as AlephWeb works.
  4. The AlephMarks prototype is a tool to share bookmarks on the Web inside a group of people that cooperate in carrying out a task. It is based on the idea of that people working on the same subject have similar informational requirements. So, they look for similar sources of information. Therefore, it would be interesting to have a tool that allows for the sharing of bookmarks collected by all the members of a cooperative group in a way that each member would benefit from each others searches.
  5. Each of the previous Aleph-based applications can cooperate in solving a user's request inside a cooperative group of people. For instance, a query to an AlephMarks or an AlephCal server can be expanded to the AlephWeb server to find out relevant information in other sites.
  6. Finally, each Aleph Package in an entity can cooperate with other Aleph Packages in other entities, sharing information to build a real, large-scale environment, using the Aleph federation mechanisms. For instance, several AlephWeb servers placed in different entities might cooperate building a worldwide search engine service formed by the conjunction of small collections of indexes built in each entity.


[1]. What is InterMed (

[2]. InterMed 96. The Conference (

[3] The COMIC Project "Computer-Based Mechanisms for Interaction in Cooperative Work". ESPRIT Basic Research Project 6225.

[4] Aleph: A large scale CSCW environment. Leandro Navarro, Gerard Rodríguez. JENC6 The 6th joint European networking conference 1995.

[5] 1995. Serra, Artur, "The I*EARN, organizational learning network." COMIC Deliverable 1.3. ESPRIT Basic Research Project 6225.

[6] 1992. Serra, A. 1992, "CMU as design university". 90 Congress of the American Anthropological Association. (

[7] Comunicació per a la Cooperació. Pangea (an APC organization). (

[8] Isoc-Cat, the Catalan Chapter of ISOC (

[9] VallesNet (

[10] BCNet (

[11] 1987 Sproull, L., Kiesler, S., Computing and change on campus. Cambridge: Cambridge University Press.

[12] 1991. Sproull, L., Kiesler, S. Connections. New Ways of Working in the Networked Organization. Cambridge, Massachusetts: MIT Press.

[13] The proposal of the Agrinet project "An initiative for an Agriculture Extension Services Network in the Mediterranean Area" (

[14] MACAM 98 (

[15] EPITELIO. Telematics against social exclusion (

[16] EUMEDNET project (

[17] COMIC, 1995 (

[18] 1993. From Small to Large Scale. J. M. Marquès, L. Navarro, M. Sarmiento. COMIC Project.

[19] AlephWeb: a search engine based on the federated structure. Leandro Navarro, Gerard Rodríguez. In Proceedings of the 17th Joint European Networking Conference. Budapest, May 1996.

[20] AlephWeb: a large scale CSCW Trader. Leandro Navarro, Gerard Rodríguez. In Proceedings of the first Workshop on CSCW and the Web. February 1996. Sankt Augustin, Germany.