CREATIVE AND INNOVATIVE NETWORK MANAGEMENT
NATO Science Series A series presenting the results of scientific meetings supported under the NATO Science Programme. The series is published by IOS Press and Kluwer Academic Publishers in conjunction with the NATO Scientific Affairs Division. Sub-Series I. n. HI. IV. V.
Life and Behavioural Sciences Mathematics, Physics and Chemistry Computer and Systems Sciences Earth and Environmental Sciences Science and Technology Policy
IOS Press Kluwer Academic Publishers IOS Press Kluwer Academic Publishers IOS Press
The NATO Science Series continues the series of books published formerly as the NATO ASI Series. The NATO Science Programme offers support for collaboration in civil science between scientists of countries of the Euro-Atlantic Partnership Council. The types of scientific meeting generally supported are "Advanced Study Institutes" and "Advanced Research Workshops", although other types of meeting are supported from time to time. The NATO Science Series collects together the results of these meetings. The meetings are co-organized by scientists from NATO countries and scientists from NATO's Partner countries - countries of the CIS and Central and Eastern Europe. Advanced Study Institutes are high-level tutorial courses offering in-depth study of latest advances in a field. Advanced Research Workshops are expert meetings aimed at critical assessment of a field, and identification of directions for future action. As a consequence of the restructuring of the NATO Science Programme in 1999, the NATO Science Series has been re-organized and there are currently five sub-series as noted above. Please consult the following web sites for information on previous volumes published in the series, as well as details of earlier sub-series: http://www.nato.int/science http://www.wkap.nl http://www.iospress.nl http://www.wtv-books.de/nato_pco.htm
Series ffl: Computer and Systems Sciences - Vol. 189
ISSN 1387-6694
Creative and Innovative Network Management Edited by
Oliver B. Popov Mid Sweden University, Sundsvall, Sweden and Saints Cyril and Methodius University, Skopje, Macedonia
/OS
Press Ohrreha
Amsterdam • Berlin • Oxford • Tokyo • Washington, DC Published in cooperation with NATO Scientific Affairs Division
Proceedings of the NATO Advanced Networking Workshop on Creative and Innovative Network Management Ohrid, Macedonia 15-19 June 2000 © 2003, IOS Press All rights reserved. No part of this book may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, without prior written permission from the publisher. ISBN 1 58603 340 9 (IOS Press) ISBN 4 274 90588 8 C3055 (Ohmsha) Library of Congress Control Number: 2003103694
Publisher IOS Press Nieuwe Hemweg 6B 1013 BG Amsterdam Netherlands fax:+3120 620 3419 e-mail:
[email protected]
Distributor in the UK and Ireland IOS Press/Lavis Marketing 73 Lime Walk Headington Oxford OX3 7AD England fax:+44 1865 75 0079
Distributor in the USA and Canada IOS Press, Inc. 5795-G Burke Centre Parkway Burke, VA 22015 USA fax: +1 703 323 3668 e-mail:
[email protected]
Distributor in Germany, Austria and Switzerland IOS Press/LSL.de Gerichtsweg 28 D-04103 Leipzig Germany fax: +49 341 995 4255
Distributor in Japan Ohmsha, Ltd. 3-1 Kanda Nishiki-cho Chiyoda-ku, Tokyo 101-8460 Japan fax: +81 3 3233 2426
LEGAL NOTICE The publisher is not responsible for the use which might be made of the following information. PRINTED IN THE NETHERLANDS
Contents Oliver B. Popov NA TO Advanced Networking Workshop Second Workshop ofCEENet Network Management CINEMA — Creative and Innovative Network Management
vii
Boris Trajkovski Address on the Second Workshop ofCEENet Network Management NA TO Advanced Networking Workshop
xi
Tomislav Djekov Welcome Address
xv
Oliver B. Popov Welcome Address
xvii
Robert Janz Managing IT in an Academic Environment
1
Oliver B. Popov Building a National Research and Education Network
13
Karel Vietsch European and United States Models of an NREN
27
George Sadowsky The Evolution ofNYU-NET in the Context of Regional and National Networking in the U.S.A.
37
Howard Davies Pan-European Networking Structure for Research and Education
51
Jan H. Guettler Telecommunications Regulations in the European Union
59
Fay Howard and Kevin Meynell Benefits ofccTLD Collaboration and Cooperation
71
Zita Wenzel ICANN and the CCTLD: Global Issues and their Relevance to the Countries in CEE and FSU
75
VI
Eugenie Staicut ".RO" - Top Level Domain Registrations
79
Peter Rastl The Human Factor in ACOnet
83
Richard Perlman Bandwidth is a Valuable Resource
87
Jacek Gajewski CEENet Educational Activities
93
Karen Edberg The Nordic Approach to Distance Education and Learning
95
Neki Frasheri Albania — Perspectives of Collaboration
103
Jan Gruntorad NRN and Commercial Activities — CESNET's Experience
105
Piotr S^siedzki Structure and Infrastructure
111
Vadim Navotny IT for Science and Education in Uzbekistan
115
Ramaz Kvatadze Georgian Research and Educational Networking Association (GRENA)
119
Avgust Jauk Access Issues from ARNES Perspective
123
Jasenka Gojsic The Competitive Environment (Human Resources in a NREN)
129
Basil Maglaris, Theodores Karounos and Xanthi Psiakki Greek Research and Technology Network
133
Tofig Babajev The Telecommunication Infrastructure in Azerbaijan
141
List of Participan ts Organization Author Index
147 155 159
Vll
Oliver B. Popov NATO Advanced Networking Workshop: Second CEENet Workshop on Network Managment
C I N E M A Creative and Innovative Network Management Research and education are essential and vital factors for the development of any society, particularly the one that is based on the free flow of information and knowledge (frequently termed as the information society). The prosperity of the citizens in this type of society is a inter alia a function of education, scientific research, and technological development, which include a variety of skills, problem solving methods and techniques that give rise to an appropriate performance. The evolving needs for competence of the general population demand primarily flexibility and adaptability of the education system. The authorities in every country should promote networking at all levels of the education process in order to create a suitable environment for a life-long learning. The mechanism of the Workshop based on the active involvement of all the participants, through lectures, presentations, and tour de table discussions proved to be highly motivating, stimulative and productive. There were seventy participants from thirty-two countries. The Workshop patron, H.E. Mr. Boris Trajkovski, President of the Republic of Macedonia, used the opportunity to deliver a major address to the nation with
viii
CINEMA - Creative and Innovative Network Management
respect to the importance of the information technology. He underlined the inevitability of the Internet and networking for every developing country. In his speech, the President emphasized that this type of events are one of the catalytic factors that steer each country into the direction of implementation the powerful and pervasive technology for the benefit of its citizens. In view of this, Mr. Trajkovski announced the formation of the National Committee for the Information Society. Some of the lecturers at the Workshop were asked to serve as honorary members of the Committee when established. The Workshop started with the lectures on the governing the ccTLD on a global scale (ICANN) and in Europe (CENTR), the differences in the approach and the awareness that the ccTLD in most cases should be considered as national asset for each country. A lot of interest generated the session on telecommunication regulative and laws. The principal lecture included a survey of the current situation in the European Union, the accession and pre-accession countries. The discussion stressed the significance of competition and deregulation, the need for a universal access and services in light of media convergence, and the parameters that should influence the privatization process of the telecommunication infrastructure in some of the countries. The fourth session was dedicated to the various models of NRENs that depend on the different cultural aspects, the tradition of funding public institutions and organizations, and the position of the government. The session was followed by the presentation of what is going on Europe with improving the infrastructure for better research and education. The main lecture was supplemented with the case studies of Poland and Azerbaijan. It is clear that it is not sufficient not have a good NREN and the appropriate infrastructure. Regardless of the bandwidth capacity, it is always possible to abuse the right and the levels of access. While it is difficult to predict future developments, due to the enormous changes in the technology, there is a need for unambiguous set of rules that will govern and guide the behavior of the network users. Sessions seven and eight were devoted to the usage of ICT in an education and research environment, mainly at universities, and how to set up the proper networking organizations, both from a technical and an administrative prospective. The presentations were colored with examples from Europe and United States.
Oliver B. Popov
ix
They addressed not only the non-for-profit organizations but also some commercial spin-offs. The last presentation in this session enumerated the work of the EC in providing the right environment for research networking in Europe. In addition to technology, even more so human resources are one of the major forces behind networking. The people who cover and work on managerial, policy and technical issues actually are able to provide a sustainable growth and existence of NRENs. How to attract the right individuals, how to educate them, how to keep them in a presence of keen competition from the commercial entities (including lucrative salaries) were among the few topics covered in ninth session. The Workshop ended with presentations that dealt with the extension of the NREN activities in the society such as distance education, and how to improve the specific knowledge in the development countries via the activities of international organizations such as CEENet, TERENA, NATO, OSI, World Bank and UNESCO. Finally, the last presentation focused on the importance of NREN for the initiation and development of the Information Society, which is a basically a foundation for a democratic and civic society. The Workshop provided a suitable environment for many bilateral and multilateral meetings. Two of them are rather promising and might have a long-term impact on the development of research and education networking in the respective regions. The first meeting was among the representatives of the national academic and research networks of Albania, Bulgaria, Greece, Macedonia and Romania. The purpose of this meeting was to initiate better co-ordination and co-operation among the NRENs in the Balkans, as well as to explore the venues for improving mutual connectivity. The second meeting was devoted to the organization of a possible Policy Workshop in 2001 in Bishkek, Kyrgyzstan. The NATO ANW: Second CEENet Workshop on Network Management received an extensive coverage by the printed and the electronic media in Macedonia. There was a press conference organized by the Ministry of Science of the Republic of Macedonia with the presence of the most important news and organizations in Macedonia. During the whole duration of the workshop, Macedonian Information Agency (MIA) was reporting on the daily events. The major address by the President Trajkovski and the excerpts from the Professor Popov's interview were given in the evening prime time news on the major television networks.
X
CINEMA - Creative and Innovative Network Management
A cknowledgments The Workshop was a mosaic of commitment and tremendous effort by many people. Our sincere gratitude goes to all members of the International Programme Committee, the International Organizing Committee, and the Local Organizing Committee. Warmest thanks to the Office of the President of the Republic of Macedonia, Ministry of Science, Faculty of Natural Sciences and Mathematics and the Faculty of Pedagogy. Genuine appreciation to the NATO Scientific Affairs Division, particularly in the area of Computer Networking, assisted by the pioneering vision of the Computer Networking Panel and the Programme Director, for their work that has had a profound and lasting influence and impact on the quality of research and education in the CEE, Caucuses and CA countries.
XI
H.E. Mr. Boris Trajkovski The President of the Republic of Macedonia
Address on the Second Workshop of CEENet Network Management NATO Advanced Networking Workshop
Ladies and Gentlemen,
It is my pleasure to address you and to welcome you to this important gathering. You have gathered here today in Ohrid, the cradle of the Slavic literacy and culture, to discuss issues important for the prosperity of every nation throughout the world, in this new millenium. The invention of computers and informatics has marked the second half of the twentieth century, and the beginning of the third millenium. The need for exchange of information has led to the interconnection of PCs and this to the creation of the global network INTERNET. In a few decades the idea that was developed in the CERN Institute, which first was used only for the communication among scholars in different states, working on same projects, has now a global character. Today the Internet access is 304 million people on line, and only last year their number increased for 78%. In 6 years there was hundredfold increase, which is a good indicator of its future importance. Today there are about 1 Billion WEB pages, and their number is increasing every day by 3 millions. The IT revolution is entering all segments of society such as economy, state administration, education, science, culture and health.
xii
Address The time we are living in is a time of Information society, where the so-called e--
business, e-trade as well as information technology is becoming major motion power in the economic growth. If the ultimate goal of information society is the production of different kinds of information and their application, than the renown economist Peter Drucker was right when he said that "in all developed economy workers who produce knowledge are the core of the working potential and power". We may say in all freedom that developed societies are transferring from industrial to information technology. Today we discuss about a so-called "new economy", some still call it digital economy. The information society lays on the converging of various media such as computers, telecommunications and audiovisual systems. This results in new types and new markets. It is very clear that a well-informed person has the choice when making decisions, and having a choice means the fulfillment of one of his fundamental democratic rights. Thus, the information which is the basis for information society is the best precondition and guarantee for a sustainable development of every democratic state. The Republic of Macedonia is part of the modern world, and with its overall political, economic and social life it constantly reaffirms its strong determination for integration into European and Euro-Atlantic structures. Part of this commitment which basically reflects my own conviction, is the commitment for infomatization of all segments of the Macedonian society with the purpose of creating all preconditions for the development of the Information society. This process is long and painful, but it is clear that it must begin by the most important system which prepares the citizens for a fruitful and happy life, the education. If only 50 years ago, the eradication of illiteracy was of crucial importance, today the introduction of computer literacy is a condicio sin equa non in building a capable active citizen in every society, and of course in the Republic of Macedonia as well. Macedonia needs a national strategy to enter the Information Society. As President of the Republic of Macedonia I will personally take initiative in this direction. I envisage the creation of a National Committee for Information Society. One of the major goals will be the introduction of a teaching program in all education levels, from elementary to high education. This program will allow future generations to use information technology. Together with the Government and the Assembly of the Republic of Macedonia we will work in adopting a legislation which will create the conditions for the development of
Boris Trajkovski
xiii
a so-called e-business, e-trade and for the implementation of the information technology. This legislation will insure free information flow taking into account the public and private interests, as well as consumer protection. The protection of intellectual rights is especially important in creating the environment for attracting foreign investors in the IT. One of the priorities of this initiatives is the development of the SMEs which will appear on the market not only as IT consumers but also as producers who will fit in the global development trends of the world economy. The deregulation and the liberalization in the telecommunication will reduce the cost of these services and this is quite important. Therefore we will work in this direction. Basically, our intention is to make the IT and Internet part of our everyday life. The businessmen to use IT in their business activities, to use Internet for the development of their business, for the design of their new products and services, and will sell the same by using the e-trade. The ministries and other state bodies must be able to deliver good and fast services to the citizens by using the Information technology. Future generations should not only learn how to use the Internet, but it should become their instrument with which they will get new knowledge. This will allow them to learn from best universities throughout the world and still stay in Macedonia. Internet together with the Information technology must become a powerful instrument in the education of our children. I am committed to this and I am thankful to you for having organized this Conference which is a step forward in this direction. It is a pleasure for me to open this Conference. Thank you.
This page intentionally left blank
XV
Tomislav Djekov Deputy Minister of Science of the Republic of Macedonia
The Welcome Address Respected Chairs, Dear colleagues, Ladies and Gentlemen,
It is a great pleasure and honor for me, as a Deputy minister of science of Republic of Macedonia, to have the opportunity to address this respectable and distinguished meeting. The topics of this workshop • • • • •
Structure and organization of networks for education and research Pan-European networking structure for research and education The EU vs. US models ofNREN Accessibility and Security ofNREN Approaches to distant education and learning
... are of interest not only for the people involved in computer, information and communication technologies, but also for a large number of participants in the research, academic end educational communities all over the world. High speed and high quality networking infrastructures which will provide information services, speed up the research activities and make possible distant learning and electronic conferences, is a must today, not only for the highly developed countries, but also for the countries of the central and east European region.
xvi
The Welcome Address
There is no doubt that the five CEEnet workshops on network technology and the two CEEnet workshops on networking policy and management, held since the establishment of Central and Eastern European Networking Association, had a significant impact on the improvement and promotion of the National Research and Education Networks in the CEE countries. I am sure that this workshop, which gathers participants from more than 35 countries, pioneers, experts and policy makers in the proliferation of computer and information technology in the areas of research and education, will have the same success as the previous ones, with a lot of new ideas and concepts and will mean a step forward toward the filling the gap between the CEE countries and the most developed countries in the area of research and education networks. Fully aware of the importance of the information technologies for the overall development of the Country, the Ministry of Science of Republic of Macedonia has supported and will be supporting activities of this kind. It is necessary to boost the development and modernization of the University network, to make it accessible for any university student. Moreover, we have to provide high quality access to Internet to all secondary schools and public libraries. This will require training of a large number of individuals in the networking technologies and their involvement in workshops and seminars of this kind. At the end, I would like to express my respects to the organizers of this meeting and my regards to all the participants in this workshop. I wish you a pleasant stay in Ohrid, in this wonderful Macedonian city, and success in your presentations.
Thank you for your attention.
XVI1
Oliver B. Popov Programme Director of the Second CEENet Workshop on Network Management
The Welcome Address TIoHuiuyean zocuodune upeiucedamejie, dpazu 2ocmu u ynecnui^u na KoH(pepeHU,ujaiua, dpazu dciMii u zocuoba Dear Mr. President, distinguished guests, dear participants, ladies and gentleman
Welcome to the opening session of the Second CEENet Workshop on Network Management - NATO Advanced Networking Workshop, that takes place in Ohrid, Macedonia. This is yet another CEENet event, fully sponsored by NATO Science Committee and with the assistance of the Ministry of Science of the Republic of Macedonia, Macedonian Telecom, Faculty of Natural Sciences and Mathematics and Faculty of Pedagogy at the University of St. Cyril and Methodius. •
• •
The objectives of the Workshop are: To facilitate a transfer of knowledge in the area of National Research and Education Networking with respect to its structure, organization, administration, and user services; To identify the principles and standards relevant to the syntactical and semantical attributes that define a research and education network; To enumerate different models and their implementation while taking into consideration the specifics of each country, that is political, economic, social and cultural differences;
xviii
•
The Welcome Address
To establish, foster and propagate the notion of human networking, that is the interconnectivity and interoperability between the key people in the CEE and FSU countries and the experts that enables the appropriate transfer of know-how, exchange of best practices and exploring further venues for joint bilateral and multilateral projects. Other objectives are:
• • •
To reaffirm the continuity with the workshop in Yaroslavl, To avoid an encore performance, and To provide a balance view by having a cocktail of European and USA based concepts.
The goal is to built an information producing and distributing facility, possibly on a national level, prima facie for the education and research community, and hence for the society in general, which must be sustainable, yet follow and incorporate all of the contemporary IT parameters. The areas that will be covered during this workshop are regulation, governance, models, global infrastructure, security, education and research, and the plausible extensions. This goal is consistent with the efforts, which are not only catalytic of the international organizations and institutions such as NATO, EC, UNDP, Unesco, World Bank and OSI. In all the countries from CEE and FSU present today, there is a clear determination (although sometimes obscured) for democracy and market economy. And they see the freedom to communicate as one of the vehicles to achieve this determination. Naturally, there is a myriad of objectives and subjective obstacles. One of the most difficult problems is the arrogance from ignorance, and that is what we are also trying to overcome by raising the awareness of government agencies and offices and the general public about the mission of a NREN. Networking for us is the way to stop the Balkan syndrome of fragmented objectives and divided interests. In his recent visit, the EU commissioner Mr. Paten during his speech stated "while you cannot change your geography, you can change your destiny". I do second his opinion, but one should not forget that with networking we could in addition change our geography, at least on a logical if not on a physical level. The presence of the President of Macedonia, Mr. Boris Trajkovski, asserts the unequivocal determination of the country for Euro and North-Atlantic integration. One of the ways to achieve this is through proliferation of CIT on all levels of the society. The major force behind this proliferation in many countries is the NREN. (On a more personal note, today is the fifth birthday ofMARNet)
Oliver B. Popov
xix
It is a privilege to have you all here in the city of St. Clement and St. Naum who are inherently part of our determination for education and search for intellectual enlightenment. We sincerely hope that you enjoy your time in Ohrid as much as we do. Thank you again Mr. President for taking time to be with us today and for your patronage of the whole event.
This page intentionally left blank
Creative and Innovative Network Management O. B. Popov (Ed.) IOS Press, 2003
Robert Janz
Managing IT in an Academic Environment
Introduction In the academic setting Information Technology (IT) is, on the one hand, an asset that is mandatory for the day to day activities in research and education. On the other hand IT is usually accompanied with innovation that has its influence in the processes that it is supporting. These two sides of IT require special thought from a management perspective. The University of Groningen in The Netherlands is taken as an example environment to highlight various management considerations. These considerations will be related to questions that are relevant for organisations that are only just starting to organise their services for the scientific and educational community. A first question to be addressed is the position of the IT organisation within the university. For example should insufficient financing within the organisation be compensated with external (commercial) revenues? And if so, to which extent should this be done? A second question one often encounters is the balance between freedom of choice versus centralisation within the organisation. This question will addressed in the setting of the University of Groningen as well as in the national setting in The Netherlands. Of course, from own experience, it is evident that there are major political and economic differences between the situation in The Netherlands and the countries that are represented in the audience. This presentation therefore does not present a blue print for other organisations, but merely points out possibilities and pit falls in managing a professional and sustainable IT environment for the academic and educational community.
2
Managing IT in an Academic Environment
IT and the University of Groningen The University of Groningen (www.rug.nl) is a classical university with 10 faculties, 19,000 students and 5,500 staff. It is the third largest university in The Netherlands and was founded 1613. The turnover of the university is approximately M€ 350. The faculties and research schools are relatively independent, i.e. at the strategic level of the Board of Directors long term issues of the university are discussed and decided on, but in general the faculties are allowed to operate freely within the boundaries set by these long term objectives. Until recently IT was viewed as a tactical issue and the actual implementation was left to the faculty level. Only the campus backbone and the administrative processes were funded by the Board of Directors. One of the strategic issues that recently (the end of 1999) was decided on was the role of IT in the processes of the university. In short in 2005 the university is to be leading in IT in the fields academic education and high performance computing and visualisation (HPC& V). Further IT is to be an integral part of all the processes of the university, with emphasise on administrative support and information exchange. In order to achieve these goals M€ 22,5 (appr. 1 % of the annual turnover during the planning period) has been allocated for new developments (on top of the IT expenses that are already being spent on IT). Thus IT has become a strategic issue of the organisation and as a result the long term implementation of the strategy was put in the hands of the Computing centre. So one of the first things that a successful IT organisation should do is fully commit the top management to IT. Not only in words but also in deeds. Within the university the Computing Centre is situated directly under the Board of Directors as an independent service oriented organisation: Board of Directors University of Groningen
1
1 Management * General director Technical director
Library
Service Centre lOfte i Computer Services * 18fte
I Faculties
Support Staff 6fte
Network Services * 22fte
User Support* 16 tte
* Management Team In essence the Computing Centre is responsible for all central IT related activities that are necessary for the IT infrastructure as a whole. Examples of these activities are the
Robert Janz
3
support of the central network backbone, the Internet connection and operational management of systems for the administrative processes. For these activities the Computing Centre has a contract with the Board of Directors. This contract describes the obligations and financial compensation (approximately 40% of the turnover of the Computing Centre is financed by the Board of Directors). The contact with the Board of Directors (and other parties) is evolving: more and more a quantitative in stead of qualitative description of the services that are being rendered is required. This issue will be addressed later on (see the chapter on ITIL). The remaining income is generated for services that are rendered at a voluntary basis in competition with internal and external parties. Although there are golden mountains to be excavated on the commercial IT market the Computing Centre hardly provides services outside of the domain of science and education. There are some external activities, but these concentrate on co-operation with regional vocational education, national programs for science and education and foreign projects with partners of the university. The core business of the Computing Centre is support of IT in an academic environment and external commercial activities for other parties will dilute the importance of this core in the total spectrum of activities. The total turnover of the Computing Centre in 1999 was M€ 10,3. How these earnings were generated from the various parties that the Computing does business with is presented in the following pie diagram: 15%
24%
12%
49%
The diagram demonstrates that although the Computing Centre is primarily a central funded part of the university nearly half of the turnover is generated from internal services. The 24% of the revenues that are generated externally stay within the domain of research and educational and in this way the Computing Centre remains loyal to its original mission. Next to the 74 fte. at the Computing Centre most faculties have their own local IT departments that are responsible for the local area networks that are connected to the backbone. The relation with these local IT departments is two sided: on the one hand there is a functional relationship concerning issues like standardisation, backbone connectivity and such and on the other hand they are sometimes true competitors in acquiring contracts for IT support within of outside of the university.
4
Managing IT in an Academic Environment
The management has always guarded the mission of the Computing Centre: it a part of the academic community and is specialised in services for this sector. The result of this strategy is that the Computing Centre is firmly anchored within the university. Other academic Computing centres who have had a more commercial approach and have broaden their range of customers have either been outsourced or having difficulties in proving their loyalty to the academic and research community.
IT infrastructure One of the consequences of the strategic ambitions of the university was that IT infrastructure needed a major redesign. Although there were no serious bottlenecks on the network yet, the architecture of the infrastructure that originated from plans laid down in the late 80's would not be able to cope with the demands that were going to be made in the coming years. Not only bandwidth would become a problem but also demands concerning the availability of the network (7 * 24 hrs uptime) could not be guaranteed. Therefore a totally new concept of the central backbone was developed. The design (that is still in place at this time) consists of a ring structure connecting three central nodes. At these nodes the faculty LAN's are connected to the backbone and one de the nodes is connected to the national network organisation to provide the university its Internet connectivity. The Wide Area Network that is the result of this configuration stretches over a oval with a length of 10 km and a breadth of 5 km. Approximately 7,500 end user stations within the university buildings are connected to this infrastructure. Additionally 8,000 users have their PC's at home connected to the infrastructure through various means (telephone lines, fixed connections and the TV cable infrastructure). As mentioned earlier the faculties have their own IT department and the operational and financial demarcation point between the IT organisations lies at the "front door" of the faculty. This concept has two main drawbacks, a technical one and an organisational one. Technically the infrastructure is fragile as, if one of the nodes fails, on an average one third of the customers has no external connectivity and the other two thirds of the customers are seriously limited in the their connectivity within the university. The second drawback is that without additional organisation changes the central organisation can not give point to point connectivity guarantees as part of the infrastructure that supports the connection lies outside its management domain. A golden rule of management in general and IT management in particular is that one can not promise deliverables if one cannot control the processes that are to deliver these deliverables. These two drawbacks had to be solved as part of the strategic movement of the university. Starting 2001 the following infrastructure will gradually replace the existing one:
Robert Janz
The central ring structure will be replaced by a quadrant structure where every corner is connected to all three other corners. Every major building of the university will have two separate connections to two different nodes on the backbone. The backbone will support bandwidths up to several Gbps and the access bandwidth ranges from 1 OMbps to 1 Gbps. Technically this concept provides sufficient redundancy to guarantee 1 * 24 hrs operation of the backbone and the connectivity to the faculty units. Even with the redundancy in place the management of the faculty LAN's is still beyond the reach of the central organisation and therefore point to point guarantees still cannot be given. In order to achieve this goal two parallel organisational movements are to be initiated. First IT is to be taken from its pedestal next to the faculties and is to be fully integrated in the processes of the university. Secondly centralised IT management should be replaced by a co-operative management based on a process and service oriented approach. At this moment the first thoughts are maturing on how to restructure the product oriented organisational structure into a service and customer oriented orgaisation. The primary processes of the university are research, education and scientific information processing. In the years to come we will see that IT will merge into these processes and most likely will not be visible as a separate discipline. A sustainable IT organisation will secure its future by seeking a close co-operation with the processes it supports and in time will have its fate intangibly entwined with the fate of the primary processes. Thus a solid and stable infrastructure is a necessary precondition for an IT environment for research and education. But it not enough! Professional management and a user oriented support structure is necessary to convince users that they need a dedicated organisation to provide them with the necessary services.
6
Managing IT in an Academic Environment
ITIL as management approach Although infrastructure is still a major problem in the countries that are represented by the audience the service oriented approach should not be neglected. Especially if sustainability of the IT organisation is at stake, the integration with the processes is very important. In order to transform from a product oriented organisation to a process and service oriented organisation a different approach to IT management is necessary. The infrastructure is no longer the issue that is being managed, but the services that users are needing in their daily work. One of these approaches is called ITIL (Information Technology Infrastructure Library). It would take a substantial portion of the whole workshop to fully explain ITIL so only the highlights will be pointed out. On the WEB there are several sites that can provide more information on this subject. According to ITIL the management of an IT infrastructure is divided into several processes and the central theme of ITIL is the Service Level Agreement (SLA): a detailed description of the features of the service that is being delivered. The ITIL processes are all directed to maintaining the service levels that have been described in a SLA. Each process covers a certain aspect of the total IT management and all activities related to this activity are carefully documented. Some of the most important processes are listed in the figure below:
Robert Janz
7
Users only have contact with the Problem Management (e.g. the Helpdesk). If a problem indicates that a certain service level is not been met, this results in a controlled modification of the infrastructure (hardware, application of operational procedures). Through Configuration Management the modified infrastructure is documented and the support department is notified of the changes. There are strict procedures regarding the information exchange between the various processes and the authorisation of the changes. With the documentation of the service levels and the procedures within the service organisation the knowledge is extracted form the IT staff and made available: 1. Users know what they may (and may not) expect of the service 2. Management is provided with the necessary information for escalation procedures and customer relation. 3. The departure of IT staff does not result in unsupported activities 4. The staff knows what to expect from their colleagues and external IT partners and they know what is expected from them. It is quite understandable that one might view this approach as a theoretical model that is far beyond the daily practice of the IT organisation that you are setting up in your countries. However within time it will also be necessary in the emerging IT countries to structure their services if one wants to provide sustainable network services. For this ITIL need not be implemented to its full extent, but if one gets accustomed to the mental framework that lies underneath the model future management will be made easier. As some point you will start charging users for the services and in return they will demand that you explain what they are paying for. And as the IT environment gets more complex with additional players explicit procedures will become necessary. Also if the services are defined together with the users they feel committed to the environment and the organisation that is responsible for the support. Clearly ITIL is something that adds to the sustainability of the IT environment as a whole.
Innovation versus operational stability One of the challenging aspects of IT management in the academic environment is the mixture of innovation and operational stability. By nature scientists and students have an explorative nature and for the IT support department there is a constant tension between providing state of the art services that at the same time are stable. Actually this is one of the specialities that is required of academic IT support. If one is too strict with procedures (as were mentioned above) users will try to get flexible services elsewhere. On the other hand if one introduces too many innovative applications or infrastructure so that the continuity of the services is endangered the users will also go elsewhere in search of stability.
8
Managing IT in an Academic Environment
At The University of Groningen several measures have been taken to relieve this tension. First the lower layers of the infrastructure have been defined as forbidden territory for any innovative experiments. As service disruption in this area leads to disruption of all the services that are dependent of the infrastructure the impact of experiments will damage the main and administrative processes of the university. In the era of the mainframe it was acceptable that on certain days the only computer of the university was put out of service for a large number crunching job of the Chemistry department. Other scientists just had to do other things in the mean time. Last year a necessary reconfiguration of the cabling to the central router on a late Friday evening resulted in a formal complaint from a department as some students missed a deadline for an end of term paper. For this reason the Computer Centre has introduced the concept of central directorship of the first three OSI layers: cabling and network protocols within the university are the domain of the Computing Centre, independent of who is the formal owner of the infrastructure. Departments that do not obey this rule are (in theory and never put to practice) disconnected from the network. The Quality of Service of the infrastructure is described in a Service Level Agreement that not only defines the rights of the users but also the duties of the local IT departments.
In the middleware and application layers above the first OSI layers more freedom is allowed, albeit only under certain conditions. The most important condition is the clear distinction between unsupported project activities and defined services. Every (larger) activity that is taking place on the infrastructure is described in a project plan. This plan not covers financial and staff related issues that are necessary but also defines the moment at which a Service Level Agreement is produced.
Robert Janz
9
As soon as the Service Level Agreement is agreed on the IT application is supported and the service is part of the service environment that was touched on in the chapter on ITIL. From this moment on development of the service is kept to a minimum in order to ensure a stable and reliable service environment for the users. It takes years before all the procedures described above are implemented and IT staff and users act according to the ideas that were presented. At the Computing Centre there is still a lot of missionary work to be done before all the aspects can be put to practice and are accepted by the various players. However it is already clear that the chosen path leads to a stable IT environment with sufficient flexibility for the academic environment.
National infrastructure
10
Managing IT in an Academic Environment
As one sees the next generation is merely a continuation of a evolution has been taken place for more than a decade. And I know that when the SURFNET connection went from 9.6 kbps to 64 kbps in 1989 we thought that we would never be able to fully use this connection. Now we are implementing gigabits per second, but there are already users who have asked when they will be able to count on terrabit per second bandwidths. The national infrastructure consists of four giga-pop's that have redundant interconnections, the universities and the research institutes connect to the giga-pop's:
Not only the national connectivity is addressed in the Gigaport initiative, but also international bandwidth will undergo a major upgrade. By the beginning of 2002 The Netherlands will have a state of the art infrastructure for research education. The applications for this network are being developed in a separate initiative: Gigaport applications. Together with the industry new applications that are eventually meant for the mass market are to be tested on the Gigaport network. Examples of these applications are video on demand, tele-medicine large scale visualisation projects. The Gigaport ambitions would have never been possible if the research and education community in The Netherlands hadn't organised themselves in the SURF foundation. This foundation is 51 % owner of SURFnet, the national network for research and education and also the owner of several other IT related companies that provide services for the academic and education community. By setting down common goals and combining the efforts toward these goals the community have proven to be a reliable partner in the future innovations. Government and industry is therefore willing to co-fund initiatives that have put The Netherlands in a leading position with regard to IT infrastructure and services.
Robert Janz
11
Final remarks There are several points that may be taken from this presentation and put to use in countries that are only just beginning to set up an IT environment for research and education: 1. A sustainable IT organisation can not survive if there is no commitment from the (top) management. 2. IT should be integrated in the processes. So setting up the IT infrastructure is not sufficient, there should also be a directed redesign of the processes where IT is being used. International donors can play an important role here by combining infrastructure support with proper management support and transfer of knowledge. 3. Co-operation and co-ordination is of critical importance in providing a sustainable IT environment for research and education. The field of IT is getting too complex to able too have all the expertise within one's own organisation. 4. The academic environment is characterised by a tension between operational stability and innovative flexibility. It is the challenge of the IT department (on local university levels as well as at the national level) to merge this tension into daily operations and remain flexible at the same time. Of course 45 minutes is too short to cover all the topics involved in managing the complex IT environment in the academic setting. Therefore this presentation is to be seen as a description of the topics to be discussed and clarified in the informal discussions in the next few days.
This page intentionally left blank
Creative and Innovative Network Management O. B. Popov (Ed.) IOS Press, 2003
\3
Oliver B. Popov
Building a National Research and Education Network "But most importantly, perhaps one needs product champions. For only these individuals can create all other conditions." Peter A. J. Tindemans
Introduction A few recent scientific and technological advancements have had such a profound influence on our lives as networking and the Internet. The Internet (arguably the quintessence of networking) offers the possibility to connect almost everyone using the same set of principles and concepts. The global adherence to open standards are behind the tremendous growth of the technology and its applicability to wide range of domains. Indeed, this Olympian attribute provides the potential to reach every segment of the society. It is a technology that compresses both time and space by giving the individual an access to information, which enriches his knowledge and the freedom of choice. While this might be a very personal view it nevertheless reflects the notion that an entity such as NREN (National Research and Educational Network) is a very important factor in bringing the Internet connectivity and networking to the research and educational communities both in developed and developing countries. However, the later is even more pertinent to the developing countries where markets are not yet developed and the determination for democracy does not match the reality of practicing it.
14
Building a National Research and Education Network
The prerequisites for the creation and the existence of research and education networks are not limited to technological proficiency. They also very much depend on policy making, management, legal and financial issues. Exhaustive consideration and detail analysis of all these phenomena are beyond the scope of this article, however we shall try to enumerate some of those issues, as well as present ways to address and eventually resolve them. There are differences between Europe and United States concerning their respective approach in dealing with many societal phenomena and structures. And that is good, since while the ultimate goals are the same, the means and the venues that differ give wider spectrum of alternatives. This might be very helpful for the developing countries when in the very beginning all options must be under consideration. This line of reasoning also applies to research and educational networking. In the States the academic communities have always in mind the market and the forces of the private initiative. Consequently, they often use the highly competitive environment to achieve acceptable prices and quality levels, as well as to induce some kind of equity. It is a common knowledge that many research and education networks, and in fact the Internet, have been developed through federally funded research projects. The same more or less is true for the Internet2 initiative. Traditionally in Europe, there has been much stronger presence of the government especially in the main sectors of the society such as education and health (defense, foreign policy and the legal system are assumed by default). This concept is also evident in the research and education networking where almost all of the national research and academic networks have been established and financed by their governments. Very often the actual implementation and operation of the NREN is assigned to a single organization that originates from the research and education environment (Geoff Huston calls is "the grand experiment"). Each NREN is usually in the rank of national agency or office under the umbrella and supervision either by one or several ministries. This of course should not be confused either by the physical location of the actual network or even its legal status in the beginning. It is clear, that in some form or another, directly and indirectly, the government plays and should play a vital role in the development an institution/organization responsible for the internal and external connectivity of the national research and education institutions. This is also consistent with the argument that both science and education have a crucial place in creating economic growth, foster national and international understanding and cooperation, and thus create the right climate for political and social stability that improve the well-being and the life of every citizen.
Oliver B. Popov
15
Motivation, Activities and Organization The ambition to define formally what a NREN is might prove somewhat elusive and will eventually produce ephemeral results. Nevertheless, for the purpose of the argument NREN is defined as an entity whose goal is to establish and to operate a computer communication network that interconnects research and educational institutions within a region (often this region is a country) and provides also external connectivity to other similar networks and to the global Internet. While provision of connectivity is the major goal, the mission is to create the right environment for the science and education communities to facilitate and improve the quality of research and education by unconstrained flow of information, provision of information resources and tools. In addition, the network may be used as a test bed for new network technologies and services. It is fair to say that the test bed feature of the NREN has become slightly sidetracked due to the enormous growth in number of users and diversity and complexity of new services. To attain its goal and mission, the NREN must perform the following activities: • • • • • • • •
Provision of network technology necessary to interconnect research and academic institutions with in the domain of the NREN and to connect to the Internet. Identification, development and distribution of network services Analyses and implementation of network technology User assistance and support Education Participation in international network organizations Transfer of network know-how to other institutions and enterprises Contribution to the establishment and implementation of national strategies for the proliferation of the information technology
NREN is much more then a communication and computing infrastructure. However, it must take care that the infrastructure either exist or must be built (the physical lines are usually purchased from telecoms, railway systems, and power distributors, and satellite vendors). While the internal infrastructure is often the responsibility of the participating institutions, the NREN should ensure that there is communication and computing equipment that will provide stable connection and intended services. This might be another venue of activity for the NREN, where it acts as a consultant to other institutions whenever they need to develop and extended the present infrastructure. There are several benefits with this approach: more or less homogenous infrastructure that facilitates maintenance and upgrading, and bulk purchasing and licensing drive down the prices (hence more money for connectivity and services).
16
Building a National Research and Education Network
Services in the form of access to databases, information systems, archives, on line computation and collaboration, and interactive learning are must for the academic community. Most of the end users do not care about bandwidth, congestion, routers and cables, all they see are their needs framed in services. NRENs provide connectivity for educational institutions. Therefore they should understand the importance and the technology of education. They have to educate their staff and through them educate their users. Most of the international networking organizations, that the NRENs are either or might become members of such as TERENA, RIPE NCC, CEENet and ISOC organize numerous educational events, conferences, workshops, and seminars on variety of technological, managerial and policy issues. International governmental and non-governmental organizations such as NATO, EC, SIDA, UNESCO, UNDP, and Soros foundation are behind the financial support of these events. These meetings not only extend the knowledge of the NREN employees, they are also a forum for exchanging experiences and best practices among peers. The partcipants have the opportunity to learn about real life problems describing similar situations in a different settings, get better understanding of the issues they face and the ideas how to resolve those problems. Moreover, the NREN staff will gain the essential pedagogical knowhow to train their colleagues and users, by replicating some of the educational events in their respective countries. That is of course a part of the rationale why the NREN should become a member of TERENA and CEENet. Another reasons are to be in the mainstream of global developments and advances in research and education networking, the opportunity to participate in joint projects and also to become eligible for cross-boarder and multinational funding. This is also a possibility to make the voice of the NREN being heard and the problems shared. The Pan-European and other trans-national organizations give a lot of support to their members and they increase your leverage when you deal with the local authorities. Sometimes membership fees will prove to be cost prohibitive, but many of these organizations have special rates for new members that take into account the state of affairs in their respective countries. There should be an active promotion of the work and the importance of the NREN through Web sites, periodical bulletins and press reports. It is a good idea to establish a lobby among young and the Internet literate decision makers. Make them users of the network. This will ensure some form of financial stability and resilience to political and economic turbulences. It will also create a position for the NREN to be involved in planning and developing long-term strategies for the IT deployment in the country. The same criteria as in other academic institutions should be used to select the staff. One should aim for the best and the brightest. Nepotism and favoritism that are so much part of the developing milieu should be avoided. The NREN should aim to be a center of excellence where there is no place for mediocrity.
Oliver B. Popov
17
Courses in networking have become a part of the regular curriculum in almost every university. Therefore, many of the NREN recruits will have a fair amount of knowledge of what they are going to encounter in the course of their daily work. Nevertheless, the profile of the staff should not be limited Computer Science and EE graduates. Networking is not only about technology, it is about people and how they can communicate with each other. The Internet is full with people coming from different backgrounds that have extended the horizon of applications and services. Young people are not only able to learn, but they are not afraid to. Naturally, a right dosage of experience and maturity must be exercised, especially in the areas of policy making, financial management and collaboration with the authorities. The disadvantages of working for a governmental institutions, where usually there are caps on salaries, is compensated by the enthusiasm that something unique is being created. The combinations of opportunities to be at the cutting edge of the technology and to learn in an international setting are incentives that will make many people to stay. However, some will leave for PNOs, commercial ISPs, or will go abroad. They of course will take with them the time and the knowledge (which at the end translates to money) the NREN has invested in them. Others will come. Eventually, the flow of know-how through the NREN will have a positive fall out on networking in the country. This is one of the pioneering responsibilities of NREN, to overcome the initial frustration with the new technology and to improve the Internet and networking literacy. The start should be with a very small structure and organization. All the resources that are available and in place are to be used, for example some people from the staff in the computing center. There should be a clear distinction between their previous responsibilities and their new ones. It will help if an additional importance and value is placed on their work. This will inevitably cause some frustrations, but they will disappear as soon as you move out from the umbrella institution and create the actual organization. Growth should be incremental. Let the number of users and services guide, otherwise the NREN might find itself isolated in the ivory tower of self-importance and obsoleteness. Set up the principles and the rules of the game as early as you can. Some of these include •
• • •
Threat all members and users equally. Promote independence of institutional preferences whenever possible. If it sounds like an "impossible mission", it is. Particularly with respect to the founding institutions and the most intensive users. Buzt, this will increase credibility and appearance of fairness. Introduce clear user policies and try not to make any exceptions. Avoid flag days and strive for transitional functionality whenever new technology or services are to be deployed. Define a small set of services and make them stable. Then proceed with introducing new ones. It is always good to build on a sound and proven ground.
18
Building a National Research and Education Network
•
Stress content orientation and value of the services, not the technological complexity and marvel behind them.
• •
Avoid regulation that might impose restrictions on the basic human rights such as freedom of speech. The need for interconnectivity and interoperability does not imply that security issues should be compromised. Different user communities require different levels of security. Each network on which the national information infrastructure is built must have a number of security procedures implemented that will prevent unauthorized
•
access to the network and the systems that comprise it. The issue of network ethics should also include the copyright provisions and intellectual integrity.
Times are certainly different from the early nineties, since the combinatorial growth of the Internet and networking has also lead to exponential growth in the volume of knowledge about them. Key people from the countries that are in the process of establishing NRENs, both on technical and managerial level, have actively participated and contributed to CEENet meetings, conferences and workshops. They have also established working relations with the international organizations and used some of the grants and donations to improve the network infrastructure in their respective countries. Then again, some of the already enumerated concerns and problems remain the same. In fact they are indispensable part of being NREN and solving them is what is all about. As indicated earlier, it is extremely important to establish an organization and a structure that will formally co-ordinate all dispersed efforts and represent the research and education networking in front of the domestic authorities and international organizations. This organization must be inclusive and should try to accommodate sometimes conflicting interests of prospective member institutions. The academic community that initially serves as a cradle for the NREN usually comes from some university department or a research unit in the academy of sciences. Previous contacts with colleagues from abroad provide information for funding possibilities, which have been quite few in networking, and then the first connection to the Internet is made. The assistance and the support from outside could be that necessary catalytic funding that makes the Internet connectivity possible for the research and education community. In the mean time some other institution realizes the importance of this activity, and does the same thing with the Internet connectivity based on the approval of another ministry and funding from other institutions. There is no justification for this kind of behavior, except self-centered interests and the abuse of the strategic role of networking.
Oliver B. Popov
19
While it is understandable that when large countries are involved with many academic institutions it is reasonable to have may be a few organizations that cover the area of research and education networking, it is still highly desirable that they work together and coordinate the representation with respect at least foreign institutions and organizations. It is indeed unacceptable, especially in small countries with limited resources that two or more ministries or state agencies based on the information supplied by two or more superficially competing institutions, exercise their rivalry this way. At the end they even try to establish two or more NRENs. Combined with primitive forms of local politics, the final result is a superposition of disconnected efforts and split benefits. And the whole picture does not look very nice in front of the international donor institutions and organizations. It also creates a flaws and strategic weaknesses when dealing with the national authorities who also sometimes have peculiar interests in the PNOs and commercial ISPs. Fortunately, the international organizations and governments are aware of these excesses. Consequently, they try to remedy the unwarranted situations by mediating between different parties, bringing them together and awarding grants only to either joint projects or projects that apply to a more general situation. This effort occasionally transcends national boarders by trying to bring academic communities between different countries, especially in regions where there is a history of intolerant behavior. Any institution that is involved in the creation of the NREN must see the new organization as bonding and unifying factor among all the scholars, researchers and students in the country and the institutions they represent. The NREN should not exercise either ignorance or arrogance towards any potential user community. To do so means that the NREN behaves in the same way as some of the PNOs have done in the past and still do until the process of deregulation is over. Here also the national independent regulative body plays an important and stimulative function. One should closely work with this institution and eventually delegate, if possible, one of its members some of its decision making structures. Despite real problems, which are relics of the past, the NREN must work also with the national communication carriers since most of the time they provide the internal infrastructure. No effort should be spared to explain that there is no place for competition, and there is a clear division of the responsibilities and the functions. On Eventually the NREN activity creates larger user community for the PNOs also The need and the desire to co-operate, associate or become a full member of EU for many countries is a major incentive to undertake regulatory reform of their telecommunication sector. The slow entrance of other carriers that generate alternatives should bring down the prices and make the operation of the NREN financially more feasible.
20
Building a National Research and Education Network
Financing and funding There are basically two distinct funding models for services provided by NREN: • •
User/service dependent or a user funding model User/service independent or central funding model
On many occasions it has been posited that is very hard to identify the optimal model for funding research and education networks, if there is one. It is an ongoing debate for the last ten years or so. It has recently experienced somewhat of a renaissance possibly due to (1) the deregulation of telecommunications in EU and the start up of a similar process in the countries of CEE and FSU, and (2) the process of establishing new NRENs again the CEE and FSU countries, and (3) the NRENs in EU have become almost indispensable entities (as networking is) in the academic communities, so there is a need for stable and efficient model of funding (in the same manner as the rest of education and research have). The short history of the academic and research networking in Europe indicates the same pattern in the beginning, namely all of the NRENs were funded from a central source or the government. On the opposite end is the so-called user dependent funding. The network recovers the costs from the institutions it connects and provides services for. The funds may come, and they often do, again from the government, but on the surface a particular institutions is free to buy network services from anybody on the market. It is fair to say that especially in developing countries, initially there were very few providers of any connectivity and usually the NREN for a prospective user was the only place to go. The picture is different today and the pool of choices is much larger. This sometimes creates a superficial argument that there is no need for NREN, and that a research or educational institution should go to the local ISP and ask to purchase services. Indeed, there are many flows in this position. Firstly, it reduces the NREN to an ISP, and there are many more objectives and activities in the realm of the NREN that are beyond the interest and the competence of an ISP. Secondly, the NREN is an extension of the old forms of academic institutions and in Europe, very few have been either replaced or surpassed by private ones, which is partially due to tradition, but more so to it is the reflection of the continuous responsibility of the government, as the primary social moderator to support the best in research and education on a long term basis. Moreover, the intellectual breadth and depth that emanates from the academic institutions represented by the diversity of ideas and concepts, creates a fertile ground for innovative services and modes of usage. Hence, the argument that it can be done by some commercial ISP negates the role of the NREN is a test bad for new technologies and services. Faced with the market risks, an ISP runs only well established and profitable services. Finally, the commercial ISPs have either very little or no experience in educational practices, which are also very important feature of NRENs. Obviously this does not mean that the NREN should enter the state of complacency; on the contrary it should be highly sensitive and responsive to the needs of its users.
Oliver B. Popov
21
Let us explore the pros and cons with respect to the two models of funding. In the case of central funding, the advantages are simplified administrative procedures, more efficient planning and strategy development as well as financial control and reporting, and direct attentiveness of the government to the needs of the academic community with respect to networking. The disadvantages are the notion that the users are out of the loop with respect to both the range of services and their quality, the transparency of the funds transmission, which makes the users think that the operation of the network, the provision of services and the cost are minor issues that also affects their behavior and adherence to the user policies, and finally the susceptibility to political perturbations and instabilities that are so characteristic for many societies in transition. Many times you will have a government official who is a great proponent of the research and education network (the motives are not important and occasionally they are very mundane) that will go beyond his way to support the development and the operation of the network. A few months later, another government is in place, and all of sudden there is a complete lack of interest and ignorance, that puts in peril everything that has been accomplished. Some of them question the existence of the national network. One should not forget that the essential attributes of a civilized society are moderation and recognition of the invariant societal values (which certainly include research and education). The user oriented model has also its advantages and those are: influence on the types of services that are either running or should be planned for the future, the freedom of the users to use the market as the incentive for the complacent behavior of their network provider, and the opportunity to be aware and control the costs and the expenditures. It might also provide additional immunity from mood swings in the government due to the users who have been used to services and would like to maintain them, in spite of the current mode of reasoning by some government officials. But most of the users, even in the academic communities are self-centered and think only of those services that are important to them and are working well, so very little money for innovations and test-bed activities. It will be very difficult to plan and develop long-term strategies, since their priorities will not always match the needs and the requirements of the users. Probably the safest (the qualification "best" is deliberately omitted) way is to mix the two models. It is very likely that one will start with a central funding model, with supplementary funds from international governmental organizations (EC, NATO, UNDP) and non-governmental organizations (such as Soros foundation that has played an extremely beneficial role in the CEE and FSU regions, particularly with respect to networking). Even when the funds are directly routed to the NREN from the government or from any other institutions, ask for a breakdown of the funds and costs (for example how much money actually is provided on behalf of some participating instruction). This will make every institutions aware that there is nothing for free an that actually someone pays for the services received, as well as fully conscious about the work and the effort that goes in provision of services. The assistance that comes from the outside will not last forever, and the
22
Building a National Research and Education Network
intent is to help you help yourself. After a while, if there is no progress, the focus of attention either shifts to another organization or to another country. As things converge towards stability move some of the services towards full user funding (primarily those that are established). For a long time, if not always, there will be a need of some centrally based funding for implementing innovative services and realization of the long-term strategic plans. If you are small country, both in size and population, it might help you to diversify you users. Never forget that they are you main asset and reason to exist. Extend the research hat to include research and development units within the industry. They might become the experimental ground for user oriented funding; will most probably enrich the spectrum of services and applications, and bring some financial stability. Extend the education hat to include both primary and secondary education whenever possible, libraries and museums. The NREN should not be necessarily directly involved in all those extended activities. The assistance may be in a form of a technical expertise, user support and education. Hopefully, a few of them will gain sufficient knowledge in some aspects of networking. At a later stage, you might use them to outsource a number of services or operations. It might prove to be of a mutual benefit to provide connectivity to various NGOs and other non-for-profit organizations. Some of them might have directly contributed in the past, some of them will do so in the future, but NGOs are very important aspect of the civic life and they, with the help of your default - academic users, might be in the first line of defense if sometimes a dubious decision by somebody in the government poses a threat to the NREN. A well-planed budget and transparent execution of it is essential. NRENs are fairly new type of institutions and not everyone understands either the work they are doing or the need for them to exist (sometimes they are treated as an overhead). The items that comprise atypical NREN budget are: connectivity (national and international), personal, equipment (hardware and software), research and innovation, and education and training. Actually, the former three might be put under a category termed as network operations, where the dominant part of the expenditures goes to the international connectivity (it varies from 50% to almost 80% depending on a country). At the end, the choice of the appropriate model will depend on the local conditions influenced by the political and social culture, touched by tradition and transformed by the transitional economic realities. The users should be always made aware of the cost, the effort put into running the NREN, and the benefits they get from the services. The NREN is usually a non-for-profit institution and organization, and as such is subject to assistance from the international governmental and non-governmental organizations. On the other hand, some of the NRENs are trying to commercialize certain segments of their work. This type of behavior should be avoided possibly at all times, but at least in the first five to ten years.
Oliver B. Popov
23
The road is very tempting. First of all, it creates conditions where one could become slightly more financially independent from unreasonable users or myopic administration. Secondly, it will be motivation for the people who work at the NREN to stay (the competition from the PNOs who are suddenly Internet stricken and commercial ISPs is fierce and cruel), since they may supplement their regular income, which might be rather low in a country that goes through economic and social transformations. PNOs and commercial ISPs might accuse you for unfairness since your institution is being subsidized and this induces market imbalance. It will eventually jeopardize the possibilities for international funding, and most of all the institution might loose the moral ground that is unique for academic communities. So far, we have restricted the discussion about the legal issues to the telecommunication laws, the process of deregulation and the existence of independent regulatory body. Within this framework, the entity of NREN has been addressed as an institution, organization and/or association. There is no doubt that the legal status is one of the key pillars in creating a stable and sustainable network for research and education. While in the very start during the so called "project state" of the NREN, the legal representation is usually confined to university department, university itself or a unit in the academy of sciences, sooner or later as things start to grow the need for financial clarification becomes a pressing argument. Financial clarity that directly influences both the responsibilities and the rights of the entities should be unambiguous and subsequently based on laws. However, most of the countries that have either entered the process of creating NREN or are about to do so, have rather short history of statehood (their national constitutions are usually not older then ten years). Consequently, in some of those countries the number of so-called system laws that have been adopted by national parliaments is modest. Few of these laws deal with the state of public institutions and organizations (including national offices, agencies, and associations). Eventually these laws will be enacted. The question is how to proceed in the interim period. If possible then the best solution is to be a national public entity (an agency or an office) with a full legal status. The second choice is to be an association of legal entities that is also a legal entity. A third one is to be association of legal entities, which itself is not a legal entity, but is under an umbrella of a legal entity (two of the most advanced networks in Europe, ACONet from Austria and SUNET from Sweden are not legal entities themselves). If one wants to introduce a user-based model there are several options for financing. In all of these options, the most important is the mechanism of granularity that will help you set a proper payment scheme that acknowledges different levels of usage. The share-holder model, where the number of shares per organization is for example proportional to the capacity of the access lines or the volume of data in certain period is acceptable, but the term "shares" does not go well with the idea of non-for-profit association. One way to circumscribe possible legal problems is to introduce the notion "categories of users", where again based on some criteria each user is a member of a category and pays accordingly.
24
Building a National Research and Education Network
Formally, the fees should be declared as a sort of membership. This might help to preserve the non-for-profit status that usually includes preferential treatment by the authorities, namely tax exceptions and duty free/custom benefits. Whatever option is considered, the final decision will highly depend on the local conditions naturally inspired and guided by similar solutions in other countries. In any case, national laws should be closely monitored for any changes and/or modifications, since the convergence towards sustainability is also a function of the legal status.
Conclusions Arguably, technology reflects the needs and the interests of the social forces and has never been politically neutral. The NRENs are case in point. There is so much politics involved in their creation that most of the time it is counter productive. The real question is how to make these institutions and their normal operation immune of political perturbations. This will also generate the necessary conditions for self-sustainability and growth. The steps and the tasks involved in the creation of NREN look sometimes formidable. But as any other innovative things in the history, they very much depend on pioneers, people with vision and courage to work for the public good. Science and education are the best way to prepare for the future. Today they simply are not possible to do without computer - communication networks. It has been stated that NRENs are new forms and extensions of the academic communities in the past and it is very likely that soon they will become, in some form or another, as indispensable entities as universities and research institutions are. We have tried to enumerate a few issues that are pertinent to the creation and the establishment of research and education network. They are based on the experience and the knowledge of many networkers involved in the creation and running the NRENs. Obviously there are no universal procedures that describe the best way to do it. One can only hope that some of the problems argued and solutions illustrated might prevent you for repeating the errors others have made and somehow make your road to sustainability shorter.
Oliver B. Popov
25
Bibliography 1.
2.
3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13.
14. 15. 16. 17. 18. 19.
Bonac, M. The Academic and Research Network of Slovenia in The Avant-garde of Information Society, Proceedings of the NATO ANW - The First CEENet/ TERENA Workshop on Network Management, CEENet Editions, Prague, 1999, pp. 42-49. Chalmers, A., Duxbury, C. The Hidden Economic and Societal Issues of Policies on Advanced Networking, Computer Networks and ISDN Systems, Vol. 28, No. 14, November, 1996, pp. 1991-1998. Dixon, T. Obstacles on the Road to Global Networking, Computer Networks and ISDN Systems, Vol. 25 Suppl. 1, pp. 9-18. Frederick, H.H. The Internet and Society, A Lecture Presented at The Second CEENet Workshop on Network Technology, Budapest, August 1996. Gajewski, J. Private Communication, 1996. Galagan, D. Private Communication, 2002. Hallgren, M. Funding an Internet Public Good: Definition and Example, Computer Networks and ISDN Systems, Vol. 27, No. 3, pp. 403-409. Hofrnokl, T., Academic Networks in Central and Eastern Europe, INTEROP'95, Frankfurt, 1995. http://www.ceenet.org http://www.gipiproject.org http ://www. isoc. org http://www.terena.nl Huston, G. Desperatley Seeking Default Internet Policy Update: A Perspective from the Pacific, Computer Networks and ISDN Systems, Vol. 26 Suppl. 1, pp. 3-14. Kaufman, P. The Implementation of a High-speed Network for DFN-community, Computer Networks and ISDN Systems, Vol. 26 Suppl. 4, pp. 179-186. Klark, D., Pasquale, J. et al. Strategic Directions in Networks and Telecommunications, ACM Computing Surveys, Vol. 28, No. 4, December, 1996. Nadreau, J.P., Popov, O.B. Networking Developments in the Caucasus Region, Proceedings of the NATO ANW, NATO Science Series, IOS Press, 2001. Nielsen, S.M. EuropaNET- Contemporary High Speed Networking, Computer Networks and ISDN Systems, Vol. 25 Suppl. 1, pp. 25-34. Noam, E. Telecommunications Policy Issues for the Next Century - Toward A Global Information Infrastructure, USIA, November, 1994. Pale, P., Information Technology, Academic Community and Future: ARNet Model in The Avant-garde of Information Society, Proceedings of the NATO ANW - The First CEENet /TERENA Workshop on Network Management, CEENet Editions, Prague, 1999, pp. 193-202.
26
20.
21.
22.
23. 24.
25. 26.
27.
28.
Building a National Research and Education Network
Popov O. B., (Ed.) National Networking Effort: Concept and Strategies in Visionary Ideas for a Visionary Future, Proceedings of NATO ANW - The First CEENet Workshop on Network Policy, CEENet Editions, Prague, 1998, pp. Popov O.B., (Ed.) The Avant-garde of Information Society, Proceedings of the NATO ANW - The First CEENet /TERENA Workshop on Network Management, CEENet Editions, Prague, 1999. Popov O.B., (Ed.) The Road to Global Connectivity, Proceedings of the First CEENet Workshop on Network Technology, CEENet Editions with NATO ANW Series, Warsaw, 1996, pp. 9-12. Rastl Peter, Coordinating Networks in Central and Eastern Europe, CEENet, Proceedings of INET'94/JENC 5, Prague, June 1994, Book 2, pp. 424/1-424/3. Schill, A., Kuhn, S., Breiter, F. Internetworking over ATM: Experiences with IP/ IPng andRSVP, Computer Networks and ISDN Systems, Vol. 28, No. 14, November, 1996, pp. 1915-1928. Swaay, van. M. The Value and Protection of Privacy, Computer Networks and ISDN Systems, Vol. 26 Suppl. 4, pp. 149-156. Tindemans, P.J. Computer Networking as an Essential Infrastructural ProvisionThe Role of the Government in The Avant-garde of Information Society, Proceedings of the NATO ANW-The First CEENet /TERENA Workshop on Network Management, CEENet Editions, Prague, 1999, pp. 53-92. Vietsch, K. Funding Models for NRENs in The Avant-garde of Information Society, Proceedings of the NATO ANW - The First CEENet /TERENA Workshop on Network Management, CEENet Editions, Prague, 1999, pp. 193-202. Vietsch, K. Private Communication, 1998.
Creative and Innovative Network Management O. B. Popov (Ed.) IOS Press, 2003
27
Karel Vietsch
European and United States Models ofanNREN
The NREN model Research networking in Europe is organised in the form of national research and education networking organisations (NRENs), which provide Internet connectivity to the research and education establishments in their countries, and which collaborate at a European level. Every country has a national research network organisation, and there is only one such organisation per country. This gives Europe an important organisational advantage over, for example, the United States. We will list some of the advantages of this organisational structure. And at the same time compare it with the US. First of all, national research networks in Europe have a manageable size. The smallest ones cater for two universities; the NRENs in the biggest countries may have some 50 universities and a few hundred schools as their customers. A similar network in the United States would have to cater for many hundreds of universities and many thousands of schools. For that reason alone, such a network does not exist in the US. The United States does have national backbones; the most important one at this moment is probably Abilene [1]. Universities can connect to Abilene at so-called GigaPoPs; there are currently about 40 GigaPoPs. A recent development is that certain regional networks are developing around those GigaPoPs. For example, different campuses of Indiana University are being connected to each other and to the Indiana GigaPoP at high speeds, and some other universities, also in neighbouring states, are being connected as well. In that way a research network is growing that covers an area of 200 miles from north to south and 150 miles from east to west, around that GigaPoP. Still, that does not compare yet to an NREN in Europe, as will be explained below.
28
European and United States Models of an NREN
Strongpoints of the European model 1. 2. 3. 4. 5.
NRENs in Europe have a manageable size NRENs in Europe speak the language of their users NRENs in Europe cover the same geographical area as their funding bodies The European NREN structure follows the traffic patterns The European NREN structure is not dependent on a particular network topology
European NRENs speak the language of their users. That statement can be taken literally: one of our handicaps in Europe but also one of our Strongpoints - our cultural Strongpoints - is that in our different countries we speak different languages. Because NRENs are organised on a national basis, the people in the NREN speak the same language as their users. But the statement can also be taken in a more general sense: because the NRENs and their users have the same cultural background, they understand each other much better than if a Norwegian would have to provide services to a customer in Turkey. Or, for that matter, a New Yorker to a school in Alabama. Thirdly: national research and education networks cover the same geographical area as their funding bodies. At the European level we tend to speak a lot about the European Commission and the funds that the EU will (or will not) make available for research networking. But we should remember that 70% of the cost of international research network traffic is paid from national sources. And of course so is 100% of the cost of national and local traffic. The fact that the customer base of each NREN coincides with those entities that a national government or science foundation is responsible for - the research and education community in a particular country - makes things much easier. This is not feasible in the US. Funding for research and education in the United States comes from many different sources, both public and private. That funding structure is much more complicated than in Europe and cannot easily be mapped on a geographical organisational structure. Another point is that the organisational structure on a national basis, which we have in Europe, follows the traffic patters. International collaboration in Europe is increasing strongly and continuously, and that is a very positive development. But it is still true, and probably always will remain true, that most of the traffic on research networks is domestic or even local, and only a relatively small percentage of traffic is international. This does not compare to the situation in the US, where traffic patters are very different. Finally, the European structure is not dependent on a particular network topology. Any European backbone will have to connect all European research networks at a national Point of Presence. In the US that is different. The Indiana network that was mentioned before may have grown around a GigaPoP, but if the Abilene topology would change or if
Karel Vietsch
29
Abilene would be replaced by another national topology, the regional network might be left hanging in the air. In the US the organisational structure follows the network topology, in Europe the topology has to follow the organisational structures. That may not always be the most cost-effective solution, but at least it is stable and one knows where one stands in the long run.
TERENA's membership and activities TERENA [2] is the association in which the national research and education networks of countries in and around Europe collaborate. The organisation was founded in 1986, and in the beginning its coverage by national members was simply defined by listing 19 or 20 countries in Western Europe. The Iron Curtain was still very much in place then. As a consequence of the enormous political changes in the early 1990s, the statutes of the organisation were changed in 1992, so that from then on all countries in Europe could join as national members. By 1995, the membership of TERENA had grown substantially. However in recent years TERENA has been losing national members again, especially in the Commonwealth of Independent States. We will come back to the reasons for that worrying development below. There are four main categories of TERENA activities that form the pillars of the organisation. Firstly, TERENA is a political lobbying group, making the case for research networking by talking to many different parties. That is a small but important task. Secondly, TERENA organises conferences and workshops. Thirdly, TERENA acts as a cradle for new networking initiatives and services. And the most important part of TERENA's work is the so-called Technical Programme. Having a large and representative membership, TERENA can bring the common opinions and interests of the European research networking community to the attention of many parties. TERENA has regular contacts with the European Commission and with the representatives in the national governments responsible for research networking. TERENA also gives general presentations about research networking to many different audiences. Very recent examples were presentations to European science ministers in a meeting about the so-called eEurope initiative and presentations at a conference of rectors of universities in Europe. TERENA also collaborates with sister organisations in other continents, such as Internet2 in the US. In recent years there has been an increased interest from industry in research networking and TERENA is in contact with major international companies to see what they could do for the European research networking community. TERENA also liaises with other organisations in the Internet world, such as the Internet Society.
30
European and United States Models of an NREN
What is TERENA? Research and education nowadays depend increasingly on electronic media and networks. Networking services for research institutes and educational establishments are provided by research and education networks. These networking organisations collaborate at the European level, thus creating a high-quality international information and telecommunications infrastructure. TERENA is the association in which the research and education networking organisations from countries in and around Europe collaborate. TERENA's work falls into four main categories: • • • •
representing the common interests and opinions of its member organisations; developing, testing and promoting new technologies and services through the TERENA Technical Programme; organising conferences, workshops and seminars; acting as a cradle for new initiatives and services.
Very important are the conferences and workshops that are organised by TERENA. They offer an opportunity for people from the European research networking community to meet each other, discuss issues of common interest and learn about new technical developments. The term "the European research networking community" does not refer only to the TERENA membership, the staff of the national research networks. There is a much wider community, in universities and research institutes, which participates actively in TERENA's technical activities, or attends TERENA conferences and workshops. The annual conferences stand in a long tradition, going back to 1985. This year's conference took place in May in Lisbon, and the next conference will be in May 2001 in Antalya, Turkey. Both technical experts and managers are very much encouraged to participate in these conferences. They offer a wonderful opportunity to meet colleagues from other countries and to learn about the latest developments. TERENA also regularly organises workshops for some 20-40 people about more specialised technical or managerial topics. The next one will be a European Workshop on Middleware in June 2000, which TERENA will host together with the Telematics Institute in the Netherlands and in collaboration with the Middleware Initiative of Internet2 in the United States. And at the end of the same month, TERENA will be organising a NATO Advanced Networking Workshop, in collaboration with EENet, in Tartu, Estonia. This will be very much a technical training workshop, aimed at young network engineers in Central and Eastern European countries. The third pillar of TERENA's activities is to act as "a cradle for new services". The one thing that TERENA does not do, is to provide operational services. There are other,
Karel Vietsch
31
professional operational organisations that have the management and provision of European operational services as their remit. But the discussions in TERENA and especially the technical development activities organised by TERENA, every now and then lead to an experimental service. Sometimes such an initiative develops into a service that is permanently needed by the TERENA membership and possibly the Internet community at large. In such cases TERENA does not continue the service itself but hands it over to a European operational organisation as soon as the service can stand on its own feet from a financial point of view. Consequently, over the years a number of organisations have been split off from TERENA, such as DANTE, Ebone and the RIPE NCC association. The most relevant of these for the NRENs in Europe is no doubt DANTE. DANTE is the agent owned by NRENs in Europe that organises and manages pan-European services. The most important of those services is the pan-European backbone network interconnecting the NRENs of the participating countries. There has been a series of these backbones - Europanet, TEN-34, TEN-155 - over the years, with ever-increasing capacities and more advanced technologies. The current backbone is called TEN-155. We are currently in a planning phase for the next-generation backbone, which will be called Geant. There the access capacity will go to the range of Gigabits per second. That network is planned to connect many more national networks than before encompassing not only the NRENs currently connected to TEN-155 but also the national research networks in all countries that are official candidates for membership of the European Union. This should mean a giant leap forward for many countries in Central and Eastern Europe. The TERENA Technical Programme is a great collection of activities, task forces and projects, where technical experts from the European research networking community that is that wider community referred to before - work together to develop and test new technologies and applications for future use in the research networking environment. The number of activities in the Technical Programme is very large and growing all the time. Some of the main broad areas covered by the Technical Programme are: Lower Layer Technologies, Videoconferencing and Streaming, Caching and Network Storage, Indexing and Searching, Security, and various other items. A full and up-to-date overview can be found at [3]. All these activities in the TERENA Technical Programme and many others are open for participation by all interested technical people. Getting involved in these activities, if only by subscribing to the relevant mailing lists, is a very good way for newly developing networks to keep informed about the technical developments in networking and the possibilities for implementing them in their own network. It is important to emphasise that all these activities are collaborative activities by people working in NRENs, universities, industrial companies and research institutes. The small TERENA staff, housed in our office in Amsterdam, is there to help organise these activities but cannot take on hands-on work themselves.
32
European and United States Models of an NREN
The NREN model - lessons learnt from the past TERENA has a history of 15 years, and the oldest national research networks in Europe also date from the mid-1980s. Other NRENs were founded later and could benefit from the experiences of the forerunners. For all pioneers in Western Europe, both the managers and engineers in the NRENs and the policy makers in the government bodies, the years from 1985 until 1993 were an important learning period. There we learned how to organise national networks and how to organise international collaboration, through a first pan-European research networking project called COSINE. One of the most important lessons learnt in that period is that it is the organisation even more than the finances - that determines success or failure. In the past ten years research networking in Central Europe has made enormous steps forward, whilst research networking in many countries in Eastern Europe has seen very little improvements, if any at all. Of course this is very much related to the general economic developments in the various countries. But organisation and management are just as important as the availability of money, and in the end probably even more decisive for the development of NRENs. Arne Moi, who for many years represented the government of Norway in COSINE, summarised in 1992 the lessons learnt in COSINE days [4]. Those lessons are still extremely relevant today, for those who want to set up or expand research networks in lessadvanced countries and for those who want to promote international collaboration in these fields.
How to organise an NREN — lessons learnt from the past 1. 2. 3. 4. 5.
Never let the prophets run a project or a service Commit the necessary funds first Have the necessary organisation in place before you start the activity Never let the other side sit in and watch your internal fighting Define a clear division of responsibilities, and stick to it
6. If you want to do something strategic, impress the strategic importance firmly in the minds of the authorities 7. If you want to do something, do it on a grand scale 8. Be vague enough to leave yourself options Lesson number one is that serious activities must be managed by down-to-earth people and not by prophets. Prophets have the answer instead of asking what must be done to reach the target.
Karel Vietsch
33
Number two is that funding authorities must commit the necessary funds first, and then worry about how they are spent. After all, committed money is not yet spent money. Number three is that the necessary organisation should be in place before you start the activity. If you have to force the horse to the water, try another. Number four: never let the other side sit and watch your internal infighting. Obviously NRENs or people trying to set up an NREN do not always agree on everything, but they should sort that out among themselves and not let their disputes be known to the outside world. Conversely, it is not very smart for governments to have the NRENs witness their political differences and internal problems. Number five is much related to that: define a clear division of responsibilities, and stick to it. Do not let politicians determine the topology or technologies of your network. And conversely: do not have engineers determine government policy. Number six: if you want to do something strategic, impress the strategic importance firmly in the minds of the administration and do not let them get away with it. As is often said, ministers are useful animals but they usually do not last very long, and then you have to train a new one. Not all of them take to training very well. In the current positive atmosphere surrounding the Internet, it should be possible to explain to politicians that setting up and expanding an NREN and stimulating international collaboration between NRENs is of great strategic importance. Number seven: if you want to do something, do it on a grand scale. That catches the attention of the politicians; and that is why they are useful animals. Small projects do not carry much weight and do not make a real difference in the real world. Finally: whatever you do, be vague enough to leave yourself options. The future is very unpredictable, certainly in the networking world. Nevertheless, they who sit and wait to see how things will develop will be the losers in the end.
The ^Europe 2002 Action Plan ^Europe 2002 is an initiative of the Portuguese Presidency of the European Union and the European Commission. One of its parts is the establishment of a very high speed network between universities and research institutes in Europe. In his keynote speech to the TERENA conference in May 2000, Portuguese Minister Gago underlined that this is something to go beyond Geant. One of the concrete targets is to realise before the end of 2001 an interconnection between all NRENs in Europe with access speed of tens of Gigabits per second, and 100 Gigabits per second as soon as possible. Minister Gago has asked TERENA to advise him on what must be done to reach that objective. In our answer, which was drafted after substantial discussion with TERENA members, we have emphasised five issues that need the urgent attention of the responsible authorities. Here we would like to mention just one of them, which is of particular interest to Central and Eastern Europe.
34
European and United States Models of an NREN
It is about issues related to the different starting positions of different countries. It should be recognised that the goal to have a trans-European network with an access speed of tens of Gigabits per second available to all NRENs by the end of 2001 is an ambitious one. It is likely that it will not be possible to achieve speeds of tens of Gigabits per second to all NRENs in that timeframe. Apart from the technology factor, the current telecommunications price structure in the wider Europe is such that for some countries, in particular some of the accession states, such capacities would be unaffordable. And even if unlimited resources would be available, the inadequacies of the underlying telecommunications infrastructures would for the moment make the establishment of such very high-speed infrastructures impossible in some of the countries concerned. There is consensus in the TERENA community that Europe's efforts to obtain "Faster Internet for researchers and students" need to balance carefully two factors that can sometimes appear to be conflicting. The first is the need to ensure that network connectivity is ubiquitous to all countries across Europe, and the second is the requirement to work on the leading edge of what is achievable. If we fail to meet the first goal we do a deep disservice to poorly connected countries, and if we fail to meet the second goal we are not using the strengths of the research community to the best effect in support of the overall societal and economic goals. The conclusion should be that the responsible national and European authorities should take additional measures to assist less-favoured regions in Europe to catch up with developments in the more advanced European countries, but that this should not lead to more advanced NRENs being held back. As to these additional measures, the Commission's Action Plan ^Europe 2002 puts an action on the Member States and the European Investment bank to achieve that national research networks are upgraded to ensure that all researchers and students across Europe benefit from equally powerful networks, where appropriate, using structural funds and EIB support. That seems to be a step in the right direction and for some of the countries concerned very substantial investments from these sources would be needed. TERENA also points out that it is the responsibility of the national governments concerned to allow and encourage competition in the provision of their national and local telecommunications infrastructure. The evidence of the past 3-4 years, not only inside the current 15 Member States of the European Union but also as experienced by our members in several of the accession states, is that infrastructure pricing indeed starts to fall quite rapidly soon after effective competition is present. Actually, a true liberalisation of telecommunications regulations in newly emerging economies can open unorthodox solutions that are not so easy achievable in the EU-15 countries. As an example, the NRENs in Poland and the Czech Republic have been able to establish high-speed networks by using the infrastructures of railroads, power companies and oil companies instead of being dependent on traditional telecom carriers.
Karel Vietsch
35
A final lesson Above, we have quoted from Arne Moi's 1992 manifesto "Letter from a citizen of Norway". In a keynote speech at the annual conference of our organisation in Trondheim in 1993, he repeated and expanded many of his thoughts [5]. Although the world has changed significantly in seven years, many of the things said then have a lot of meaning for us today. A final quote may show this: Europe is a fairly loose collection of regions, at present organised in hierarchies of national states. The regionalism of Europe is its strength and its curse. Europe is a manyfaceted place that even if it has much in common, it still has its differences that have not been ironed out in political contexts over hundreds of years and probably will not go away, no matter how much Eurotalkyou apply to it. Instead of homogenising Europe, if that is what we are trying, let us be proud of our diversities, keep them and see how we can benefit from them, and turn them to our advantage. Let us not waste our time on fighting each other, region by region, national state by national state, but see how we can facilitate and ease co-operation with each other, respecting our differences by defining what is good for us and let the rest be.
References
[1] http://www.ucaid.edu/abilene/ [2] http://www.terena.nl/ [3] http://www.terejia.nl/tech/ [4] Arne Moi, Letter from a citizen of Norway, Oslo 1992 [5] Arne Moi, Keynote speech at the 4th Joint European Networking Conference, Trondheim 1993
This page intentionally left blank
Creative and Innovative Network Management O. B. Popov (Ed.) IOS Press, 2003
37
George Sadowsky
The Evolution of NYU-NET in the Context of Regional and National Networking in the U.S.A.
Introduction The purpose of this talk is to relate to you the development of the Internet in one United States research university, New York University. I will observe that its development was intermeshed with that of the regional network, NYSERNet (New York State Education and Research Network) which the University helped to being into existence and in which it continues to participate actively. A strong cooperative environment has existed among the universities in New York State and NYSERNet that has substantially helped all participants. This experience may be relevant for the CEENET region. New York University is the largest private university in the United States. It is composed of 13 separate schools, and has approximately 15,000 undergraduate students, 16,000 graduate students, and about 15,000 people at any one time who are taking continuing education courses. It was established in 1833, and for much of its history was oriented purposely to students in the New York City area. However, since World War II it had become increasingly national and international, and now boasts among the highest percentage of foreign students among U.S. universities. One school, the Courant Institute of Mathematical Sciences, was instrumental in the development of computing and networking at NYU. The Institute was founded with Professor Richard Courant, a famous mathematician, who emigrated from Germany to the U.S. before World War II. After the war ended, The Institute attracted substantially more talent, some of it from the Los Alamos project, which had used applied mathematical methods extensively in the design of the first nuclear weapons.
38
The Evolution ofNYU-NET in the Context of Regional and National Networking
As a result of this background, the Institute quickly entered into a major research relationship with the U.S. Atomic Energy Commission (now the Department of Energy), in which mathematical modeling and simulation were important techniques. As a result, when digital computers were first commercialized, the Courant Institute received in 1957 an IBM 704 computer for use for many different kinds of scientific calculations. It was one of a small group of companies and industries for which automatic digital computation was essential for progress. The Courant Mathematics and Computing Laboratory (CMCL) was quickly formed, and it did seminal work in the development of computational methods and their applications to problems in applied mathematics. Courant has remained in the forefront of this type of computing activity to this day. In 1965, a Control Data 6600 was installed (serial no. 4), followed by a succession of high powered computational engines from various suppliers.
Early Experimental Networking (1975-1983) In the early days of scientific computing, there were different types of computer, all having different operating systems, programming languages, and data formats. Interoperability was the exception, not the rule. The post World War II climate was quite favorable for the growth of scientific research, and applied mathematics research groups sprung up at other research centers and universities. Much of the research required collaboration among this community, and often required use of distant computing facilities to implement such collaboration. This requirement was at the heart of the initial ARPANet. Using file transfer and remote login techniques, it would be possible for someone at one location to use a computer at another location as if they were physically present. Because of the very large research productivity gains that this mode of operation promised, mathematicians and computer specialists became very excited about achieving this goal, and the initial ARPANet results spurred them on. Courant was one of the leaders in this field, and the CMCL staff contributed substantially to parts of the ARPANet protocols and applications in the 1970s. In 1981, with the spread of university computing opportunities outside of the sciences, Courant was asked to take over the academic computing support responsibility for all of New York University, which it did. At that time, networking was still experimental. The mode of development seemed to be towards disciplinary networks such as CSNET for computer sciences, HEPNET for high energy physicists, and BITNET for mail and file transfer. This was possibly encouraged by ARPA which, although it had the closest thing to a general purpose network at that time, restricted its use to those working on government and military activities. However, one should note that the then management of ARPANet
George Sadowsky
39
had no concept of how large their network would become, since the pre-TCP/IP protocol set, NCP, had address space for addressing at most 256 networks of computers. In 1983, the ARPANet made a very large and significant transition from its earlier protocol set, NCP, to TCP/IP. It was thought that the theoretical capacity of TCP/IP to address 4 billion hosts would suffice for the indefinite future, exemplifying the sense of the time that the Internet would likely remain a relatively specialized network. There were other competing networking efforts at the time. UUCP (Unix-to-Unix copy Program) was used across dial-up links to implement a store and forward network for delivering mail files. Similarly, FidoNet was a network created by a grass roots effort both in the K-12 community and by NGOs operating in developing countries to develop a cooperative volunteer store and forward network using dial-up telephone circuits.
Technological Backdrop Those of us who work in information technology are the beneficiaries of technological progress unmatched in hardly any other industry. Since 1955, the performance-to-price index for information technology hardware in general has been in the range of 25-30% per year. The same measure for the telecommunications industry has historically been considerably lower but in recent years, however as this technology has depended to an increasing extent upon the semiconductor industry and digital communication using optical fiber media, technological progress has accelerated sharply. Nor is there an end in sight, at least for the next 10 years which is generally the limit of product development vision. The cumulative effect of this rate of progress can be measured quantitatively, but it is felt qualitatively, as entire major areas of applications move from being infeasible to being feasible. Markets spring up without much warning, and the demographics of capitalism accelerate, with firms being born and dying with great rapidity. We have appropriately chosen to call this phenomenon Internet time, and it determines the speed of our clock.
Awakening of the Academic and Research Community In the early 1980s, the National Science Foundation (NSF), an executive agency within the U.S. government, was increasingly the target of reports that U.S. scientists were falling behind other countries because of the lower levels of investment in supercomputing in the U.S. In highly publicized testimony, the astrophysicist Larry Smarr reported that he had to travel to Germany to do his research effectively. The NSF has a mandate to encourage and support research in the sciences, and it works through branches corresponding to subject matter areas that fund programs and projects.
40
The Evolution ofNYU-NET in the Context of Regional and National Networking
NSF responded to these complaints in the mid 1980's by funding the establishment of 5 supercomputer centers, and added a component for funding regional networks that would connect to an NSF national backbone for the purpose of allowing researchers to access the supercomputer centers through a network rather than having to travel to the center itself. In retrospect, it is ironic that since that investment decision the relative importance of supercomputers has diminished while the relative importance of data networks has grown enormously. This three-tier model of network expansion — campus networks, regional networks, and the national backbone, formed a solid base for the expansion of academic networking. Starting with a backbone of 64 Kbps on the backbone, capacity increased in 1988 using fractional Tl circuits, in 1991 using T3 circuits, and finally became commercial in 1995 at ever increasing bandwidths. We no longer speak of a backbone because the topology has become much more complex in the five years since that time. The mid 1980's were a time of significant excitement in the academic and research community. The rapid development of mini and microcomputing technology yielded the concept of the 3M machine, a computer that had 1 megabyte of primary memory, could execute 1 million instructions per second, and had a 1 megapixel display unit. This conceptualization of the scholar's workstation was thought to be an ideal computing environment for much of the work of that time. In addition, the deployment of broadband Ethernet and the expansion of the ARPANet presaged network connections ot powerful external services. NYSERNet (New York State Research and Education Network) was the first regional network to establish connectivity in its region. NYSERNet was established in 1986 to "advance science and education in New York State by means of high speed telecommunications techniques, and to assist... in gaining access to ... resources outside of New York State by such means. " In addition, it connected the one supercomputer center within New York State, at Cornell University, to the net. NYSERNet is a not-for-profit organization, and its Board of Directors consists of representatives of the major institutions which it connects." New York state is one of 50 states, containing about 20-25 million people, most of whom live in an L-shaped region connecting the New York city metropolitan area, Albany to the north, and Buffalo to the west. NYSERNet was founded by the leading research universities of New York State. On the one hand, an organization like NYSERNet had to happen; in no way could individual universities have afforded the cost of individually connecting to the national network. On the other hand, it was a happy coincidence, since it laid the groundwork for the formation of personal and institutional relationships as well as cooperative programs that have served the organization and its members very well during its existence.
George Sadowsky
41
First Generation Regional and Campus Networks First generation regional networks were characterized by low bandwidth, few subscribers, and experimental learning. In the case of NYSERNet, an initial experiment to contract the network to the regional telephone service provider yielded unsatisfactory results, and NYSERNet decided to build its own network, using circuits leased from the company. The network infrastructure consisted of Tl circuits. In 1990, the network design and operations team wanted to enter the then emerging commercial IP industry. An amicable settlement was reached whereby the team, led by Bill Schrader, would take ownership of the physical networking assets and would leave with the staff to form a new commercial company, Performance Systems International, Inc., and would offer a commercial service using PSINET. NYSERNet would contract with PSI for IP service for a five year period, and in return, NYSERNet would be given a portion of the common stock of PSI. At about the same time, other companies were making similar decisions. Alternet, under the leadership of Rick Adams, founded UUNet, and established companies such as MCI, BBN and Sprint were beginning to enter the field. Advanced Networks and Services (ANS) was formed in 1990 to manage the second generation NSFNet backbone, using IBM RS/6000 minicomputers as routers for T3 circuits. During the next few years, the Internet prospered, especially with the introduction and spread of the World Wide Web, and infrastructures such as PSI benefited from that activity. ANS did especially well; after the next evolutionary step of NSFNet - Network Access Points for peering and the commercialization of the backbone - they sold their network to America OnLine (AOL) and became a charitable foundation. Networking was evolving at NYU during this time. Since NYU has been an early entrant into networking, it had deployed early technology. In 1985, as a result of the GreeneAT&T decision, NYU installed its own internal telephone system, and at the same time, laid a broadband coaxial cable connecting all main buildings for both television and data communication. Several thousand nodes were connected into the 1990's through buffered repeaters and low cost terminal control units. (This is very similar to the same technology that is used now in home cable Internet connections.) The shortcomings of coaxial cable transport soon became apparent. Reliable connectivity depended upon high frequency analog carriers, with active amplifiers throughout. Tuning the entire network was a major problem. Second, the entire network was bridged. Occasional broadcast storms were frustrating and their sources were sometimes difficult to locate. Occasionally someone would plug a television source into an Ethernet jack and the network would be rendered unusable. The system used two unidirectional 5 megahertz channels, and there would be occasional interference when adjacent analog channels were used. For these reasons as well as increasing utilization, in 1993 a decision was made to
42
The Evolution ofNYU-NET in the Context of Regional and National Networking
migrate to NYU-NET-2, a fiber based, routed network centered initially around a DEC Gigaswitch, using FDDI as the major backbone transport with Ethernet tail circuits. The NYU-NET-2 undertaking started in 1993-94 was planned for 2-3 years, but in fact has lasted much longer. A good part of this result was that previously we perceived network implementation and upgrading as a series of discrete steps, separated widely in time. Such a concept was born in the earlier period of experimental networking, which was project based and related to the funding cycles of government agencies. With NYU-NET-2, we learned that network evolution was moving from a step function to a more continuous process as large parts of campuses wanted to become connected to the net.
Second Generation Networking: After PSI and NYU-NET-1 Near the end of the five year contract with PSI, Inc., NYSERNet began to outgrow the capacity provided by PSI. In addition, NYSERNet management as well as university representatives had become familiar with some of the disadvantages of not having control of their network which they had enjoyed prior to 1990. A decision was made to invest in its own network again. The decision was made possible by its investment in PSI. Before that time PSI had launched an initial public offering (IPO) and its stock was commanding respectable value. NYSERNet therefore liquidated its position at a significant profit, and used a part of the funds to invest in new network infrastructure which, in cooperation with Sprint and Verizon (then NYNEX) provided a statewide T3 network, managed by external contractors but controlled by NYSERNet. At about the same time, it was the judgment of NYSERNet management that in an era of rapid commercialization of Internet services, regional academic and research networks had a limited future. A decision had to be made between static continuation, which was feared would lead to acquisition or death, or commercialization. A decision was made to go the route of commercialization again, and from that decision in 1995-96 was born AppliedTheory Corporation. The scenario repeated itself; much of AppliedTheory was owned by NYSERNet, the physical assets were transferred to AppliedTheory, and a five year service contract was concluded. AppliedTheory inherited the existing relationships with Sprint, which provided the WAN connectivity and overall network management within New York State, and with NYNEX, which provided the local loops.
The Internet Explosion (1995-1999) Although the growth rate of the Internet has been consistently high from its beginning, the impact of this growth, coupled with public awareness, came in the mid-1990's.
George Sadowsky
43
The applications explosion provided much of the fuel. Applications such as Gopher, which provided an entirely new and powerful way to organize and access information hierarchically and associatively, were smothered a year later by the emergence of the World Wide Web and Mosaic based upon the initial HTML version 1.0. Since then the Web has expanded even more, with Netscape, Internet explorer, new versions of HTML, XML, Java, back end servers, Perl scripts, applets and servlets. During this period of time, there were substantial challenges to campus networking. At NYU, infrastructure expense levels were beginning to be quite visible in computing budgets. Bandwidths that had seemed very excessive at the beginning of the period were increasingly inadequate as applications evolved, new users joined the net, and applications became more bandwidth intensive. User expectations were shifting; instead of regarding the network as an experimental entity as it was regarded in 1990, greater reliability and availability were expected. The network began to be exploited for business applications, which heightened expectations. At the same time, the net began to be an attractive target for hackers of the worse kind, so that security issues concerning the net not only were increasingly visible to its users, but had to be dealt with by network security staffs which were established and grew. Universities, having a large population of students who were going through an experimental period in their lives, suffered more than most. Misbehavior, and how to deal with it in an environment that championed free expression, became a major issue.
Internet - 2 NYSERNet was quite right in forecasting their future as commercial growth and stagnant demise. During these years every regional network in the United States was either acquired by a commercial ISP or went out of business in some other way. The academic community realized that it had lost control of the Internet and were now at the mercy of commodity Internet providers. With increasing congestion and no remedy in sight, the Internet-2 project was started. Internet-2 was to be a separate network, technically stable, offering high bandwidth and different qualities of service for experimental applications. The quality of service dimension (QoS) was especially important, since it held the key to rational pricing, investment, and rationing of the resource. Internet-2 has developed substantially, and has added a substantial amount of capacity for academic use. Experimental applications are being developed, although at a disappointing rate. More disappointing, the use of the network is not being restricted to experimental applications, but is being used for all traffic between participating institutions. In theory there is a more restrictive acceptable use policy but it is not enforced. In addition, work on QoS has been more difficult than what was foreseen, and progress has been slow.
44
The Evolution ofNYU-NET in the Context of Regional and National Networking
Finally, the ability to access Internet-2 on an end to end basis is causing many campuses to have to make relatively expensive additions and changes to their campus networks. Internet-2 topology is similar to that supported by NSF 15 years ago. There are regional networks, now called regional aggregation points, or gigapops, that are connected in turn to each other. Coalitions of geographically proximate members are charged with the responsibility of creating and operating the gigapop and connecting themselves to it. NYSERNet participates in Internet-2 through an unusual gigapop that is 500 miles long and several optical fibers thick, which uses the right of way of the New York State Thruway, a limited access highway that connects New York City to Albany to Buffalo. The gigapop operates at OC-12 speed, and has connections at both ends to Internet-2 backbone provider points of presence. This architecture would not have been possible without using the capital gains realized on PSI stock in 1995, which provided a moderate size endowment for NYSERNet. This research network, called NYSERNet 2000 and sponsored by NYSERNet, is in addition to NYSERNet's connections to the commodity network. Within NYSERNet, we have seen a slow move toward meaningful applications on NYSERNet 2000. Perhaps the majority of them so far have had to do with geographically distributed artistic synchronized performances. In my opinion, many of the applications on a national basis have demanded low latency to be successful, with high bandwidth being a remote second attribute of choice. To the extent that this continues, quality of service research and implementation offers a great deal of hope, even over networks that may be somewhat congested at the present time. NYU's reaction to Internet-2 is, I think moderately typical. We have benefited from the increased bandwidth, and scientists who share large data files with their colleagues have done well. On the other had, new applications have been slow to emerge, which leads to the question of whether there is a "killer application" in Internet-2 space. One would think that some form of enhanced desktop videoconferencing would be forthcoming and would claim such a title, but this has not happened, at least not yet.
Third Generation NYSERNet Networking NYSERNet continues to provide commodity Internet service to its members through AppliedTheory. From single T3 lines spanning the State of New York in 1996, AppliedTheory has expanded to an OC-12 network spanning much of the eastern part of the United States, and also including California and Washington State. NYSERNet's member institutions are upgrading their connections from T3 to OC-3 to meet local demand. NYSERNet, committed to pushing the networking frontier in New York State, is assisting the early adopters to make such moves by offering operating subsidies from its endowment.
George Sadowsky
45
The Future NYU-NET's future is easy to see in the large, but not in the small. We anticipate that demand will continue to grow at the same rate. With all student residence hall rooms now networked and network participation rates at 60% and climbing, growth will continue. A serious challenge that student computing presents is how to provide available bandwidth for academic use in the face of recreational networking activities such as Napster and its relatives that consume enormous amounts of bandwidth. Increased reliability and availability will be critical. Telephony engineers speak of five-9's reliability, i.e. 99.999% uptime. We are not there yet. Spread spectrum wireless technology is invading Internet space just as most of our buildings have become wired. What are the benefits of providing mobile computing, and what are the costs and the opportunity costs of doing so? What are the residual security issues and risks in implementing the current state of IEEE 802.11 based wireless technology? How quickly will we have to, or want to, make the transition to Ipv6, in light of using up the Ipv4 address space? How should regard voice over IP (VoIP) technology at this point in time? To what extent does the connecting of other non-IP hand held devices to the local network contribute to the academic mission or the business operations of the institution? There are many more questions than answers at this time.
Policy Issues are Important Universities and similar institutions differ from Internet service providers (ISPs). Whereas an account with an ISP generally enables the subscriber to use the services provided for anything legal, academic and research institutions - and the networks that they manage - furnish network services to their various constituencies in furtherance of their mission. Earlier in academic networking the U.S. National Science Foundation formulated an acceptable use policy (AUP) for all users of its network that stated essentially that the network was not to be used for commercial or other non-mission related activities. Such AUPs are a useful rationing device in a field characterized by high and rising costs. Similarly, institutions need to ensure that all connected institutions are paying their own way. If downstreaming of connections is permitted i.e. a participant attached to the network through one or more participants, then it is not unreasonable to establish rules that ensure that downstreamed participants pay some share of the cost of maintaining the network. Alternatively, network connections could be engineered and priced assuming that all bandwidth would be in constant use, and then downstreaming policy could be left to the institutions connected to the network. This could lead to substantially higher costs overall.
46
The Evolution ofNYU-NET in the Context of Regional and National Networking
Policies regarding privacy, security, and appropriate behavior can be a significant issue in universities. By the nature of the institution, networking has some experimental component in higher education. The difficulty comes in defining what behavior is appropriate and acceptable, and what behavior is inappropriate and unacceptable. One strategy is to lay down a long list of rules; this generally generates a competitive reaction in students who then work to see what unacceptable modes of behavior still fall within the rules. Another approach is to provide general principles and illustrative cases. Regardless of the approach chosen, enforcement will be needed, and education may be one of the best tools for minimizing the need for enforcement. Issues of content are always present. Censorship of content goes against most democracies, yet allowing all content regardless of network effect may cause inadvertent denial of service for others. The spread of Napster earlier this year illustrated this tradeoff in a dramatic manner. The tensions between freedom of expression, content choices, privacy, and institutional mission are sometimes not easy to reconcile.
Economic and Financial Issues Early wide area networking has been substantially subsidized in many countries, at multiple levels. Early use was experimental, limited in scope, ad not a part of the essential operations of an institution. Costs to the institution were limited and often covered by research grant and contract funds. Most networking is no longer experimental, and the costs of providing what is becoming a new and essential infrastructure to an entire institution is very high, the more so because the rapid technical advances underlying the infrastructure imply a short replacement cycle or substantial opportunity cost. These costs have grown to the point where central administrations often lean toward the side of recovering them by direct user charges. This is easy to do on a very approximate basis, but still difficult to do in a more exact manner. At NYU for example, the cost of just our Tl Internet connection from 1990 to 1995 was approximately $25,000 per year, which was very modestly subsidized. From 1995 to 1999, the cost of just our T3 Internet connection was about $105,000 per year. This cost was more heavily subsidized, since the equivalent commercial price of such service was between $250,000 and $300,000 per year. Finally, starting at the end of year 2000, the cost of our OC-3 connection to the commodity Internet will be about $250,000 per year, less than half of the commercial cost of such a connection. Many other universities are paying for their service at or close to commercial rates. Within NYU, historically we have set a zero marginal price for the use of the existing network, but have charged offices schools and departments the cost of the equipment to
George Sadowsky
47
build, extend, modify, and upgrade the network. The labor for this work has been paid for by central funds. This policy was appropriate for an experimental network. Usage pricing is appealing, but difficult to accomplish. With appropriate records and accounting and data base programs it is possible to charge for an IP address, or for use of a network jack. However, it is extremely costly to charge by volume of transmission, i.e. by the packet. One can approach this result by sampling, with the resulting inequity being determined by the sampling error of the process implemented. One can price also by specific services rendered, such as installations, deinstalls, and trouble shooting calls From an economist's point of view, such a system is in equilibrium if the aggregate charges collected from all sources (including possibly government grants and other sources from outside the institution) are sufficient to pay for the cost of operating the network and providing the network services, and in addition, to pay for the cost of the renewals, upgrades, and network expansions needed to keep the services current when measured against peer institutions. While there are not a large number of choices, the choice is not trivial, and may well change as the technology evolves. Different institutions will come to different conclusions based upon specific circumstances and practices within their institution and their country.
Smart Policy Decisions or Luck? Both NYU-NET and NYSERNet are examples of successful activities. They deliver services with reasonable reliability, and they have evolved in an affordable manner that has generally met almost everyone's needs. To obtain a better performance in either case would have taken considerably greater resources that were not available. I believe that there were significant decisions taken that helped both of these projects to be successful. For NYU, it was smart policy to get involved with networking early and to continue to be involved in the forefront of networking as much as possible. This policy, although probably never enunciated formally as such, allowed the development of a core technical staff experienced in networking that allowed us to make investment decisions both with regard to timing and technology - that were in the university's best interests. It allowed us to be more than proportionately influential in steering the NYSERNet direction. Perhaps the most beneficial policy that emerged from our relationship with NYSERNet was the cooperative group and activities that emerged. While collaboration on the first generation regional network was necessary because of financial considerations, what emerged was a group that cooperated on a much broader basis, sharing resources, experience, software, and support. Such cooperation worked toward the interests of almost all members of the group, and allowed us as a group to get more and do more with less resources, an important consideration in a period of academic recession.
48
The Evolution ofNYU-NET in the Context of Regional and National Networking
On NYSERNet's part, it was smart to recognize that it could do more for its members as a buying consortium than its members could do separately. Under the strong and able leadership of Richard Mandelbaum, NYSERNet acted as a cohesive body with respect to every significant investment decision that was made. It forged close partnerships with private sector firms to push the frontier of networking for the benefit of all of its members. Common interests and the common cultural aspects of the academic and research community were easily identified and assisted in common cooperative work both internally and with external organizations. NYSERNet's relatively unique achievement was to be able to leverage the value that it created by spinning it off into the private sector. Twice NYSERNet reinvented itself by splitting off the major part of the organization and letting it develop and compete in the private sector. The returns to ownership in these companies have been very good so far, and have allowed NYSERNet to emerge as the only remaining regional network O(of the original set commissioned and supported by NSF) and one that has sufficient resources to assist its members, individually and collectively, to stay on the networking frontier. Nevertheless, there were also elements of luck in the paths that were taken. These are best identified in hindsight. First, NYSERNet did not really understand how important the equity positions in PSI and in AppliedTheory were. It was not clear what the two stock issues would do in terms of valuation, but it was not anticipated that they would become so financially important as they have become. Second, there were no negative competitive relationships among NYSERNet members that disrupted good cooperation between them. Third, we had the benefit of strong and knowledgeable leadership from Richard Mandelbaum, one of the most prominent networking pioneers in the United states. And finally, our timing with regard to the strength of information technology stock issues in the equity markets has been fortuitous.
Conclusion One might ask whether the NYU-NET and NYSERNet experience is really too U.S.centric to serve as a model for developing countries. I believe that our experience applies, although not exactly. In support of this, I note that CEENet countries and regions have plenty of market opportunities, although perhaps not as large, as well as opportunities for entrepreneurs. These countries have both local and international partnering opportunities to build networks together and to capitalize upon the value added that they have generated by being early adopters. In most countries there is a strong academic tradition and a recognized professional class. On the other hand, access to the capital markets is clearly not as strong, and there are different legal and regulatory environments that may not be as supportive as needed.
George Sadowsky
49
It is clear now that the growth, use and exploitation of internetworking is of crucial importance to learning and research. Growth and evolution of today's networks is inevitable, and the financial implications of this evolution may be severe for universities, as well as other levels of education. University administrations often do not understand this issue. It is therefore important to capitalize the intellectual value added and obtain returns from it. In this regard, the academic sector can be its own enemy, since it often has difficulty extracting financial value from intellectual achievement. Partnerships and cooperative activities have worked within New York State, and are more likely to lead to identification and exploitation of value that can be capitalized for economic return. With some luck, and with proper partnering, academic and research groups can play their role in this world by experimentation and innovation, activities that historically they have done well at, thereby benefiting multiple sectors of civil society.
This page intentionally left blank
Creative and Innovative Network Management O. B. Popov (Ed.) 1OS Press, 2003
51
Howard Davies
Pan-European Networking Structure for Research and Education
TEN-155 Organisation
Cyprus Israel
16 NRENs + DANTE as Coordinating Partner Policy Committee takes strategic decisions
52
Pan-European Networking Structure for Research and Education
DANTE holds all supply contracts and has responsibility for operational management Funding support from EC via QUANTUM CY, IL connections managed via Q-MED Total cost about 40 MEUR/year Supply contracts terminate in Oct 01 ... . . . but include price review clauses
TEN-155 Structure Based on 34/45/155(7622) Mbit/s SDH ATM layer used for • traffic engineering within the network • separation of European and US traffic • Support of VPNs (including new protocol tests) Services provided • Basic IP • (including interconnection with commercial internet in Europe) • Managed Bandwidth Service
Cyprus Israel
Howard Davies
Geant - Principal Requirements 2.5 Gbit/s "core" as soon as possible Tens of Gbit/s soon afterwards Continuity with Quantum/TEN-155 Include all FP5 countries • integrate CY, IL, PL • connections to BG, EE, LT, LV, RO, SK • funding for CH, CY, CZ, HU, SI Interconnection with research networks in other regions Open procurement Support research traffic • from EC funded projects • access to test beds • including "experimentation with disruptive technologies' Traffic measurement
Infrastructure Issues Technology/performance Availability of capacity Cost of capacity (Increasing) cost of equipment Resilience 24 x 7 operation Interaction with commercial ISPs Equitable cost sharing/funding model Infinite capacity vs. QoS
Network Layers Now
53
54
Pan-European Networking Structure for Research and Education
ATM: Good or Evil? •
•
Merits • Traffic engineering • Traffic separation • for management • for testing • Flexible/optimal use of capacity • Available NOW Disadvantages • Performance (cell) overhead • Added technology layer • complexity • operations/training • cost • Not available (?) at high bandwidths
Conclusions ATM has been useful, but will be superseded; Geant will take us into the optical era
The Energis network
Howard Davies
European Capacity http://www.gtsgroup.com http://www.flagtelecom.com/index_e 1 .htm http://www.kpnqwest.com/pressroom/eurorings-atlantic.jpg http://www.level3.co.uk/uk/network/networkplan http://www.francetelecom.com/vfrance/actualite/cornmdosp/ebackbone.htm http://www.energis.co.uk/groupframe.htm http://www.alcatel.com/telecom/snd/refs/cibles/maps/med.pdf
Transatlantic Capacity for Fixed Cost 1000.0
Resilience NREN and end-user dependence on service 24 x 7 x 365 availability becoming essential Ideally, no single point of failure • equipment • environment • international circuits • local loops Availability of support staff, access to equipment Trade-off between cost and duplication Geant aims for dual links to all countries
55
56
Pan-European Networking Structure for Research and Education
Interaction with Commercial ISPs • Cost/charging issues • Need for (additional) income • Danger of mixing research and commercial activities • Distortion of competition if government funds are used to subsidize services • Better relation with suppliers if NRENs are not seen as commercial competitors • Solution: careful separation of R&E and commercial activities
Cost Sharing •
Factors • relative weight of different access capacities • geographical variation in circuit costs • traffic imbalance • resilience/exposure to loss of service • TEN-155 Model • access factor • core charge to cover equipment, management, core circuits • access charge to cover geographical cost variation • reduction for singly linked NRENs • 60/40 core/access ratio set to give "sensible" result
Infinite Capacity vs. QoS Infinite Capacity • • •
Capacity available is rising rapidly (Geant = 4 x T155) Performance is adequate Simple; restrict complexity to those domains that justify it QoS
• Demand is also rising (x2.5 per year) • New applications (HDTV = 40 Mbps streams) • Traffic expands to fill capacity available • New applications will demand high performance • End-end service must be managed across all domains
Howard Davies
Geant Status Proposal submitted to EC, 31 May Positive evaluation (with comments), 7 June Evaluators comments being addressed EC/NREN contract negotiation to be completed
In parallel, Procurement process formally started, 5 May Invitation to Tender planned for early July
Geant: Future Activities Tender will determine • Initial configuration of 2.5 Gbit/s circuits • Possibilities for connection of the new countries • Timescale for further additions • Timescale for upgrade to >10 Gbit/s • New transmission/switching technologies used Work required on implementation of • IP QoS • IP VPN • transition to IPv6
57
This page intentionally left blank
Creative and Innovative Network Management O. B. Popov (Ed.) IOS Press, 2003
59
Jan H. Guettler
Telecommunications Regulations in the European Union A Moving Target
EU regulatory regimes
J
Full liberalisation 2
Directives 1999 Review
2
Full liberalisation
Directives 1992 Review
I
Partial liberalisation
Directives Green Paper
1986
1992
1998 1999
2002
60
Telecommunications Regulations in the European Union
Type approvals The Mutual Recognition Directive (1991) • • •
European equipment standards Notified bodies / Certified laboratories Mutual recognition
Approach unsuccessful, because • approval regime quite burdensome • European standards not possible for legacy networks • thus, national regimes remained in place
Type approvals The RTTEDirective (1999) •
• • •
"Standards" • Operator network interface specifications • Voluntary equipment standards Equipment declared to work with given network interfaces Manufacturers declarations National authorities • Notified body • Development of test specifications for radio equipment • Control of technical file • Surveillance authority (ex post)
The 1998 regime (1) •
Liberalisation directives • All services open to competition • Subject to competition law • Mergers and concentrations • Abuse of dominant position
The 1998 regime (2) •
Harmonisation directives • Independent national regulator • Transparency • Appeals
JanH. Guettler •
•
Licensing • Open and transparent procedures • Maximum set of conditions Significant Market Power regulations • Delivery obligations • Interconnection obligations • Unbundling/non-discrimination/cost orientation
The 1998 regime (3) •
Harmonisation directives (cont.) 4. Universal service • Fixed telephony at reasonable cost • Data protection • Against misuse of personal data
Implementation issues 1998 Regulatory Package Fair competition • Interconnection • Tariffs • Interconnection rights • Number portability • Carrier selection and pre-selection • Equal access • Timing Harmonisation of European regulations
Interconnection Tariffs • •
Call termination / origination European Commission Recommendation • Long Run Average Incremental Cost • Benchmark figures / best practices
61
62
Telecommunications Regulations in the European Union
Interconnection Facility based competition vs. Service provision Facility based competition: Operator has its own infrastructure Service provision: Operator leases transmission capacity from other operator Basic issue: Which type of organisation is entitled to interconnection tariffs (wholesale tariffs) Underlying question: Which policy is best suited to promote investments and competition
Interconnection rights The Interconnection Directive permits differential treatment: Infrastructure owners vs service providers Minimum set vs. Extended set of categories Some countries: More favourable terms for infrastructure owners: Belgium, France, Spain Most countries: Same terms and conditions for all operators licensed for public voice telephony Germany, Greece, Ireland, Italy, Portugal, Spain, Some countries: Interconnection rights for all operators of public telecommunications (including Internet) Denmark, Finland, the Netherlands, Norway, Sweden, UK • •
Number portability Issues: Who is obliged Timeliness
Number portability Must be provided by: All fixed operators: Austria, Denmark, Finland, France, Germany, Greece, Ireland, The Netherlands, Norway, Spain, Sweden, UK
Jan H. Guettler Incumbent: Belgium
Number portability Implementation target 1.1.00 for fixed regional portability: Early: Denmark (1.7.99), Finland (97/98), France (1.1.98), Germany (1.1.98), the Netherlands (1.1.99), Norway (1.7.99), Sweden (1.7.99), UK (97) On target: Austria, Belgium, Italy, Spain Late: Greece, (1.1.03), Ireland (1.7.00), Portugal (1.1.02)
Number portability Extensions beyond portability for fixed telephony within same geographic region: Examples: Denmark: will be extended to across-region and across-service Finland: across-region and across-service France: will be extended to across-region (on-demand) Germany: applies to mobile, but obligation is currently suspended Norway: no geographic regions Sweden: all services requiring numbers in the national numbering plan UK: wide range of services
Carrier selection / pre-selection
• •
Issues: Timely introduction Options
63
64
Telecommunications Regulations in the European Union
Carrier selection Must be provided by: All authorised voice operators to all authorised voice operators: Austria, Finland, Germany, Italy, Norway, Sweden, Switzerland SMP operators to licensed voice operators: Belgium, France, Greece, Ireland, Portugal, Spain, UK By SMP operators to all operators: Denmark, the Netherlands
Carrier pre-selection Must be provided by: All operators: Austria, Germany, Norway, Sweden, Switzerland SMP operators: Belgium, Denmark, Finland, France, Greece, the Netherlands, Spain Incumbent: Ireland, Italy, Portugal, UK
Carrier pre-selection Timeliness: Early: Denmark (1.1.99), Finland (1.10.98), Germany (1.1.98), Norway (1.6.99), Spain (99), Sweden(l 1.9.99), Switzerland (1.4.99) On deadline 1.1.2000: Austria, Belgium, France, Ireland, Italy, the Netherlands Late: Greece (1.1.03), Portugal (1.7.00), UK (1.4.00)
Carrier selection and pre-selection for mobile operators No obligation: Austria, Belgium, France, Ireland, the Netherlands Both selection and pre-selection: Denmark (not yet implemented), Germany (not yet implemented) Norway (implemented for selection), Spain (for international calls) Switzerland (selection implemented for international calls)
Jan H. Guettler
65
Pre-selection: Sweden (not yet implemented - selection may be imposed later) Selection: Finland (for international calls), Portugal (for international calls) UK (for BT Cellnet and Vodaphone, but not yet implemented)
Carrier selection What services can be provided by selected operator (in addition to fixed long distance and international calls) Local calls, mobile calls, special services: Austria (some special service restrictions), Denmark, Ireland, Norway, Sweden, UK Mobile calls, special services: Belgium, Spain Local calls, mobile calls: Italy, Netherlands (local calls priced as long distance) Mobile calls: France (from 1.7.00), Germany None: Finland
Carrier selection Local calls
66
Telecommunications Regulations in the European Union
Carrier selection Long distance calls
Carrier selection International calls
Jan H. Guettler
67
Transmission Fibre capacity 1999: 6400Gbit/s - (Nortel at Geneva) Speech requirement: 64Kbit/s / 9.6Kbit/s 6400Gbit/s : lOKbit/s = 640 million conversations - (1280 million people) 6400Gbit/s : 2Mbit/s = 3 . 2 million video channels
Regulatory consequences Competition in basic telephony services on the basis of carrier selection (or preselection) will become increasingly difficult and in the end impossible because it is competition based on usage charges for telephony Competition in basic telephony will therefore increasingly be based on access competition with a operator/customer relationship sustained by periodic subscription fees.
Local loop unbundling
68
Telecommunications Regulations in the European Union
Local loop unbundling Status Physical access: Austria (€12.35), Denmark (€9.24), Germany (€13), Italy (not yet implemented), The Netherlands (€10.19), Spain (currently Bitstream), UK (1.7.01) Physical access and Bitstream: Finland (€10-15) Discussions: France, Ireland, Norway, Sweden No plans: Belgium (CableTV), Greece, Portugal (to be reviewed in 2000)
Study on Potential value of European Regulatory Authority • •
•
Part of 1999 Review Major survey conducted • Little support in general for ERA • Support for ERA associated with level of discontent • Lack of harmonisation identified as a major issue Recommendations: • No new European regulatory institution • Strengthen current institutional means • More use of guidelines/recommendations/codes of conduct • Telecom Committee with harmonisation responisbility
1999 Review - Regulatory proposals (1) Regulatory Framework • • • • • •
NRAs independence / tasks /duties / obligations Procedures for harmonisation / dispute resolution Telecommunications/Broadcasting network convergence • (but not Broadcasting content regulation) Management of scarce resources (frequencies / numbers) • Both auctions and administrative procedures likely to be permitted Rolling back of Significant Market Power regulations Institutional arrangements
Jan H. Guettler • •
69
Communications Committee High level communications group
1999 Review - Regulatory proposals (2) Authorisations •
General authorisations for all networks and services • Individual authorisations only for assignment of frequencies and numbers • Notification with maximum: • Declaration of intent to start service • Co-ordinates of the undertaking • Short service description • Frequencies: • Maximum 6 weeks (extendable to 6 months for competitive bids) • Obligations to grant frequencies and to implement CEPT agreements • Numbers: • Maximum 2 weeks • Confirmation of interconnection rights and rights-of-way: Maximum 1 week • Authorisation fees to be linked to turnover and to be limited to adm.costs
1999 Review - Regulatory proposals (3) Interconnection • • • •
•
Roll-back of Significant Market Power regulation Requirement for market analysis with regard to dominance Markets defined for Interconnection/Network Access/Digital TV Where dominance exists: • Broad powers for NRAs to select conditions from a maximum list • (Virtual Mobile Networks not obligatory, but may be demanded by NRA) Local loop unbundling obligatory (separate Directive)
1999 Review - Regulatory proposals (4) Universal Service, etc. •
Universal service scope • Remains as before (fixed telephony) • Procedure for re-evaluation
70
Telecommunications Regulations in the European Union
• •
• • •
Designation of USO operators • Could include tenders or auctions for part or all of the obligation USO cost • Assessment of burden according to guidelines • Compensation through general taxation or transparent sharing arrangement Price regulation • For SMP in public telephony: Price caps/floors to ensure comm. levels Number Portability • Extended to mobile operators Internet access to schools at cost oriented tariffs
1999 Review - Regulatory proposals (5) Protection of personal data and privacy •
• • • • •
Technology neutral: "Traffic data" : "traffic data" shall mean any data processed in the course of or for the purpose of the transmission of a communication over an electronic communications network Enabling of value added services Safeguards on location data (positioning a mobile user) Right not to be listed in directory Protection against unsolicited marketing communications Privacy compliant software and hardware (Internet) • Threath to develop mesures unless industry brings forward solutions
1999 Review - main directions •
• • • • •
Lighter authorisation procedures • Individual authorisation only for numbers and frequencies • For the rest "post card" registration "Significant Market Power" to be harmonised with "Dominance" • 50% market share rather than 25% Local loop unbundling requirement More flexibility to NRAs to impose interconnection obligations, but at the same time stronger European co-ordination Procedures to review scope of universal service Mobile operators to offer number portability
Creative and Innovative Network Management O. B. Popov (Ed.) IOS Press, 2003
11
Fay Howard & Kevin Meynell
Benefits of ccTLD Collaboration and Cooperation
Overview of CENTR
CENTR is an association of Internet Country Code Top-Level Domain Registries such as .uk in the United Kingdom and .es in Spain. Full Membership is open to organisations managing an ISO 3166-1 country code top-level domain (ccTLD) registry, and there is also a category of Associate Membership for organisations managing other TLDs. There are currently 34 Full Members and 3 Associate Members in CENTR who collectively register over 6 million names. This represents around two-thirds of the total domain names registered by ccTLDs worldwide. The CENTR Project started under the auspices of the former RIPE TLD Working Group in March 1998. This was in response to the US Government Green Paper which did not recognise the position of the ccTLD registries, and was informally funded by contributions from participant registries. The following year, it was legally established as a 'not-forprofit' company in the UK. CENTR is now funded by membership fees based on self-selecting bands (see http:/ /www.centr.org/joining/) which are currently between EUR 1,200 and EUR 36,000. There is a small secretariat (currently 3 persons) co-located with Nominet (.uk) in Oxford, UK which allows operating costs to be kept minimal. The organisation originally had a European focus, but there are no geographical restrictions to membership. Indeed, CENTR currently has several members from outside the European Region.
72
Benefits ofccTLD Collaboration and Cooperation
CENTR provides a forum to discuss matters of policy affecting ccTLD registries and acts as a channel of communication to Internet governing bodies and other organisations involved with the Internet. It promotes the interests of ccTLDs and lobbies on their behalf, provides a focal point for enquiries about ccTLDs, and documents the different practices of the registries. It also aims to encourage harmonisation and promote better service for users. As a result, CENTR is facilitating a number of collaborative activities on political, legal and technical issues. One of the most important activities is the development of ccTLD Best Practice to augment RFC 1591 which defines the role and responsibilities of ccTLD Managers. There is also an ongoing activity to establish a contract for services between ICANN and ccTLDs. Other work includes consideration of dispute resolution mechanisms (e.g. WIPO ADRP), involvement in the consultation process for the proposed .eu TLD, investigating how data protection applies to ccTLDs, and how the single market affects ccTLDs in EU/EEA and EU candidate nations. CENTR is also working on a number of technical issues that include secondary server arrangements, DNSSEC, whois, internationalised domain names, and registry-registrar protocols. Finally, CENTR is collaborating with RIPE to migrate domain name information out of the RIPE Database as this is having a detrimental impact on performance.
ccTLD Constituency CENTR participates in the ccTLD Constituency of the ICANN Domain Name Supporting Organisation (DNSO) as one of the recognised regional TLD associations; the others being AFTLD (Africa), APTLD (Asia-Pacific), LACTLD (South America and Caribbean), NATLD (North America) and IATLD (non-aligned). The ccTLD Constituency usually meets during ICANN meetings, although activities are mostly conducted remotely. Like CENTR, the ccTLD Constituency is looking to establish a formal relationship with ICANN and to receive certain services in return for their payments. When ICANN was established, it was envisaged that it would be funded by a 'tax' on domain name registrations. Unfortunately, this means that ccTLDs are collectively expected to contribute 35% of the ICANN budget even though ICANN does not provide ccTLDs with any services (e.g. the root servers which are still controlled by the US Government), nor does the ccTLD Constituency have any direct representation on the ICANN Board of Directors. Furthermore, there is much uncertainty about the accuracy of the domain name counts used to calculate the fee levels, as well as the fact that some ccTLD registries do not charge for registrations. The ccTLD Constituency has consistently rejected the concept of basing fees on number of domain names registered and instead proposes a alternative system of fee bands.
Fay Howard & Kevin Meynell
73
Why Collaborate? There are many advantages to be gained from participating in the ccTLD Constituency and the regional TLD associations. Most ccTLDs are not in competition with each other and are happy to exchange experiences. The boom in the use of the Internet means that many registries are moving away from their academic origins to become independent organisations often answering to ISPs or governments themselves. At the same time, commercialisation has also raised policy and legal issues. Some registries have already experienced the initial transitional problems and are in the position to offer advice to others. Where issues have still to be resolved, the costs of solving these can be shared to reduce the sometimes prohibitive cost to a single registry. ccTLDs are also coming under increasing regulatory pressure from external parties such as governments and international treaty organisations. As a result, registries need to demonstrate they are capable of self-regulation. Individual ccTLDs have a small voice within the Internet community, but collectively their views are much more likely to be taken seriously. This is important at time when the governance structure of Internet governance is being formulated.
Conclusion The relationship between ccTLDs and ICANN is currently under discussion and there is still an opportunity to provide input into the process. The key concerns include establishment of a formal relationship with ICANN, contracts for root services and the I AN A function, and an appropriate ccTLD level of funding. CENTR also wishes to have a review of the democratic processes within the ICANN structure, most specifically with respect to the DNSO. If agreement can be reached on these issues, most ccTLDs are likely to recognise and support ICANN in the future.
Useful Links CENTR ccTLD Constituency ICANN DNSO IANA Government Advisory Committee Proposed EU TLD
http://www.centr.org/ http://www.wwtld.org/ http://www.icann.org/ http://www.dnso.org/ http://www.iana.org/ http://www.noie.gov.au/gac/ http://www.centr.org/eu-links.html
This page intentionally left blank
Creative and Innovative Network Management O. B. Popov (Ed.) IOS Press, 2003
75
Zita Wenzel
ICANN and the CCTLD: Global Issues and their Relevance to the Countries in CEE and FSU
Internet Assigned Numbers Authority • •
IANA = Jon Postel "Dedicated to preserving the central coordinatingfunctions of the global Internet for the public good." • Functions: • Protocol parameter assignment • IP address allocation • Domain Name System (DNS) coordination • Root server system coordination
History of 1C ANN • • • • •
Funding from US Government ending Privatization needed Need to contractualize all relationships Need to provide bottom-up representation Need to provide international representation
76
ICANN and the CCTLD: Global Issues and their Relevance to CEE and FSU
Overview of ICANN
Internet Corporation for Assigned Names and Numbers
Global Issues for CCTLDs • • • • •
Relationship with ICANN Relationship with governments Best Practice document Internationalized domain names CCTLD constituency
CCTLD-ICANN Relationship • • • •
Root server system and function Domain Name System coordination Uniform Dispute Resolution Policy Funding
CCTLDs and Governments • • • •
For the good of the local community Fair and equitable Stable and reliable Able to function internationally
Zita Wenzel
77
Best Practice Document • • • • • •
Technical administrative requirements Redelegation requirements Use RFC 1591 as a starting point See also Internet-Draft-Klensin-1591 GAC document regarding CCTLDs CCTLD working group has draft
Internationalized Domain Names •
• •
To use native language: • Domain names • URLs • Email IETF working group (IDN) requirements Multilingual Internet Names Consortium
CCTLD Constituency • • • •
CCTLD DNSO representatives • Nii Quaynor (.GH), Patricio Poblete (.CL), Dennis Jennings (formerly-.IE) Best Practice document by July meeting CCTLD secretariat (AFNIC/AP proposal) CCTLD mailing list
Attributes of CCTLDs • • • •
Representing local Internet community Technical administrative competence In country-use versus open registration For profit versus not-for-profit
78
ICANN and the CCTLD: Global Issues and their Relevance to CEE and FSU
Specific Issues for CEE and FSU •
• • •
Regional representation • Current groups are Asia-Pacific, Africa, Western Europe, Latin America, North America Differences in governments and economies Differences in languages and cultures • Cyrillic MoU with Georgia, Azerbaijan, Armenia • CEETLD? FSUTLD?
Creative and Innovative Network Management O. B. Popov (Ed.) IOS Press, 2003
79
Eugenie Staicut
".RO" Top Level Domain Registrations
The domain name registration under the .RO top level was initiated by the National Institute for R&D in Informatics in 1992 when the first leased line was established to Vienna University for providing connection to EARN. We had the opportunity to work with very qualified people at Vienna University who advised us from the very beginning to start with TCP/IP protocol and NJE protocol over TCP/IP. So, from the very beginning we had both Internet and EARN connectivity. The nameservers running the .ro ccTLD are: • ns.rnc.ro • ns-a.rnc.ro • nsl.univie.ac.at • ns.eu.net • sunic.sunet.se • ns.austria.eu.net • ns.eu.net • sparky.arl.mil. At the beginning, maybe like in any other country, the activity concerning domain registration was small, only 1-2 registration application per month were received. Gradually, especially during the last two years, the number of registrations increased and therefore the complexity of the work has increased.
80
".RO" Top Level Domain Registrations
The dynamic of the number of domain registrations is presented in the following table. Table 1. Month
Year
Number of .ro domains
Number of hosts
March
1993
2
15
March
1994
17
113
March
1995
28
519
March
1996
236
2,531
March
1997
536
9,491
March
1998
1,224
15,315
March
1999
2,676
27,436
March
2000
4,894
35,844
April
2000
5,532
50,191
In running the domain name registry for the .ro, we were guided by the RFC 1591, but this document presents only the main guidelines. Due to the increase of the use of Internet, especially for commerce, we were faced with problems that did not exist at the beginning, like for example: • •
How to treat registrations for trademarks, for famous and well known names; What should be the domain name dispute resolution?
This was the time when we learned about the CENTR and we joint it. For us, joining the CENTR was very useful since we had the opportunity to learn how the domain registry is organized in other countries and how the problems we faced are being treated. In making the rules for domain registration, we paid attention to minimize the conflicts arising from domain name registration and at the same time to minimize the time for registration. Usually, people want the registration to be done in the same day they sent the application. Within the registration application, we require the applicant to specify that he takes over the responsibility for legal use of the domain name and that does not infringe the rights of any other party.
Eugenie Staicut
81
We also tried to meet the user requirements as much as possible. When we started the domain registrations, we adopted a flat structure, after the consultation with the user community at that time. When the commercial user community increased in size, especially during the last two years, many users asked us why there is no possibility to register under .com.ro, so at present both flat and non-flat domain structure is allowed. Registrations can be made either directly under .ro or under .com.ro, the applicant does the selection. Regarding the payment, at the beginning the registration was free of charge. At present, there is a one-time fee for registration. The payment of the fee is a condition to complete the domain registration. The main problems we would like to pay attention in the near future are the following: • •
•
To improve the dispute resolution policy; To build an automatic system for domain registration. We have already started to do it. Till now, the number of registrations was small, but the number of registrations is increasing, so we need that the registration is done automatically; Until now, we used the RIPE whois server for publishing the registered domains. We have already installed a local whois server and we will no longer use the RIPE whois server in the future.
The participation to the CENTR activity helped us to solve some of our problems and I am sure that the participation will be also fruitful in the future.
This page intentionally left blank
Creative and Innovative Network Management O. B. Popov (Ed.) IOS Press, 2003
Peter Rastl
The Human Factor in ACOnet Obviously NRNs are concerned mostly with technology, but there is more: It's humans who make the NRNs for humans to use them. A few thoughts and experiences from ACOnet
What will make a NRN worthwhile? •
•
Creation phase How to select the right initiative? How to set up a stable organization? Operation phase How to maintain a durable infrastructure? How to cope with changes and developments?
83
84
The Human Factor in ACOnet
Creating a NRN is dealing with humans: •
your peers: creating cooperation How do you deal with competitive initiatives? • your staff: creating expertise What will make you qualified to do the job? • your users: creating services Are you sure you understand the needs of your user community? • your sponsors: creating funding Why should someone pay for your NRN?
Operating a NRN is dealing with humans: •
your users: preserving your services Are you fast enough adjusting to your users' changing needs? • your competitors: preserving your mission How do you behave to the Commercial Internet? • your colleagues: preserving your contacts Do you care for the Internet on an international scale? • your authorities: preserving your position How do you contribute to the benefit of your national Internet?
ACOnet's peers •
Consolidation of competing groups ACONET Association has been founded in 1986 • Decision about ACOnet's mission Service provider vs. research institution • Selection of proper technology Production quality, usability, innovation • Motivation for making efforts Non-profit user orientation vs. ambition, vanity, career, earnings
Peter Rastl
ACOnet's staff •
Small crew integrated in Computer Center NRN services are managed jointly with university services
•
Highly motivated and qualified crew Freedom to work on their own, but keep a spirit of common goals
•
Keep contact with state-of-the-art work Confront the staff with more advanced environments
•
Tackle the problem of enticing away Try to keep your staff even if there are higher salaries elsewhere
ACOnet's users •
ACOnet is a backbone network ACOnet's users are the local network providers
•
Vienna University is an ACOnet user We use the same services ourselves as other users - no privileges
•
Regular contact with all users React to their requests and explain them your problems
•
Accounts are made transparent to users Regular financial reporting, regular adjustment of usage fees
ACOnet's sponsors •
Initially we had no funding from ministry Offer services nevertheless, and educate ministry about networking
•
Cooperation with industry Sponsorship from IBM was instrumental for our start-up
•
Keep neutrality Be sure that you can always talk to future sponsors
•
Secure funding from more than one source Users must understand that your services are worth real money
85
86
The Human Factor in ACOnet
ACOnet's competitors • •
• •
ACOnet is no commercial ISP Service only for research, education and cultural institutions Internet Services Providers Association ACOnet helped to establish ISPA and cooperates in many areas Vienna Internet eXchange (VIX) ACOnet offers neutral exchange services to commercial providers Domain Name Administration ACOnet operates registry, but transferred responsibility to ISPA
ACOnet's colleagues • • •
Neworking is applied global cooperation ACOnet is a member of several international organizations Exchange of experience Most has already been invented elsewhere, but we have to implement it at home ACOnet's early initiatives in CEE This has already been dealt with in the Tbilisi Workshop
ACOnet's authorities • • •
ACOnet is strictly not-for-profit Nobody should be able to impute anything beyond your mission Responsible approach to monopoly services Could users benefit from deregulation of your monopoly services? Credibility through competence ACOnet offers support to government authorities
Always offer others a share of your success!
Creative and Innovative Network Management O. B. Popov (Ed.) IOS Press, 2003
87
Richard Perlman
Bandwidth is a Valuable Resource!
How will you protect your bandwidth investment? Remember Murphy's Law Given the Chance... • Everything that can go wrong, will go wrong
Intentions Don't Always Work out • • •
The automobile was thought to be a means to solve a major pollution problem. Horse Scheisse But, in retrospect, the cure was worse than the disease
The Internet was created to... • • •
Provide a communications network that could survive a nuclear war It was not designed with today's uses and applications in mind We still use SMTP... • ... Simple Mail Transport Protocol
88
Bandwidth is a Valuable Resource
So, what are the dangers to your network • • •
Misdesign • Sorry, you have to solve this one :-) Misappropriation / Attack • Someone is making a specific attempt to disrupt your network performance Misuse • The network is not being used in the way you intended
Where does the real risk lie? •
Outside or Inside your network?
Outside threats include: • • • • •
Crackers Joy Riders Thrill Seekers Taggers But, this is not our topic for today ...
What are the internal threats to your valuable bandwidth •
The biggest risk is your own assumptions • How your users will use your network • What purposes your network will serve • Why you built your network in the first place
Assumptions •
Symmetry of Data flows... • The client server model • Users are clients. • 90% of data flows TO the user • 10% of data flows FROM the user
Richard Perlman
This is NOT TRUE any more • Napster • GNUtella • Fully distributed file systems • Shared memory distributed multi-processing virtual reality multi-person environments... • Don't laugh, there may be one on your network already!
This is NOT TRUE any more •
Napster • At 12:00 GMT (04:00 to 07:00 in USA) • 3,206 nodes (I.e. servers) • 1,390 Gigabytes • 343,751 individual files • Typical files are 2 to 4 Megabytes
Assumptions •
Lots of free quality information • Reality...
89
90
Bandwidth is a Valuable Resource
Loss of Content • • • • •
Percent of usable content dropping Quality of free content dropping It's like having a new car, with no roads to drive it on A corollary to the Peter principle... • Junk will expand to fill available space. Or... • Nature abhors a vacuum (and spamers too)
Assumptions Meet Reality •
The Internet means free access to lots of data. Is this true?
•
Political, religious, cultural limits • Are ideas really safe • How much freedom are you willing to tolerate • You check on how deep the pool is before you dive in!
Assumptions Meet Reality •
French case against Yahoo access to "Nazi" sites • Does a nation (region, school, etc.) have a right to limit what can be. • Served • Accessed • Is it even practical or possible to attempt this level of control?
Assumptions Meet Reality •
• •
Proposed English RIP law • Police can order YOU to give them your password • You cannot tell anyone when that has happened • A business, school can be ordered to monitor communications This is all designed to combat Child Pornography Multi-Byte character sets in domain names
Richard Perlman
91
A nation or ethnic group should not be forced to compress the richness of their "Alphabet" into 7-bit ascii. But, should we be denied access to large parts of the net simply because we cannot enter a domain name?
Historical concept of "Acceptable Use" •
When the Internet was still in non-commercial form there were distinct limitations on its use • Does this situation still exist? • Should it?
Acceptable Use •
Do you have an Acceptable use Policy and Agreement for your network? • Why? • Why not?
•
Some things are clear... • Commercial vs. non-commercial • Legal vs. illegal • But, laws vary by city, province, country... • Too much limitation can stifle creativity and thought • Limitations on valid use will drive users underground
Will Fear keep us from enjoying our promised land? •
If you have to wall off your network to protect it from insiders and outsiders what do you have left?
Freude und Freiheit • •
This region of the world has special associations with the word and concept of "Freedom." What will be our attitude toward the new e-freedom
92
Bandwidth is a Valuable Resource
What can you do to protect your network investment •
Keep it free and open
•
Invest in content • Silicon and wires are fascinating ... ... but, in the information society, content is King. • Do we really put the same effort toward quality of content that we do for the connection?
We are really expanding the human bandwidth • • •
Working together to build regional networks Collaboration CEENET!
Creative and Innovative Network Management O. B. Popov (Ed.) 1OS Press, 2003
93
Jacek Gajewski
CEENet Educational Activities
CEENet (co-)organizes: •
• •
Annual Network Technology Workshops (Warsaw 1995, Budapest 1996, Zagreb 1997, Bratislava 1998, Budapest 1999, Budapest 2000) Biannual Network Policy Workshops (Tartu 1997, Tbilisi 1999) Biannual Network Managerial Methods Workshops (Yaroslavl 1998, Ohrid 2000)
Network Technology Workshops consist of three tracks: • • •
ENGIneering the Networks NETwork KNOWledge Systems WIRed Education
Each workshop is preceded by WebCT distance learning phase. Workshop course material is printed as proceedings and published on the Web: www.ceenet.org/workshops.
94
CEENet Educational Activities
# Students # Lecturers / Tutors # Admin. Staff # Working days
DL - pre course to Network Technology -200 6/9 1 9 weeks 10 h / week
Cost
Workshop (Network Technology)
Policy & Network Managerial Workshops
96 15/3 3 + LOG 9 days 9 h / day 150 k$
-30 -20 3 + LOG 3 days 7 h / day
9 K $ + inkind donations
75 k$
(45 k$)
CEENet supports The development of vendor constructed courses i.e. CISCO Academy (75 k $ OSI grant for travel of trainers)
CEENet aims To establish quasi-permanent training facility at CEU CC in Budapest - working name SONETA (CEENet investment of 30 k $). This facility could and should be used by others!
Creative and Innovative Network Management O. B. Popov (Ed.) IOS Press, 2003
95
Karen Edberg
The Nordic Approach to Distance Education and Learning
The traditional distance education courses are associated with correspondence courses, where the handwritten letter was a central feature. The interaction between learner and teacher was carried out by the exchanging of study material, assignments and response to assignments, delivered by ordinary mail. Face-to-face meetings and telephone guidance were later being used as well. The last decade has seen a rapid change in the delivery of study material and in the interaction between teacher and student via the web, not to speak of the student-to-student interaction in the same medium. The Norwegian Association for Distance Education today defines distance education in the following way: "Distance education is where the teacher and the student(s) are separated both physically and/or in time. Technical means are being used to present learning material and to mediate communication between teacher and students as a help in the learning process" (www.tunet.net). The web offers new possibilities for bridging distances of different kinds but there is a risk that the belief in technology itself as a solution will overlook an important issue. "Information technology does not solve the problem of learning but it changes the conditions for learning." (Sa'ljo 2000?) Learning is still a cumbersome activity that takes place within the individual learner and in the interaction between people engaged in a shared learning process. The use of information technology also affects the individual teacher's work. "One of our teachers sighed and longed for the times when the postman arrived once a day" said a Norwegian networking pioneer, professor Morten Plate Paulsen. He quoted a nostalgic colleague at the electronically managed distance education institute, Nettskolen, at NKI, se below (Ekberg 2000). The quotation illustrates the dramatic change in the conditions for distance education; from the exchange of handwritten letters with a potential for delivery at the very most five or six days a week, compared to the high speed web communication every day around the clock. What we see today is the whole spectrum of technologies being used, side-byside and in combination. How does this affect the providers of education? What happens on
96
The Nordic Approach to Distance Education and Learning
the national level to meet new demands? This article will take a look at some of these conditions and discuss trends related to distance education or rather to the concept of flexible learning in higher education. The focus will be on some of the Nordic countries, namely Norway, Sweden and Finland, thus making no promises to be exhaustive since Denmark and Iceland are missing. The reason for my choosing Norway and Finland, beside my own country, Sweden, is rather personal and not at all scientific: I happened to visit various distance education institutes in Norway last year and with Finland I have special connections to people in the business! It is worth pointing out, though, that there are frequent contacts between all Nordic countries on different levels and many examples of cooperation, such as a Swedish-Norwegian research programme, an evaluation of ICT projects in Norway, Finland, Denmark and Sweden (www.distum.se). The Nordic countries have a long tradition of distance education and of adult education. In Sweden two correspondence institutes, Hermods and Norens Korrespondens Institut, NKI, started in the late 1900th century. NKI was established in Norway as well, and it is now an independent foundation, Norsk Kunskaps Institut (www.nki.no). Since the late 1950-ies and early 1960-ies there are also good opportunities for people who lack secondary education to study as distance students at two state funded centres in Sweden, the National Institute for Distance Education ("Statens skola for vuxna" in Norrkoping and in Harnosand respectively). Since the 1970-ies adults become elegible for univiersity studies at their local adult education center run by the municipality. An extra effort was made in the midnineties to bridge the educational gap among both employed and unemployed people in Sweden when an educational scheme called "Knowledge Raising" (Kunskapslyftet) started. The providers of adult education form a network of distance nodes to promote cooperation. In Norway the competence raising education is run not only by the state but also by private institutes, like NKI and the other well-known provider, Norsk Korrespondance Skole, NKS (www.nks.no ). The state is joining several public institutes into one organisation, Statens ressurs- og voksenopplaeringssenter (SRV). The new organisation includes both an educational publishing company and a research centre. Finland has an Open university which allows adults to qualify for higher education. The Open university is run by several universities, and local study centres, for example folk high schools, that provide support for these students. The students are admitted without being elegible for higher education. This means that the structure of distance education differs between Finland on the one hand and Norway and Sweden on the other. Finland chose to build on existing structures instead of establishing special institutions for distance education. In the 70-ies distance education was introduced into many existing higher educational institutions, among others ten universities and colleges. The distance education activities at Finnish universities are usually organised by the centres for continuing education, which work as an integral part of the conventional universities. Thus the Nordic countries did not establish specialised distance universities like the British Open and Distance University but adopted the "dual mode" model, which is especially true for Sweden and
Karen Edberg
97
Norway. The "dual mode" model means that universities and colleges integrated distance education into their traditional activities. "Concepts like 'decentralised education' and 'extended classes' are more correct in those cases than 'distance education1." (www.distum.se).
Nordic national organisations and networks All Nordic countries have national umbrella organisations for distance education, with similar names: The Norwegian Association for Distance Education (NADE), The Finnish Association for Distance Education (FADE), The Swedish Association for Distance Education (SADE) and so on. The SADE policy may serve as an example: The organisation "is open to individuals and organisation with an interest in distance education and flexible learning" (www.sverd.org). It is also Sweden's representative in international networks for distance education and the same goes for the neigbouring countries' organisations. An organisational difference is that the Finnish and Norwegian organisations are not open to individuals, only to organisations and institutes. NADE has published quality norms for distance education and is managing the webb site called Tunet, www.tunet.net. NORDUnet is the umbrella organisation for the Nordic academic networks like Sunet, Funet, UNINETT (Norway), DAREnet and Isnet. Since these networks are well-known to most readers of this article, I will just mention them here. One of many interesting emerging features supported by the networks are the nationwide library networks that are being created (www.sunet.se, www.funet.fi).
Nordic trends Political trends Like in most Western countries the concept of life long learning is being discussed: "Learning is a lifelong project coinciding with work and leisure time and alternating between them" (Sinikka Larsen 1999). This means for example that people have to consider changing their jobs both once and twice during their working life. It also means having to catch up at their existing jobs. The demands for higher education are becoming more differentiated, with a clear just-in-time aspect. It means short courses at the right time and for people who will have the necessary life/work experience to qualify for the course but not the paper qualifications needed, the so called non-traditional learners. Let's take Norway as an example. In 2000 a competence reform was launched, a state funded project of 50 million NKR (about 6 million dollars). The money was being given to joint projects that would involve a workplace and an education provider. It was not allocated to individuals, but to groups of people within the project. The projects should
98
The Nordic Approach to Distance Education and Learning
focus on new competence at work and thereby apply new technology. People with low education should also be given a chance to raise their competence. This reform obviously is directed to people who are already working, since there is hardly any open unemployment in Norway. More money is made available in another round of projects with a doubled budget for the year 2001 (100 million NKR, see www.tunet.net). The right to get leave of absence for studies has been recognized for many years in the Nordic countries. In Sweden the government is about to produce a bill suggesting an Individual Learning Account, a savings scheme that would allow people to take time off for studies, when needed during working life. There are different subsidies, e.g. reduction in payroll taxes, being suggested to encourage the individual and also the employer to participate and to support the system. The studies do not necessarily have to be connected to their jobs: "The definition of competence development /.../ should be broad and generous" (SOU 2000:119, p. 2). The study allowance system in Sweden does not always fit the individual needs and it is an expensive solution. Another issue that is crucial to life long learning is the validation of competence. There is a Norwegian project investigating a change of rules for admission to higher education in order to make it easier for people get access to relevant courses. Now people may have the required qualifications but they lack the offical paper to prove it (www.realkompetense.org). Similar ideas were expressed by the Swedish Minister of Education in his inauguration speech at the Distum conference (see below). He said that distance education should be shaped in such a way that new groups of students would be attracted to higher education. The regulations of higher education must be changed to meet the needs of the new communication society.
Organisational trends A general trend is that the distance education institutes and the universities are getting closer to each other. A problem for "dual mode" institutions is to finance the development work that distance education requires. A new authority was created in Sweden in 1999, the Swedish Agency for Distance education, Distum (Distansutbildningsmyndigheten,) in order to support development in the field of "flexible education", which is the term used by Distum. The corresponding authority in Norway is SOFF (Sentralorganet for flexibel laering i h0yre utdanning) which is older, formed in 1989. Besides supporting development projects it also forms a platform for cooperation within higher education. The universities has profited by the administrative and pedagogical knowledge that has been built up by the old correspondence institutes, while the universities have contributed with their subjects and their right to issue exams. Both parties have gained but in the long run it is an open question who will profit most from the cooperation (Holmberg 1998).
Karen Edberg
99
Pedagogical trends The risk for ignoring pedagogical issues was mentioned in the introduction of this article. It is therefore of the greatest importance to focus on the these aspects in distance education. The last decades of the 20th century have seen a shift in the pedagogical ideology from teaching to learning. The traditional lectures are being questioned as inefficient ways of delivering information. The focus should be on the needs of the learner and thus the organisation of the courses and the methods have to vary. Computer mediated communication (CMC) or Information and Communication Technology, (ICT) match the shift from a pedagogical point of view. The teacher's interaction with the learner and also the interaction between the learners are important. Different pedagogical models influence each other - the democratic study circle, the distance education and the university traditions. Instead of speaking of distance education the concept of flexible learning is widely used. My university, Mid Sweden university, sums up the concept of flexible learning in the following three statements: • Education which focusses on student learning • Flexibility in time and place and in teaching and learning methods • Use of appropriate technology in an networking environment. (Policy document 1997-2000) Since we have four, even five, campuses united into one university, we have special reasons to promote a networking solution, bridging the distance by using the flexibility that technology offers. The first item, the most challenging one, is a vision that is very well expressed by our visitng professor Liz Burge, from Canada, University of New Brunswick. "Learners are becoming their own information architects rather than being photocopiers of the professor's lecture notes" (Interview May 17, 2000). I will also let Liz Burge pinpoint the challenges that the educational institutions are facing: • Technology is not the issue. Learning is. • Products are not the focus. Learners'needs are. • Conveying information is not the task. Facilitating learning is. • Shaping the learner to the system is not the challenge. Shaping the system to the learner is. (Distum conference, May 2000)
Summing up The general driving forces for distance education could be summed up in four categories, namely as a way to a. bridge physical distances
100
The Nordic Approach to Distance Education and Learning
b. level out social "distance" c. make a profit d. serve as a social and economic spring board for an elite (Hansson 2000). The Nordic countries fall roughly into the first two categories, which also explains our relative absence on the world market, compared to the United States of America, and to some degree also Canada and Australia, where individuals are prepared to pay large tuition fees for higher education. Both in Finland and in Norway there is a greater acceptance for tuition fees than in Sweden where all public education is free and so far the private competitors have been few. Students pay a fee for courses at the Open university in Finland (FADE) and in Norway the universities have to charge individual adult students. In Sweden universities and university colleges do contract education, but not on an individual basis. Here is a point where the Nordic networks may be vulnerable because of different traditions and different attitudes. Other tensions are revealed in the fact that private institutes are recruiting good teachers from the regular universities because the former are able to pay more (Ekberg 2000). The overall trend in the Nordic countries is, not surprisingly, a strong belief in the need for easy access to higher education during an individual's lifespan. ICT plays an important part as an agent in the process. Twelve Swedish universities and university colleges are joining forces in offering different teacher training programmes using ICT. Improving the technical infrastructure is an important prerequisite for a democratic development and a lot of money is put into the physical networks, the national broadband schemes. Both Finland and Sweden (and Denmark!) are joining existing networks and distance education consortia into so called virtual universities on national bases. So, cooperation rather than competition is the hall mark of educational providers in the Nordic countries.
References [1]
Burge, Elisabeth J., 2000: Distum conference, 17-19 May 2000, Harnosand, Sweden
[2]
Ekberg, Karin, 2000: "Hur ar laget i Norge?" I: Per Distans, nr2,jum'2000.
[3]
Finnish Association for Distance Education - Finland URL: http://oyt.oulu.fi/fade/eng/fadeesc.htm Accessed: June 2000
Karen Edberg [4]
Hansson, Henrik, 2000: Distansutbildning och ideologi i ett international It perspektiv. URL: www.distum.se Published: Nov. 16, 2000. Accessed: Dec 28, 2000.
[5]
Holmberg, Carl, 1998: "Norska aktorer inom distansutbildning." Utbildningsdepartementet, 1998-03 -13.
[6]
Individuellt kompetenssparande. SOU2000:119 URL:http:// naring.regeringen.se/propositioner_mm/sou/pdf/sou2000_l 19_sum.pdf Accessed: Dec 2000
[7]
Individual Learning Accounts, IKS — starting year 2002. Final report. Commission on individual learning account. SOU 2000:119.
[8]
Kunskapsstrategi for utbildning och forskning 2000-2004 URL: www.minedu. fi/j ulkaisut/information/svenskaU/2/ 2.html Accessed: December 2000
[9]
Larsen. Sinikka, "The Development Prospects of Open Universities." In: Universitas Helsingiensis 2/1999. Lahti Research and Training Centre of the University of Helsinki URL: http://www.helsinki.fl/lehdet/uh/21999etu.htm Accessed: June 2000
[ 10]
The National Agency for Distance Education, Distum URL: www.distum.se Accessed: December 2000
[11]
The Norwegian Association for Distance Education UTL: www.tunet.net Accessed: June 2000
[12]
Realkompetensprojektet UTL: www.realkompetense.org Accessed: December 2000
101
This page intentionally left blank
Creative and Innovative Network Management O. B. Popov (Ed.) IOS Press, 2003
103
Neki Frasheri
Albania Perspectives of Collaboration Within the framework of the regional collaboration and interconnection of the NRENs, several problems are identified in Albania. These problems may be classified in two groups.
I. Infrastructure and connectivity problems During the last years the telecommunication infrastructure in Albania is improved considerably. The main national backbone is built recently using optical fibers, as well as three main fiber international links with Greece, Italy and Croatia. There are two telecommunication companies in Albania, Albtelecom for the fixed telephony and AMC for the mobile. AMC is just privatized and bought by Telenor, so it is expected for new posiitive developments in the near future. Albtelecom is under consideration for the privatization. Despite the existence of optical fibers, the connectivity remains problematic and all ISPs, including the academic network, are forced to use private satellite links for international connectivity and wireless equipment for metropolitan dedicated links. Theoretically Albtelecom may give leased lines, but in practice it is difficult to be achieved due two factors: • •
The technical availability of Albtelecom to give leased lines, both in multiplexing and on local loops. The high prices Albtelecom applies for leased lines.
The future is uncertain due to the fact that Albtelecom is working to become a national backbone operator. If they would be successful in this enterprise, the things may be improved. But also there is the risks that Albtelecom may enter aggressively to the Internet marked and create many problems for the private ISPs, leading to a monopolization of the market.
104
Albania - Perspectives of Collaboration
Taking into consideration the local situation and the requirements to go forward with the regional collaboration and interconnection of NRENs, an official request is under consideration for the Albtelecom and the Ministry of Economy and Privatization to ask for a reduction of prices in the case of research and education acctivities. At he same time other options may be available and considered, as mobile telephony for example.
II. Institutionalization of the Albanian Research and Academic Network The Albanian NREN started its operations at the end of 1995 with a project leaded by UNDP, and was extended in June 1997 with the Internet project of Soros Foundation (OSFA). It is limited in the area of Tirana, where more than 12 faculties and institutes are connected with dedicated wireless links, as well as dial-up connection are given for many other institutions and individuals for the academic community. The network is operated by the Open Internet Center of OSFA, with the support from INIMA and Polytechnic University (Department of Electronics). This informal consortium is known as "@net", and its activities include: • • •
Keeping secondary level domain names for ccTld of Albania, Giving connectivity, as well as email and web services to the academic community, Organizing local workshops on Internet technologies and services (1997,1998,1999).
After three years of intense activity, OSFA is considering the termination of the Internet project, but it may continue funding of Internet related activities. As result, before the Albanian academic community and its institutions there is the responsibility to institutionalize the "Albanian Research and Education Network" and continue the activities initialized by OSFA. There are two critical actors on whom this institutionalization depends the Ministry of Education and Science and the Academy of Sciences as funding bodies. The academic community is becoming aware about the problem, and we expect some progress in the near future. In this context, INIMA is insisting in the institutionalization of the academic network considering it not simply as an Internet provider but as an inter-institutional research and education laboratory.
Creative and Innovative Network Management O. B. Popov (Ed.) IOS Press, 2003
105
Jan Gruntorad
NRN and Commercial Activities CESNET's Experience
Presentation Overwiev 1. 2. 3. 4. 5. 6. 7.
Evolution of CESNET Network Reasoning for commercial activities Creation of CESNET, A.L.E. Organization of CESNET, A.L.E. Overview of activities Strategic partnership (selling) of CESNET Network Lessons learned - future plans
Evolution of CESNET 1990 1991 1992 1993 1994 -
EARN introduced to Czechoslovakia and run by Czech Technical University in Prague Internet connection over 9,6 kbps links to University of Linz, Austria Grant for building country-wide IP network for Czech Universities (20 mil. Kes - Czech Ministry of Education) CESNET network (64 kbps backbone) in 9 cities operational (run by Czech Technical University in Prague) CESNET Advisory Board nominated by the Czech Ministry of Education services provided also to other organizations than R&E
106
NRN and Commercial Activities - CESNET's Experience
Commercial activities Why were started ? demand for services from outside academic community no "stabile" funding available have "reasonable" salary level to keep skilled staff Negative impact bandwidth providers consider NRN as competitor 1995 -
Czech Ministry of Education terminates funding of the CESNET network 1996 - creation of CESNET, Association of Legal Entities - tender for implementation of the TEN-34 CZ network and pilot applications - start of the TEN-34 CZ programme (1996-1998) - 550 mil Ke (Government Board for Research and Development, Czech Ministry of Education)
Creation of CESNET, Association of Legal Entities Reasons: - end of central (ministry) funding, universities had to pay for services - decision making process - project TEN-34 CZ
Organizational Scheme of the CESNET, A.L.E. 1999 i
Dire clor
1 ...
! I i
Manager
i j l \
employees 6 University employees HighSch.
1 Operational and Technical Dep.
:
I 1
Research and Development Dep. I
,
!
i
Manager
j I
I employees 13 University employees 15 HighSch.
:
\
Supervisory Board _J 1
Board ol Directors -J)
|
H
Secretarial Office 2 employees
1 Marketing and Commercial Dep.
:
I i I
Financial Dep.
,
Manoger
!
employees 3 University employees 5 HighSch.
I I
Administrative Dep.
__ I \ j 1 j
,1. Manager
1 employees 3 University employees 2 HighSch.
] ] I :
1 Manager
,.,
: 1 }
1 employees 5 University employees 4 HighSch.
Jan Gruntordd
1997 -
107
TEN-34 CZ network (34 Mbps backbone) in 9 cities operational start of migration of the R&E users from CESNET network to TEN-34 CZ 1998 - part of backbone upgraded to 155 Mbps - migration of the R&E users to TEN-34 CZ copmleted - CESNET network commercial ISP 1999 - contract with Czech Ministry of Education (Government Board for research and Development) for 5 years (1999-2003) support of "High Speed Network for National Research and its New Applications" - preparation of strategic partnership (selling) of the CESNET Network
Activities 1. 2. 3. 4.
Operate TEN-155 CZ Network International projects (QUANTUM, TF-TANT, NICE,...) National pilot servicies Oparate CESNET Network (90 partners), e.g. issue 8000 invoices/month
108
NRN and Commercial Activities - CESNET's Experience
TEN-155 CZ Network Topology
Strategic Partnership (CESNET Network) January 1999 - 3 studies about evolution of Internet in CZ Spring 1999 - selection of advisor (6 candidates) - Ernst & Young Summer 1999 - start of public tender for strategic partnership for the CESNET Network (66 companies were approached)
Strategic Partner Requirements ready for a very large investment will contribute with alraedy available products and know-how
109
Jan Gruntordd
Fall 1999 1 st round of the tender 6 offer received: BT, Contactel, Czech Telecom, Global One, GTS, Telenor Winter 1999 - negotiation with selected 5: BT, Contactel, Czech Telecom, GTS, Telenor Jan. 2000 - 2nd round: 1. Contactel, 2. Telenor, 3. GTS Feb. 2000 - contract for selling of the CESNET Network signed with Contactel May 2000 - contract comes to effect, start of moving staff and technology June 2000 - handing over should be completed, 45 staff members moved to Contactel premises, 32 will stay with CESNET, A.L.E.
Organizational Scheme of the CESNET, A.L.E.
2000
110
NRN and Commercial Activities - CESNET's
Experience
Some key indicators
Staff
1997 1998 1999 2000
Income CESNET
Income TEN
(mil. Kc)
(mil. Kc)
43 53 64 83(32)
78 127 178 699(54+645)
323 285 300 302
What to do with the money ? (645 mil. Ke = 17 mil. USD) 1) use interest only for new projects and contribute to salaries of CESNET staff 2) use for network operation 3) distribute to universities and Czech Academy of Sciences
Lessons learned commercial ISP is a very competetive business commercial activities need a lot of human resources different priorities in commercial and academic world different skill of staff needed no competitor for telco (projects possible)
Future plans of CESNET, A.L.E. focus on R&E activities (TEN-155 CZ, GEANT, TF-TANT, MoU with 12, ...) sell know-how, e.g. consultancy services, training,...
Creative and Innovative Network Management O. B. Popov (Ed.) IOS Press, 2003
Piotr S^siedzki
Structure and Infrastructure (POL-34 Network)
Historical background 91 -NASKNREN 93 - MAN programme 97 - POL-34
Structure Consortium of MAN networks PSNC - formal representative Bodies: • Strategic Committee • Technical Committee • Working Groups • Network Management
111
112
Structure and Infrastructure
Present status
PIONIER project Polish Optical Internet Advanced Applications, Services and Technologies for the Information Society 2001-2005 2000 - preparatory phase
Piotr Sqsiedzki
PIONIER: Polish Optical Internet
'SZt/JXIN
BYDGOSZCZ
\\ARSZAWA"
113
114
Structure and Infrastructure
Core Technology • • •
Dark fibre DWDM / 1-routing ATM/PoS/GBE/IPoF
Testbed • • • • •
ISThmus 2000 conference WrocOaw-PoznaO (180 km) DWDM 5 x 2,5 Gbps Gigabit routers Applications: • distributed computing • distributed archiving • VoD
Thank you I
Creative and Innovative Network Management O. B. Popov (Ed.) IOS Press, 2003
Vadim Navotny
IT for Science and Education in Uzbekistan
Geography
Short history 1996: CAFE 1997: UzSciNet 1997: Cybernetical Institute
115
116
IT for Science and Education in Uzbekistan
Situation beginning of 1999 • activities were not co-ordinated • high ambitions • government decree
Partners SOROS Foundation UNDP CEENet Nato
UzSciNet backbone
h
Vadim Navotny
Present situation Joint project UNDP, OSI, NATO, AoS Connections with: • State Fund for Information (NATO) • PERDCA (NATO) • NGO's (UNDP, OSI) • Agencies of UN (UNDP) Talking with universities
117
This page intentionally left blank
Creative and Innovative Network Management O. B. Popov (Ed.) IOS Press, 2003
119
Ramaz Kvatadze
Georgian Research and Educational Networking Association (GRENA)
GRENA has been founded in 1999 by Georgian Academy of Sciences Tbilisi State University Georgian State Technical University Tbilisi State Medical University Tbilisi State Pedagogical University Georgian State Agrarian University High Energy Physics Institute of TSU Open Society - Georgian Foundation
Goals of GRENA Development, construction and maintenance of the computer and communication infrastructure in Tbilisi to connect academic and research institutions into unique information system. Providing connectivity of academic and research institutions in the regional cities - Kutaisi, Batumi, Poti, Gori, Zugdidi, Rustavi to Internet through Tbilisi. Connection with international networks. Development and construction of information nodes and networks.
120
Georgian Research and Educational Networking Association (GRENA)
Steps of development •
1994 NATO / INTAS grant Georgia Academic & Research Network Project.
•
1997 NATO grant Improvement of the Georgian Research and Education Network Capabilities.
•
1999 NATO grant Development of Networking Infrastructure of Georgian Research and Educational Networking Association.
•
1996-2000 OSGF - equipment end operational expenses.
At present more then 350 institutions are connected to Internet through HEPI, ICM and Internet Center WOW.
Connectivity • •
HEPI - Satellite Station - bandwidth 128 kbps to DFN. WOW - 448 kbps leased line from commercial provider Sanet.
Basic services of GRENA • • • •
Internet connectivity for educational and research institutions, libraries and other organizations through radio modem, leased line and dial-up connections. Design and development of LAN for these institutions. Standard information services -WWW hosting, FTP, e-mail, etc. User training and support.
Staff Administration - 2 pers. Technical department -20 pers.
Ramaz Kvatadze
121
This page intentionally left blank
Creative and Innovative Network Management O. B. Popov (Ed.) IOS Press, 2003
Avgust Jauk
Access Issues from ARNES Perspective Access policy issues Who the users are • Institutions • Individuals AUP definition • Contracts with organizations • Signed agreements with individuals AUP enforcement • Is it important? • How to do it (AAA?) • Limits on usage? • New access technologies
Access BW issues How to get enough BW? How much is enough (applications)? • (sub) 2 Mbit/s • 10 Mbit/s • 100 Mbit/s • 1 Gbit/s Technology: Leased lines, Cable, xDSL, Radio, SDH, Fiber, VPNs Price/performance, TCO issues
123
124
Access Issues from ARNES Perspective
Expected changes in the market: lower prices EC recommendations Telekom Slovenije 64 Kilobft/s capacity circuit 2 Megabit/s capacity circuit 34 Megabit/s capacity circuit
80 350 2.600
Euro/month Euro/month Euro/month
417 Euro/month 1.292 Euro/month 18.090 Euro/month (*)
Prices of Telekom Slovenije versus EC recommendations
Avgust Jauk
125
Financial evaluation of options at ARNES Comparison of costs • One time costs • Monthly costs • Total cost over a period of 5 years
Total costs over 5 years (EURO)
i' y ; :
• f
126
Access Issues from ARNES Perspective
Total costs over 5 years (EURO)
Conclusions Demand for bandwidth is growing very fast Tight budgets, costs-effective solutions needed Liberalization of the market expected since Slovenia is to join EU Considerable decrease in prices expected, avoid signing long term contracts Seek for long-term solutions Dial-up only as a temporary solution!
Conclusions Fiber has the best price/performance ratio • Lease it, long term contracts might be acceptable • Build it if sensible lease not available • Share costs with other players • Use good relations with local communities
Avgust Jauk
Act as a test user • Good relations with carriers • Try to influence developments Follow market developments in your country and abroad Follow what other NRENs are planning
127
This page intentionally left blank
Creative and Innovative Network Management O. B. Popov (Ed.) IOS Press, 2003
Jasenka Gojsic
The Competitive Environment (Human Resources in a NREN)
CARNet - overview • •
Government agency, founded in 1995 since 1994 connects the whole AC 120 academic and research institutions in 27 cities • 2000 budget = 11 M DEM • four programs in 2000 • Network of excellence • @hr • Friends • CARNet ID
CARNet - human resources •
30 employees • average age 31 • 60% female (18) • 53 % engineers (16) • 73% high education (22) 10% masters degree (3) • 20-40 students • international students • 70 coworkers
129
130
The Competitive Environment
HR experience • • • • •
Strategy development Organization Recruiting Turnover Education & Learning
Evolution of HR focus 1991. 1993. 1994. 1995. 1997. 1999.
Get onto Internet Wire the community Support end-users Provide examples Promote IT & CARNet Share & Help
Organizational forms - evolution 1991. 1992. 1995. 1998.
Idea & first 1M USD - project team Doers Formal organization: four divisions F lat, project based organization
Work forms - today •
distributed organization • five cities • Osijek, Pula, Rijeka, Split, Zagreb • three locations in Zagreb • teleworking • also from home • while travelling • cooperation: • 30 eployees • 70 contractors • 50 projects
» » » » » »
Technicians Technicians Teachers Facilitators Marketing Consultants
Jasenka Gojsic
Recruiting • • •
dedicated person for HR ! long-term and short-term needs criteria: • culture fit (for team-building) • intellectual abilities • basic knowledge • self-manageable • potential to grow with CN development
Turnover •
•
// is natural! For various reasons: • personal development • new ambitions • curiosity • needs IMPORTANT: • concentrate on core people • motivations, organization, speed... should fit the core, not the margin
Education & Learning •
• • •
• •
per employee per year • 10 days • 4000 USD English as a company language keeping up with technology managerial development • MBAs • KSG - Harvard project management, team work multidisciplinary • psychologists, defectologists, teachers • economists, librarians, lawyers • marketing, journalists
131
132
The Competitive Environment
Future • •
•
• •
strong corporate culture • identification & pride profile: • curious, smart, innovative, forward • intrinsic motivated people • broad interest • future, progress, other people, ... learning possibilities • employability • responsibility of employees & employer • knowledge management • to protect organization's investment basic needs should be covered spin-off commercial companies
Creative and Innovative Network Management O. B. Popov (Ed) IOS Press, 2003
133
Basil Maglaris Theodores Karounos Xanthi Psiakki
Greek Research and Technology Network
GRNET S.A. •
Shareholder: General Secretariat of Research & Technology (Ministry of Development)
•
Interconnects: 50 Institutions (Universities, Research Centers, Technological Institutions) with 150.000 Internet users
•
Operates Athens Internet Exchange - AIX: A TTP among all major Greek ISPs
•
Provides International Connectivity: TEN-155, Regional Networks
134
Greek Research and Technology Network
TRAFFIC 5/2000 Incoming Traffic Outgoing Traffic International Incoming Traffic International Outgoing Traffic AIX Total Traffic (ISP Peering)
4549 8421 2457 6496 1292
GB GB GB GB GB
Basil Maglaris & Theodoras Karounos & Xanthi Psiakki
INTERNATIONAL CONNECTIVITY •
TEN-155 (QUANTUM), GEANT Athens PoP 34 Mbps (155 Mbps, June 2000)
•
Q-MED Quality R&E Network Technology for User Oriented Multimedia in the Eastern Mediterranean
•
MEDNET Regional R&E Backbone infrastructure to interconnect the 12 Third Mediterranean Countries (funding under MED A project?)
TEN-155 (QUANTUM) •
High-Speed Connectivity • 17 European NRNs • US Connectivity - Internet2
•
Managed Bandwidth Services • VPNs serving 1ST 5th Framework Projects • Multimedia - Multicast
•
TF-TANT Advanced Pilots • DiffServ • MPLS • WDM • IPv6
135
136
Greek Research and Technology Network
Cyprus
Internet! • • • •
University Corporation for Advanced Internet Development (UCAID, 170 US Universities, Research Centers, Industrial Affiliates) Abilene 2.5 Gbps Backbone International Partner (Connection to Abilene) Participation of GRNET in Internet2 Tech. Workshops, WGs, Experiments
Q-MED •
Connectivity to TEN-155 Israeli NRN - MACHBA (34 Mbps, via London PoP) Cyprus NRN CYNET (4 Mbps, via Athens PoP)
137
Basil Maglaris & Theodoras Karounos & Xanthi Psiakki
Creation of a regional network Involving Italy, Greece, Cyprus and Israel Pilot applications Improve Networking QoS Promotion of Regional Academic Networking Cultural bonds
April 1999 Topology
Cyprus Israel
MEDNET High Quality Regional Connectivity Lebanon, Syria, Jordan, Israel, Turkey, Palestinian Authority, Egypt, Algeria, Malta, Cyprus, Morocco, Tunisia, EU Mediterranean Countries Training Pilot Application Projects Improve QoS Cultural Exchange
138
Greek Research and Technology Network
A basic framework for the evolution of GRNET Research-oriented • •
Testbed for advanced network protocol experiments Using a VPN infrastructure • Different - separate equipment (even duplicate "research" equipment at PoP)
Services-oriented •
Not just plain connectivity, but also value added (reliable and dependable) services • Directory services aware / enabled • Public Key Infrastructure • Mobility aware • Bandwidth brokers • Multicast • Videoconferencing, H.323, MCUs • Voice support
IP-oriented (1) •
• •
Native IP QoS support • Do not rely on "QoS by overprovisioning" • DiffServ model: Best effort (BE), Premium - Olympic medals (AF), virtual leased line (EF) • RSVP support (hard Qos) for experiments Native MPLS (VPN and TE) support • MPLS Switching depends on technology used Native IPv6 support
IP-oriented (2) •
Native IP VPN support (based on MPLS) • Guaranteed bandwidth # VPN • VPNs should not be based on different 1's: • Includes MBS as IP MBS • Includes Best Effort service (= commodity Internet) as separate VPN • Includes Voice network
Basil Maglaris & Theodoras Karounos & Xanthi Psiakki
Other • • •
Effectively use the results and the lessons of Internet! and TF-TANT ATM as separate service (maybe over public ATM networks) US and "commodity Internet" connectivity
139
This page intentionally left blank
141
Creative and Innovative Network Management O. B. Popov (Ed.) IOS Press, 2003
Tofig Babajev
The Telecommunication Infrastructure in Azerbaijan
Research Institutes and Educational Institutes
State and Government organizations
The General Structure of the Network
University
142
The Telecommunication Infrastructure in Azerbaijan
TV Tower
The Internet connection of BSTC and its Sumgait and Nakhichevan branches
Lenkoran 2001
The network of the State Committee of Science and Engineering of Azerbaijan
143
Baku Scientific and Training Centre
Projects & Organizers
STACCIS( 1996-1999) EU/UNESCO/ UNIDO/EDNES Internet Training (since 1995) UNESCO/UNIDO/BHtish Council A AS (1996-1997; TACIS/SOFREGAS (France) IAS (1998-1999) TACIS/SOFREGAS(France) BSTC (1995-1997) UNESCO/UNDP Sumgait CC (1997-1999) UNESCO/UNDP
The Telecommunication Infrastructure in Azerbaijan
144
Nakhichevan CC (1998-2000) UNESCO/UNDP EDNES Azerbaijan branch (since 1999) EDNES Assembly WISTCIS (2000-2002) EC/EDNES Assembly IAIA (2000-2001) FAO(Rome) GSGP (2000) TACIS/SOFREGAS (France) MULTIMEDIA Courses (2000) UNESCO INCO-COPERNICUS EIS-CCE/NIS (1999-2001) EC/IST IDEALIST EAST (1998-2000) EC/IST
Baku Scientific and Training Centre http://www. bstc. azeri. com e-mail:
[email protected]
BSTC
.Teleeoirimuaications department
Training department
Sumgait branch
Software department
" WEB department
Nakhichevan ,Computer Centre
Scientific department
, Institute of CIT
Departiaent of C1T for children
EDNES : Azerbaijan branch-
145
Tofig Babajev
\
Main build ng of a Academy (if Science
| [PC] [PC] ••• [PC] I
N
IPX/SPX
Ethernet
1 «-< && ;
| Ethernet
\ I
Decnet, JNET Multinet VAX/VMS DSV-11 VAX
i
UMD HUB AT-5000
Dec Server 200
1 1 1 1 1 1 Terminals
Moscow
!
A
1
FM] pyil
1
Cisco Router 2501
i i 1
1
i
Uecncf
TfP/IP IU7ir
i! |
The building of Presidium Academy of Science .. Moscow Decnet, JNET Multinet WAVA/IK VAX/VMS DZQ11 VAXSolaris v.2.1
SUN 10 Multiplexor M i l Dial-Up
ofAAS I
Cisco Router 4000 i
LAN NetBEUI
Netware 3. 11 and 4.01 ... [PC] LAN
' |—[
| rjjg nh . . . nh
!
"-I-1 IPYAPY
!
1
SUN 4
Cisco D
Router 2501
Windows 95, Windows NT
The Hardware Structure of the Network
1
—HJ!~|J
i i i
Ethernet TCP/IP
Cis n Rou ter 25 1 L
Building of IMM of AAS |
HUB AT-500 i
|
[M][|l[M]
i
| BSU
I
m in uiai-up | | | ||| Cisco AS-500
! i
| j
i j
IKA "JA 1;
This page intentionally left blank
147
List of Participants
Albert TONEYAN Yerevan ARMENIA Alexander FINAEV Dushanbe TAJIKISTAN Anne MARDIMAE Tartu ESTONIA Asomiddin ATOEV Dushanbe TAJIKISTAN Avgust JAUK Ljubljana SLOVENIA Cristian GARABET Bucharest ROMANIA Ernst G. BOOS Almaty KAZAKSTAN
148
List of Participants
Eugenie STAICUT Bucharest ROMANIA Fay HOWARD London UNITED KINGDOM FisnikMARKU Skopje MACEDONIA Franck BOISSIERE Brussels BELGIUM George SADOWSKY New York USA Howard DAVIES Cambridge UNITED KINGDOM Ihor KATERNYAK Lviv UKRAINE Iskra DJONOVA-POPOVA Skopje MACEDONIA Jacek GAJEWSKI Warsaw POLAND
List of Participants
Jan GRUNTORAD Prague CZECH REPUBLIC Jan GUETTLER Oslo NORWAY Jasenka GOJSIC Zagreb CROATIA Karel VIETSCH Amsterdam NEDERLANDS Karin EDBERG Sundsvall SWEDEN Kevin MEYNELL London UNITED KINGDOM Kvatadze RAMAZ Tbilisi GEORGIA Laimutis TELKSNYS Riga LATVIA Maray TAMAS Budapest HUNGARY
149
150
List of Participants
Marko BONAC Ljubljana SLOVENIA Michael PSHENICHNIKOV Bishkek KYRGHYZ REPUBLIC Mihkel KRAAV Tartu ESTONIA Neki FRASHERI Tirana ALBANIA Nicolai SANDU Bucharest ROMANIA Oliver ILIEV Skopje MACEDONIA Oliver B. POPOV Skopje MACEDONIA Orlin KOUZOV Sofia BULGARIA PavolHORVATH Bratislava SLOVAKIA
List of Participants
Peter RASTL Vienna AUSTRIA Petras SULCAS Vilnius LITHUANIA Piotr SASIEDZKI Warsaw POLAND Pulat PULATOV Dushanbe TAJIKISTAN Richard PERLMAN Berkeley USA Riina REINUMAGI Tartu ESTONIA Robert JANZ Groningen NETHERLANDS Sandro KARUMIDZE Tbilisi GEORGIA Sergey KUSHCH Tashkent UZBEKISTAN
151
152
List of Participants
Theodores KOROUNOS Athens GREECE Tofig BABAYEV Baku AZERBAIJAN Turgul YILMAZ Istanbul TURKEY Vadim NAVOTNY Tashkent UZBEKISTAN Vahan AMIRBEKYAN Yerevan ARMENIA Valerian LEVINSKI Kishinev MOLDOVA Vasilis MAGLARIS Athens GREECE VugarZEYNALOV Baku AZERBAIJAN Vyacheslav SHKARUPIN Kiev UKRAINE
List of Participants
Wojtek BOGUSZ Warsaw POLAND Xanthi PSIAKI Athens GREECE Zita WENZEL Marina del Rey USA
153
This page intentionally left blank
Organization NATO-country Co-director Jacek Gajewski CEENet / Warsaw University Warsaw, Poland
Partner-country Co-director Iskra Djonova - Popova Saints Cyril and Methodius University Skopje, Macedonia
International Programme Committee Zita Wenzel University of Southern California Los Angeles, USA Jan Gruntorad CESNet Prague, Czech Republic Peter Rastl ACONet / University of Vienna Vienna, Austria
156
Organization
Robert Jam University of Groningen Groningen, The Netherlands Oliver B. Popov (Chair) Saints Cyril and Methodius University Skopje, Macedonia Francois Fluckiger CERN / University of Geneva Geneva, Switzerland
International Organizing Committee Oliver B. Popov Saints Cyril and Methodius University Skopje, Macedonia Fisnik Marku Ministry of Science Skopje, Macedonia Iskra Djonova-Popova Saints Cyril and Methodius University Skopje, Macedonia Jacek Gajewski (Chair) CEENet / Warsaw University, Warsaw, Poland Jean-Paul Nadreau University of Laval Quebec, Canada
Organization
Local Organizing Committee Biljana Stojcevska (Chair) Saints Cyril and Methodius University Skopje, Macedonia Svetlana Grncarovska Saints Cyril and Methodius University Skopje, Macedonia Sanja Lazarevska Saints Cyril and Methodius University Skopje, Macedonia Tomislav Kik Sitografika - Studio for Visual Design Skopje, Macedonia Svetlana Unkovska SAVANA Travel Agency Skopje, Macedonia
157
This page intentionally left blank
159
Author Index Babajev, T. Davies, H. Djekov, T. Edberg, K. Frasheri, N. Gajewski, J. Gojsic, J. Gruntorad, J. Guettler, J.H. Howard, F. Janz, R. Jauk, A. Karounos, T. Kvatadze, R. Maglaris, B. Meynell, K. Navotny, V. Perlman, R. Popov, O.B. Psiakki, X. Rastl, P. Sadowsky, G. Sajsiedzki, P. Staicut, E. Trajkovski, B. Vietsch, K. Wenzel, Z.
141 51 xv 95 103 93 129 105 59 71
1 123 133
119 133 71 115 87 vii,xvii,13 133 83 37 111 79 xi 27 75