JISC

Education content: Contribute, Distribute and Retrieve

Introduction
Broadcasting
Interactive television
Data broadcasting and multicasting
Internet access
One and two-way connectivity
VSAT networks
Costs and Service considerations

An introduction to different types of content

'If content is King, then connection is God' is a statement that is often heard in discussions around content distribution, no matter if it is about educational, business or commercial content. One cannot live without the other. Creating content is of no use if the creator cannot convey the message one way or another to his/her target audience. The type of technology that is right for the message, depends on many different aspects of the content and of the audience. In this chapter we will look first at the different types of content. It is important to have a clear picture of this before deciding further on the technological solution for the communication.

First of all, we should agree on the fact that content needs to exist in some material form. This can be in audio: the human voice, a conversation, a discussion, music, a speech, a lecture, a play… all are originally auditive information only, but all can be recorded and/or transmitted. Content can also exist in written form: a handwritten note, a letter, a newspaper, a magazine, a book, a course, course notes, a written assessment etc. Content is sometimes packaged in visual information: a drawing (eg, a drawing on the blackboard), a sketch, a map, a graph, an image, a photograph; even in moving images: a demonstration, a show, a film, a video, a computer animation, often accompanied with suitable audio information: for example, the soundtrack that goes with the film, a voice over on a videotape, or sounds accompanying the animation.

The above information can exist in its original form (the sound of your voice 'as you speak') or in a captured form (a tape recording of the same voice). The format of capture can be analogue (eg, the traditional audiocassette recorder is an analogue recording technology) or digital (eg, the audio CD). In both cases the content is no longer exactly the original, but a representation of the original. A collection of content representations (eg, a library, a collection of cassettes or CDs, but also an archive of messages that were exchanged between a number of people) can also be considered some typical form of content. Some of these can be considered libraries or resources, some are almost like databases, others start to look like Knowledge Bases. These resources are sometimes static (the content does not change once the resource is created, eg, a library with all the works of Shakespeare). In other cases the content is dynamic and the resource can be changed interactively by all their users at all times, for example, the contents of a Newsgroup on the Internet.

A less obvious type of content is data itself, although they have to be structured in some kind of a database to be usable and useful of course: for example, a contacts database or an information management system. Web sites are also examples of this: in essence they are archives with files, hierarchically connected and linked with each other. The files within such an archive themselves can contain all types of contents as described above: sound, images, text, and more.

Another form of content is a procedure or a programme: this can be in the form of a software but it can also be less tangible, for example, in the form of a procedural description: for example, the description of how to perform certain dance movements or a cooking recipe easily fall in this category although one could argue that they need to be materialised in one or the other way: in text, in an aural explanation, with drawings or pictures etc. A more conventional example of this content is a computer program of course: this can be a copy of application software, for example, a text- processing or a spreadsheet application; it can also be some training programme, a simulation, a virtual reality application or a computer game where interaction between man and machine is required.

The most evolved form of content is a combination of all elements, where different types are combined in such a way that they make an effective and balanced mix. Blending is the term that is being used nowadays to describe this. It is often used in education to describe the combination of a technology-based delivery mode (eg, videotaped materials) with classroom teaching and learning but the definition can be used also to more elaborate mixed concepts: learning environments with classroom lessons, videoconferences with CD-ROMs etc.

Websites can be considered the ultimate way of converging different types of content. A web site is a collection of web files on a particular subject includes a beginning file called a home page. These web files can consist of various types of media: text, audio, images (still, moving, animated, virtual) and interactive elements: software applications and games, virtual reality applications, simulations etc.

How users want to access content: unicast, multicast, broadcast

Depending on the nature of the content and the interest of the user in a particular content, there are different modes in which content is being delivered. Let's look at the various bits of information and communication that we receive or consume on an average day.

A handwritten, personal letter that arrives by ordinary mail ('snail mail' as it is being called nowadays) is unique: there is only one single copy, it delivers a personal message and travels from the sender to the receiver along a unique path. Newspapers on the other hand are less unique: maybe in your street there are a number of people that have subscribed to the same newspaper: they will receive the same newspaper at the same time, but each will have his/her own copy. In total there are maybe tens of thousands of newspaper copies that left the printing press at the same time on their way to their subscribers. Another kind of information is the door-to-door publicity: when campaigners want to hit the biggest mark, they organise a door-to-door distribution campaign for their leaflets: there is no subscription needed, everyone with a letter box will receive the content.

These are three types of delivery modes: the first, a personal communication mode, is called unicast: the communication is strictly individual. It applies to personal letters, to making a phone call, to email, to sending an SMS, but also to certain information retrievals from a web site: for example, when one performs a specific search in Google, Google will return with a customised and individual response. Nowadays, unicast communication is still the most important communication mode. Consider the following generalisation: while newspapers take up 90% of the volume and weight of mail, they only account for 10% of the revenues of the Post, letters and other personal mail are net subsidisers of newspaper delivery. In Finland, revenues from SMS messaging have surpassed in 2002 the revenues that commercial TV stations make from advertisement. Email is still the killer application on the Internet, more than the web, and its share of the Internet traffic is still growing. Unicast is a way of communicating rather than of distributing information: normally sender and receiver know each other and acknowledge receipt of the message.


Unicast communication model

Broadcast mode is when the message goes to all potential users that are within reach of the network or the signal. Broadcast here means that the signal is cast (sent out) in all directions at the same time. A radio or television broadcast for example, is a programme that is transmitted over the airwaves for public reception by anyone with a receiver tuned in. When used in relation to email it means distribution of a message to all members, rather than specific members, of a group such as a department or enterprise. Broadcast mode is a typical way of sharing information with as many as possible: normally a broadcaster is more interested in getting his/her message out to as many as possible, without knowing the individual receivers personally, rather than to communicate back and forth with each of them.



Broadcast communication model

Multicast is communication between a single sender and a specific group of receivers, where the sender transmits a specific content set to a specific target group. A typical and traditional example of multicasting is the newspaper that is delivered by mail. The same content travels almost all over the area and past every door, but is only delivered to the subscribers. Pay TV, where people subscribe to certain programmes on TV, is another good example: the signals are transmitted all over the area, but only people that are allowed to decrypt the content will effectively see the content. In Internet terms, multicasting is used to send content to certain well-defined IP addresses, or to a specific range of addresses.



Multicast communication model

Normally, switched and wired networks are ideally suited for enabling unicast communication: for example, the wireline telephone network allows for a unique channel of communication between two parties. Because of the switching capacities of the network, it is possible to use the network highly efficiently: many parties can have simultaneous distinct exchanges while the overall capacity of the network is only restricted to its own switching capacity and the number of participants. Adding a new participant to an existing wired and switched network however, requires establishing an additional physical connection between this new end-user and the switching point³. 

3.  Wired networks are not necessarily always switched networks. A Cable TV Distribution network, for example, is not switched and is typically a broadcast medium: it is used to distribute an identical bouquet of content from a central source to all cable-connected subscribers at the same time, very much indeed like terrestrial (Hertzian) television reaches all viewers that are within reach of a transmitting antenna. 

To make the distinction between different communication networks even more complicated, you should note that Cable TV distribution networks are increasingly converted into two-way interactive networks that also allow for typical unicast applications: in countries with a high penetration of Cable TV distribution, it is becoming increasingly possible to use telephone and Internet access services via the Cable network.



Switched and wired network

Wireless and radio networks are basically much better suited for broadcasting activities such as radio and TV. Intrinsically they are less suited to enable a large number of discrete two-way communications because of a number of limitations: the fact that the transmission medium is shared (the radio spectrum or available bandwidth in the ether), plus the fact that access to the transmission medium is almost completely under the control of the transmitting party (without the coordination that takes place in a switched network environment) makes it necessary to carefully regulate and agree on the use of the available radio spectrum. Transmission of a radio or TV programme to many is easier via this wireless and unswitched environment because the signals transmitted basically reach all the participants that are under the coverage of the transmission footprint. Adding new participants (typically called 'receivers') to such a network requires only installation of end-user equipment (or 'receiver') at the new participants' premises.

Non-switched wireless networks

The fact that a network is wireless does not necessarily mean that it does not allow for switched (or unicast) communications. The wireless telephone network (GSM or mobile phone network) is a good example of wireless switched. Another example is Internet via satellite. Both are typical unicast applications. To establish a one-to-one connection they work on the basis of a request for time-limited use of part of the available wireless spectrum. Concurrent uses of the available total bandwidth are possible as far as its capacity reaches. This issue is expressed in the contention rate, which indicates the maximum number of participants that can be sharing the service at one given moment. While it is the norm of the network provider to have some level of contention (for wireless and satellite networks, the contention rate varies between 1:20 and 1:50⁴), it is best to obtain the lowest contention rate possible, in order to get the best service and to reduce the potential for a customer to experience congestion (communication interrupts) or reduced download speeds.

The network service provider has to ensure that when the number of users of the network increases, also the number of available channels (the total amount of available bandwidth) increases at a similar pace. Otherwise the contention rate will drop and users may start to complain of bad connection service. This is what happens when the number of users of mobile phones increases far faster than the network capacity: the consequence is that many connection attempts fail because of a network overload. Similarly in satellite-based Internet access, users compete with each other for the available bandwidth: the satellite service provider has to keep up with the bandwidth demands of the user population and increase the total capacity accordingly.

4.  Note also that wired network service providers (Dial In or ADSL) deploy their services on the basis of a similar contention rate, comparable to the rates given here, otherwise their service offer would become too expensive.

Interaction

The concept of communication comprises a certain form of interaction. Interaction in terms of information and communication technologies has three dimensions: the first dimension is topology (the actors between which the interaction takes place), secondly there is a time dimension, and then lastly there is an aspect of symmetry.

Topology

Interaction can happen between a human being and a machine: for example, between someone sitting at a computer and the application on that computer. Computer-based training programmes are of that type. Interaction of that type can range from simple actions (switching on and off the TV set can be considered the lowest level of interaction) to complicated and multi-layered activities: for example, the kind of activities that take place on web-based learning environments. Simulations and CD-ROM-based learning applications fall also within this category.

Interaction also happens between human beings: discussions, dialogues, telephone calls, letters etc, all are examples of interactions. Some happen between two parties: a telephone conversation normally takes place between two people. A lecture in a classroom is an interaction between one (the teacher) and many (the students). An electronic news group or a bulletin board is an example of many to many people interacting with each other.

When selecting an information and communication technology to support one or the other interaction topology, there are a number of obvious choices that can be made: for a private communication between two people, telephone is an obvious choice when these two people cannot meet physically. When one person wants to address many people at the same time who are dispersed over a large geographic area, broadcasting via radio or TV may be an appropriate means. Whenever there is a large number of people involved, groupware systems or news groups and billboards can be a solution.

Time

Another aspect of interaction is the element of synchronicity: in some cases it is necessary to communicate directly and without any delay between the exchanges: telephone conversations are the obvious example, videoconferencing too. Radio and TV broadcasts are synchronous too: if the receiving parties in the target audience are not watching, listening or recording while the programme is being transmitted, then their opportunity to receive is lost. Email is a typical asynchronous person to person communication system: the message is made available until the addressee has time to collect or read the message. In that sense, email compares again to the letter that is posted and sent via email or courier. Email may go as fast as the speed of light, but email does not require the receiver to be present at the other side of the communication chain at the moment the message is sent. Therefore it is called asynchronous.

Teaching in front of a classroom is synchronous: it only happens when teacher and students are effectively present at the same time.

Symmetry/asymmetry

A third quality aspect of communication is the symmetry of the exchange or transfer of information. A normal conversation between two people can easily be understood to be symmetric: both parties have an equal say. In a classroom on the other hand, with one teacher lecturing to a hundred students, it is obvious that the communication is not so equal: the information flow from professor to students will normally be many times larger than vice versa. The same principle applies to technology-enhanced communication. Depending on the symmetry of the communication flow, it will be necessary to opt for the appropriate communications service: when the parties are equal contributors, videoconferencing may be the preferred solution. Email normally also is symmetric in the sense that all subscribers have equal possibilities to contribute. Radio and TV are asymmetric: the content flowing from the transmitting party towards the receivers is much larger and of much higher quality than the communication that listeners or viewers will be able to return.

Communication and information exchanges by means of the World Wide Web display a similar character. Browsing the web is clearly an asymmetric activity: the browser sends very small amounts of data to the server when he/she clicks on a hyperlink to request a page view. The webserver on the other hand returns a lot of data (the requested page view with all the page elements such as text, images, sounds, interactivity etc) towards the end-user. Browsing the web can effectively be called asymmetric communication. The webmaster and his/her web development team who are responsible for the content offered by the webserver on the other hand are more contributing towards the webserver than retrieving: their communication path may well be rather symmetric or even inversed asymmetric (more contribution than retrieval).

Matching a pedagogical model with an educational information and communication technology selection

The learning and teaching methodology may mean that one or the other information and communications technology has to be selected. The table opposite gives an indication of how each of the most common used technologies specifically addresses the communication quality issues of topology, synchronicity and symmetry.

In this chapter we have not referred to satellite communications technology at all. This will be done in the next part. We believe however, that it is important to understand the basics of (educational) content delivery issues before one can apply specific technological solutions to them. In the next part we will attempt to clarify the potential relationship between the content, educational technology selection and satellite telecommunications.

Although it is not easy to categorise, it is useful to describe the various ways in which satellite communications can be configured and used in an educational context. In this section, we will provide some broad categories, describe the distinguishing features of each category and, where appropriate, the various sub-sections that occur within each category.

It is important to begin with a number of reservations to avoid misleading the reader. First, the technological environment is changing fast and therefore certain distinctions that applied in the past are no longer relevant and emerging services often span two or more of the categories described. Secondly, even where technological change is not a factor, there is a lot of 'carry-over' from one category to another and we have tried to point this out when it occurs. Thirdly, many terms like 'Interactive Television', 'Videoconferencing' and 'Multicasting' are being used differently, and there is little common understanding of terminology in this field. These distinctions are often region specific, for example, the term 'Videoconferencing' is used differently in Europe than in the USA. Different uses of the same terminology often occur depending upon sector or industry. For example, the broadcasting industry uses terminology differently from the computing industry.

Broadcasting

Most people are familiar with the use of broadcast radio and TV to support education, as this has been a common means to provide educational service to potential learners for many years. Probably the first educational TV programme was Sunrise Semester, broadcast from Chicago in the USA in 1959. Continuing until the early Sixties, Sunrise Semester featured a single broadcaster, a teacher, standing in front of a class with a camera shooting over the heads of the students. The initiative ceased because it was not economically sustainable.

Traditional educational television is one-way, sent by the broadcaster to the end-user at a fixed time and according to a set and pre-ordained schedule, as in the case of School TV and Open University programming carried by the BBC.

Categorised by its ability to address large potential groups of users, it is common for broadcasters to use satellite technology to transmit their signal, particularly in regions of the world where terrestrial broadcast services are unsuitable. The satellite technology used in such instances is usually a large-scale professional service involving large transmitting Earth stations and significant technical resources for production and play out of programmes. Reception equipment comprises a small dish with a satellite receiver. Many specialised broadcast channels offering, for example, sports, financial information and targeted programmes, are broadcast in this way with the set-top box acting as a filter ensuring that all licence and other fees are paid before the end-user gains access to the channel(s).

Satellite supports good-quality audio and moving video images thanks to its high bandwidth transmission capability. Production costs are relatively high and there is no possibility for interaction. Increasingly however, educational broadcasters are looking to embed this medium in a learning environment supported by other means. It is common nowadays to find associated web sites, help-desks and other support services for people accessing educational resources in this way. Well-known educational broadcast services in the world, like the China Central Radio and TV University, are transforming this type of service into a more interactive model (see next part on 'Interactive Television').

Numerous examples of this type of application exist including EMMA, BBC Education, Swedish Educational Broadcasting and InTeleCom in the USA. Some are associated directly with public service broadcasting organisations. Others operate more as educational programme producers that are selling programmes to broadcast television stations.

Interactive television

The term 'Interactive Television' is one of the most confusing terms. It can refer in fact to any form of interaction with a television or broadcast service. This ranges from tele-polling, choosing camera angles in sports events, to using shopping channels, to video-on-demand. Many interactive television services use satellite technology to support at least part of the communications chain. Typically an interactive television service will involve various types of media, each supporting different functions. Let us look at a number of different samples to explain the range of applications possible in an educational context.

Video-on-demand

This kind of interactive television service usually involves a broadcasting station setting up a service whereby viewers can choose to have programmes sent to them on demand from a video server. This can be on an 'instantaneous' or 'near-instantaneous' basis: depending on the service, the programme is either instantly available to the end-user, or available later. The broadcast of the requested programme can be done via satellite. The ordering or request system is usually facilitated via a terrestrial telecommunications network or even via the Internet. These kinds of services are increasingly on offer from cable operators and there are some examples of educational services using this service. The service offered by Les Amphis de france is a good example and is described in chapter 6. This type of service in an educational context demands considerable technological and organisational resources but is very useful when considered as a way in which educational institutions can manage access to large resources of video-based material. It is essentially pre-recorded and little opportunity for any 'live' interaction exists. Material can be maintained in a central service and then accessed when required by the institution or individual wishing to make use of the material.

One-way broadcast with asymmetric return

This model of interactive television services is increasingly common. What happens is that viewers watching a broadcast programme interact with those in the studio or support network via telephone, Internet chat or email messages or via a videoconferencing link. In this, the broadcast is in the form of listener feedback or questions to the studio panel. The system is essentially 'live' and can be configured in a number of ways. It supports synchronous communications although elements can be asynchronous, for example, using email for question and answer sessions after the 'live' event. Broadcast via satellite is often the way in which the signal from the studio is sent to the viewer, the return link is usually via terrestrial means. The quality of the signal from the studio to the viewers is pretty much always of higher quality than the incoming one (from the viewer to the studio).

In the educational context such networks can be set up either as 'closed' networks, whereby those taking part are known to the educational provider, or 'open' where this is not the case and anyone with the means to receive the broadcast signal can interact with the studio and ancillary services. Interaction can take the form again of questions and discussion, polling and feedback, even group work in remote sites with feedback provided to the central studio via remote group leaders is possible.

This kind of broadcast using satellite one-way and various different sorts of return channels is commonly used by commercial training channels like the Computer Channel or by large corporations including Ford and Daimler Chrysler for training staff. It is also used by those providing specialised medical educational services like Rockpointe Broadcasting in the USA or Plymouth University in the UK. There are a number of examples of this application in the educational domain, in our list of sample projects; these include the African Virtual University and Consorzio Nettuno. 

Data broadcasting and multicasting

Data broadcasting and multicasting networks are often supported via satellite systems. These are essentially one-way communication networks offering data to the end-user, such as video files, web site contents, analytical and statistical information, applications (software updates) or any other form of information that can be digitally stored. The end-user stores the transmitted data on either a PC or on some sort of set-top box. These set-top boxes are developing into Personal Video Recorder (PVR) or Multimedia Home Platform (MHP) systems, supporting standardised data broadcasting and multicasting. The systems increasingly use the Internet TCP/IP protocol in the management of the data transmitted. The way the network is configured may include satellite download for one channel of the network, ie to the end-user using the satellite capacity to handle the bandwidth requirements, which may be high if there is a lot of video in the material being downloaded. The way the network is configured can allow for specific addressing, ie multicasting to specific recipients so the service provider knows exactly who is receiving the data and when it is sent out, or it can simply be broadcast to a wide variety of recipients. This kind of network is used frequently by educational providers as it allows for secure and managed distribution of data resources and a couple of examples of this type of application have been included in Chapter 6, including Espresso and the Austrian AVD project.

Internet access

Many Internet Service Providers use satellite services to support one or both channels of their service to connect their subscribers to the Internet backbone. This is particularly true of developing regions where the terrestrial telecommunications infrastructure is poor. The type of infrastructure is normally based on relatively large-size antennas and significant resources for hosting, providing gateways, proxy servers, security etc.

When considering how this use of satellite fits into an overall educational perspective, it is interesting to recall that the Internet really took off as an academic network. Very often, it is educational institutions and universities in particular taking the first steps in providing an Internet service and Point-of-Presence (PoP) in regions or communities where such services were not previously available. As university campuses extend their reach and educational providers of one kind or another seek to reach new learners the question of creating new PoPs arises. Satellite is often the only way such services can be extended and there are important implications here regarding licensing on a national basis and on how the provision of such PoPs can be made sustainable. It is important to follow market trends in this rather fragmented ISP environment to make sure that educational providers have access to good quality and reasonably priced Internet connections either though a commercial service or by operating their own service. Satellite technology can be used to network ISPs should the need arise for outreach to underserved areas or to set up a content delivery network.

One and two-way connectivity

New satellite technology and more specifically Internet via Satellite can provide high speed IP connectivity via satellite with all the advantages of commercial digital television: wide uptake, high quality of service, scalability and data transmission capabilities.

Until very recently the vital return connection, the interactive connection or back-channel, happened via terrestrial lines, mainly through a dial-up modem connection via telephone line. This is a logical configuration given that most Internet applications are typically asymmetric - the traffic from client to server is usually much smaller than vice versa. This is usually in the order of 5 to 10% upload versus download, except for content creation and contribution. However, the return channel can also be supported via satellite as explained in the previous chapter.

The availability of these services via satellite for the end-user is an important development that has terrific potential for educational authorities wishing to serve remote learners or learner communities. Information about the likely players in this field is contained in Chapter 7 of this report. Typically such services operating in the Ku- band or mixed Ku-/Ka-band offer asynchronous network configurations of typically 56-128 Kbps outgoing (from the end terminal) and 200 Kbps to 8 Mbps incoming. Service operators typically target so-called SoHo end-users (Small Office/Home Office), but they are equally of value for educational users. They tend to operate with relatively low-cost, simple to install end-user equipment, which is hosted on either a separate box or as software installed on a PC. Dish sizes from the service providers currently active in the market range from 90 cm to 150 cm. Unlike the satellite services utilised by ISPs, these networks are aimed at the single end-user or as a gateway to a small and local area network (LAN). Licensing and network configuration are important issues to consider when considering these types of satellite-supported services in the educational context. How the network is configured depends upon the practical use that the organisation offering the service wishes to make of the service, and so there are a number of sub-categories of potential educational use here. In broad terms we categorise them as the following.

Virtual Classroom scenario: in this case, the individual end-user station is part of an educational network whereby other learners, teachers and resource people and materials are remote from the end-user. The system is used for a variety of applications that are a surrogate for normal 'classroom-type' activities. These can include quasi-synchronous communications (usually online chat), asynchronous communications using a closed bulletin board type system and a common store of resources usually housed on a remote server, which are available to the user on demand or as part of a multicast set-up where digital materials are sent to the end-user's storage device via satellite and accessed when necessary.

Resources-based learning scenario: in this case the teacher and immediate learning peers are in the same location and use the satellite service to access resources when required. These resources can be accessed either with an open Internet-type connection or to a closed Intranet hosted remotely at the location of the satellite up-link server. This is a similar network set-up as described in the table on page 65.

See examples of projects using this type of service - SchoolSat and the JISC 2-way Satellite Access trial.

VSAT networks

VSAT communications can be used to set up virtual private satellite networks of one kind or another and are very often used in the commercial world to provide entire communications networks to outlying companies and institutions. Using relatively small-size dishes (by comparison with broadcast-type applications) such institutions use these kinds of services to support telephony, data communications and videoconferencing between a closed user set. VSAT networks are very common for example, in the banking world where commercial banks use them to provide a secure and comprehensive communications network. The exact configuration of such a network will depend upon the resources invested by the network operators, the bandwidth capacity required and often the location of the end-users. They can be used in both synchronous and asynchronous configurations. Many organisations favour an entirely closed system for both security and management purposes, given that the overall control of the system is completely in the hands of the owners and can be managed and controlled centrally. Such networks require high up-front investment and compliance with national and regional licensing authorities.

In the educational context, VSAT networks have significant advantages and allow the organisation flexibility in controlling the educational environment - within the technical constraints of the chosen network of course - to create learning scenarios with the precise media mix required. A good example of this kind of network is the Global Development Learning Network (GDLN) set-up by the World Bank and described further in Chapter 6. Interestingly, this network was built on the existing VSAT communications network used by the World Bank for all its communications requirements. Now an independent network, it seeks to collaborate with other institutions with common aims, interested in sharing the resources of the GDLN.

Meeting educational needs by means of Satellite Networks

Before discussing the application of satellite technology in education further, it is useful to take a look at how generic educational activities can or might be supported by satellite services and where such services are most suitable. The following table provides an overview of this topic.


Overview of Satellite Supported Education Applications

Costs and Service Considerations

This section will provide a basic list of the costing considerations that service providers need to consider. These include hardware at both the send and receive sites, software and server considerations, bandwidth costs, licensing, maintenance and service. It will also cover aspects such as training, operation and personnel support and maintenance. The section starts with three examples of cost models, related to three of the most familiar satellite applications for education and training. In the next part we will elaborate on the different issues that surround the costing structure. Prices are calculated on the basis of averages and can only indicate a range of order. Most of the prices are susceptible to change especially in regions with a lot of competition. It has to be said that prices do not only differ from country to country (in some countries huge mark ups are applied with no apparent method of accounting), but also that fortunately prices tend to drop, sometimes slowly but steadily.

Cost models

Satellite TV broadcast

The cost structure for satellite TV broadcasting is the simplest. We assume that the content is pre-recorded in an acceptable format and quality. The costs of production are not taken into consideration as they are not directly related to the transmission format: it can be assumed that they would be the same if the content were to be distributed on VHS tapes or via broadcast TV.

The costs can be divided into costs occurred at the originating side and costs at the receiving side of the chain. Costs can be quantified easily because they are directly related to the duration of a programme. We present below the cost for a one-hour programme, on a purely occasional basis. Tariffs will change according to the choice of satellite, the frequency of use, the number of transmissions, the time during the day of the transmission, the flexibility of the broadcast and several other parameters.

At the originating side:



Satellite TV Broadcast Costs

Transmission in digital format is cheaper because it occupies only a small part of the bandwidth and therefore it can be combined easily with other content materials.

At the end-user side (cost per site):

Satellite TV Receive Cost

The cost at the end-user side is per installation, not on an hourly basis. The depreciation time for this type of hardware, given the status of transmission standards, can be estimated to be at least 5 years. The number of receive site installations is unlimited in number, it is only limited geographically by the footprint of the chosen satellite and transponder. Because the cost at the originating side does not alter with the number of receive sites, it is easy to see that the cost per user per hour decreases by the quantity of clients.

One hour broadcast in digital video to one single receiver will cost US$ 1,100 per user per hour. For 1,000 hours of broadcast (that is one hour per day over 3 years time) transmitted to 10,000 viewers the cost will be about US$ 0.1 per hour per user.

Note: the above example is for non-supported content materials. If - as is mostly the case in educational TV programming - additional support (support materials, tutoring, helpdesk, etc) needs to be put in place, the cost to do so will be affected by the numbers of applicants.

VSAT community network

As in the previous example, we will provide an example of cost calculation without directly referring to the cost of the content creation, the pedagogical model, the standard or format of content exchange or the specific type of communications. A VSAT network, furthermore, can be tailored around specific requirements: architecture (star or mesh), bandwidth and power are just a few of the parameters that come into play. In the example below we describe the cost for a star network with broadband capacity (2 Mbps) available 24 hours a day all year long. This system allows for a mixture of videoconferencing with data and multimedia distribution and Internet connectivity. The specific costs for these applications are not included in the example below.

A VSAT network is a centrally organised and controlled networking system. Costs cannot be discriminated so easily between originating (central education provider) and individual clients' costs. Because scale of deployment plays an important role, it is more appropriate to calculate the total cost of operation for the whole network.



VSAT Network Deployment Cost

The table above takes into account the following satellite communications-related cost factors:



In VSAT networks, dimensioning starts to play an important role: the number of clients will indicate whether it is cheaper to share a central hub or to invest in an own hub, the available bandwidth and the connectivity to the backbone will also depend upon the dimensions of the network.

The above cost gives an indication for a full-time high-bandwidth secure private communications network. While the investment cost may be frighteningly high at first sight, it is important to look at the operational costs (total cost over usage period per terminal) and it becomes clear that VSAT is a solution for broadband communications that can effectively compete with terrestrial solutions, even in countries where good quality terrestrial connections are available.

Broadband Internet via two-way satellite

Small two-way satellite Internet systems using Ku- and Ka-band are becoming more and more popular. They can be considered VSAT systems but they differ from the previous example in the sense that the network is set up by satellite providers who service a less homogeneous community of users. The service usually offered is limited to some form of high-bandwidth Internet access (in various bandwidth levels according to the customers requirements), sometimes extended with customised applications that are implemented to serve a particular part of the audience by multicasting, special Intranet type of applications etc. In theory there is no need for the education provider to worry about the investment and running costs at the hub or server level of the system: it is enough to invest in client stations and connection fees. The service itself (connection of the hub server to the Internet backbone, satellite bandwidth sufficient to service all clients, maintenance and operation of the service) is not a cost issue to the content provider. Specific centralised costs for the course or content provider are related to contribution costs (bringing or 'porting' the content from the designer or producer to an access point within high-bandwidth reach of the hub server, or preferably even on the hub server location itself), or the investment in specific server-side applications or tools, if they are not supported by the service provider.

In this case, economies of scale do not affect either the end-user or the education provider. They are absorbed in the commercial plan of the service provider. However, it may be worthwhile discussing the deployment plan for a network with the service providers involved, especially when it is very large and demanding, when certain levels of quality of service are required or when expansion is foreseen. It may turn out that migration from a public or semi-public service towards a VSAT community solution using the same technology is a more efficient solution.

Costs can be estimated easily:

Two Way Satellite Internet Cost

Compared to the VSAT network, costs are very low. Compared to terrestrial broadband network costs are competitive, certainly when taking into account the availability of the connection virtually everywhere almost instantly.

The big difference with the VSAT network solution is the fact that the network capacity is shared with a virtually unknown number of concurrent users, which may affect the network performance. The speed of download depends upon the number of users logged on and consuming bandwidth simultaneously, ie the contention rate. The same is true for Cable connectivity and to a lesser extent for ADSL, where different connections need to share the same bandwidth.

Another disadvantage compared to VSAT, is the limitation of application implementation that content providers most likely will incur: this will hinder for example the use of synchronous bandwidth demanding applications such as videoconferencing.

Service costs

With the term 'service costs', we refer to all costs that add up to the total cost of the delivery of a service (be it an occasional offer or an ongoing service). We have distinguished so far a number of cost elements that are directly related to the delivery and thus end up on the final bill of the content provider and/or the end-user.

Furthermore, in this part we will go briefly into the trends that are or will affect price evolution of these cost elements. This will involve some assumptions and crystal ball gazing, because as we know from the last 10 years, economic factors tend to change within periods of 6 months due to the rapid development of new technologies, the stormy nature of the Internet and the changing economics in which the largest players, especially the telecom companies, find themselves.

Cost elements

This section seeks to identify the different cost categories that come into play when talking about ICT-based education and training. The highly important management issues that are related to the change that ICT-based education can bring into institutions are not examined here, because they can be considered external to the choice of the technology and delivery mechanism. It should be kept in mind that management plays a key role in the process and that even when certain technical choices impose themselves, it may be that managerial criteria change the options.

It is not our intention to cover all issues of ICT-based education and training in great detail. What we would like to do is indicate how the choice for a particular learning and teaching methodology and all its implications can influence the cost of network technology or capacity. We do not make a judgement on which methodology is best suited or most effective, this depends completely on the individual situation of each provider. Although some methodologies seem to be more appealing then others at first sight, their attractiveness should not mislead the user from the ultimate goal: enhancing education by means of technology support. Stepping into the pitfall of 'technology support for technology sake' should be avoided at all cost.

Pedagogical Model

New information and telecommunication technologies such as the Internet, multimedia, videoconferencing, simulation etc. allow for innovative or extended experiences for teachers as well as learners. Content providers, institutions, teachers and trainers find new means of getting their contents out to new and larger audiences. Learners on the other hand, have the possibility to gain access to educational resources much less dependent on place and time.

The ways in which this happens are varied: there are tele-courses via broadcast TV and radio, videoconferences between multiple locations, courseware is distributed on CD-ROM or DVD instead of in printed format, courses are delivered via email or FTP, courses are distributed in such a form that a complete electronic or virtual learning environment is created, vast electronic libraries as well as research databases are being made available, virtual laboratories enable learners and researchers to expand their experiments. It should be noted that most often, a complete course or content package consists of parts of each model, combining traditional modes such as face-to-face teaching, lectures, personal tutoring and print-based materials, with technology-supported modes that bring into the media mix elements that can be provided best in an innovative way.

To illustrate the vast choice of options that the content provider has, we would like to provide some examples that indicate how the mode can influence the choice of a particular technology: this list is not exhaustive and does not necessarily relate to the choice of satellite technology only. Our intention is to indicate how intricate the interplay is between all the different elements.

Telepresence classroom

The education mode that closest resembles the traditional classroom, as we know it, is the telepresence classroom in which the teacher teaches remotely to his/her learners by means of camera, microphone and supporting materials. The teacher can see, hear, and interact in real time with his/her students. Students can be miles away, even dispersed over a (limited) number of sites. Videoconferencing, and to a lesser extent audio-conferencing, is one of the most appealing applications of ICT in the domain of education and training, but not necessarily the most successfully implemented. Videoconferencing, gives the impression that a quick and easy transition can be made from classroom teaching and training models towards ICT-based education: setting up a camera and a microphone in the back of the classroom, adding the hardware to transmit the images and sound of the teacher plus the equipment to receive these images and sound at the remote sites, already seem to fulfil a number of requirements to reach remote learners. However, it is often forgotten that videoconferencing significantly adds to the traditional classroom requirements (seats, opening hours, accommodations, etc) a number of technical requirements (AV equipment, telecommunications provisions, technical support, etc). It requires from administrators, teachers and learners a certain level of adaptation: classroom organisation; style of teaching, lecturing and tutoring; format and style of supporting lesson materials; interaction with the learners; all need to change to a certain extent to make videoconferencing a successful application.

ISDN is most probably the best network choice for the time being because it allows for a standardised global interconnection, it is widely available in the developed world, and it has a quality of service level that assures that performance is correctly predictable. Satellite technology and especially VSAT can support videoconferencing effectively especially in those areas and locations that do not have a reliable terrestrial telephone network. The newer type of VSAT systems not only allow for videoconferencing between the different VSAT stations of similar type, but also allow for seamless integration into an ISDN environment, thus allowing connectivity amongst systems in almost all parts of the world.

Depending on the quality requirements, one can choose different levels of compression. Satellite latency contributes to the difficulties of using videoconferencing: not only is there a delay caused by the treatment of the originating audio and video signal (compression and encoding) plus the subsequent processing of the same signal at the receive side for display, but in addition to this delay comes the satellite latency due to its remoteness from Earth, another 0.25 sec. In total, delays can add up quickly to several seconds when the encoding and compression take a long time. It is our experience that with a minimal amount of awareness training, most users will accept a total delay of maximum 1.2 sec. From 0.5 sec delay a 'walkie-talkie' effect starts to emerge. More than 1.2 sec makes normal interaction almost impossible.

In many cases it may not even be necessary to carry out interactive sessions live: it may be just as easy to distribute the audiovisual content in another way (via broadcast TV or radio) and when the time factor doesn't play such a large role, even on tapes, on CD, or DVD. It may be possible to set up an interactive system between learners and teachers and tutors separately from the content distribution mechanism: learners can view and consult the content and ask their question, post their assessments via telephone, email, fax etc.

When the requirements of the AV media quality and the time constraints are not very high, the Internet may be chosen as the way to deliver the content. For videoconferencing, the public Internet is simply not mature until additional protocols such as RSVP and Mbone are implemented and the overall bandwidth availability has increased. Although videoconferencing applications are being used on the Internet nowadays, we clearly see an uptake of home and recreational use rather than mission- critical business or education applications.

When putting in place a private or corporate network solution on which sufficient bandwidth is available to every client (be it over satellite or other links) it is worth considering the use of IP-based videoconferencing protocols. This protocol adopts the transfer technology that is also used on the public Internet but applied in a private controlled network, with much less restrictions and better controllability.

Virtual Learning Environments

With the advent of the Internet as the network of networks from the mid 90s onwards and its acceptance as the standard worldwide for communications between universities, research institutions, administrations, commercial organisations and individual end-users, education providers started to recognise the possibilities of this new medium to not only reach new audiences but also to enhance existing education. This led to the creation of a new model of ICT-based education and training: the electronic or Virtual Learning Environment, integrated digital environments where the teaching and learning activity is organised and in large parts takes place. In other words, these applications are used for delivery of learning content and facilitation of the learning process. They can be used to connect learners and training departments electronically whether at the same location or dispersed over a wider area. Many electronic learning platforms have grown out of communication and collaboration tools with an important additional set of features and functionalities that make them more suitable for training purposes. Almost all platforms currently available are based on client/server architecture. In many cases, the client, located on the user side, is simply a web browser that is used to access HTML pages on the server. Although it is still possible to create and adopt learning environments that are completely stand alone (Computer Based Training and CD-Rom educational software) the vast majority of learning environments take advantage of enabling connections to teachers, tutors, peer learners, administrations, additional resources such as electronic libraries, simulations, the public Internet etc.

To understand what learning environments can do it is useful to consider the functionalities assigned to them.

Layered model of the electronic learning platform

Each of the functions described in the diagram above fulfil a more or less complicated task that has an effect on the other aspects of the environment: production of the learning materials for instance can include the making of elaborate multimedia materials, the building of assessment tools for the learner, the provision of communication mechanisms, libraries etc.

There are many different learning environments commercially available on the market- ranging from simple applications that concentrate on just one functionality, to complicated 'Swiss Army knife' type environments that can perform every possible function the user can think of.

A key aspect of the networked learning environment is its use of some way of connecting the different parties involved: creators, teachers, tutors, learners and administrators. As a consequence, the success of the environment is highly dependent on the selection of the appropriate communications network. Most are making use again of the IP protocol for packaging and transport and allow the user to access the environment using a standard browser. All IP-supported applications can therefore be made part of the environment: conferencing, email, chat, simulation, interactivity, multimedia, etc. The inclusion of some or more of these applications may have quite drastic implications on the requirements for distribution to the end-user, especially when the end-user is remote.

There are two possible ways of approaching this problem. The minimal way is to restrict the functions that one implements in the environment to the accepted minimal level of network capacity for the least served end-user. However, especially in badly served regions, this may result in the elimination of a number of important desirable applications. The maximal way is to build the platform completely according to the pedagogical and technical requirements of the content provider and to impose the inherent network requirements to the end-users. By providing a specific networking solution to support the environment, the content provider has the advantage of complete control over the total environment. This can be achieved best by the implementation of a VSAT type corporate or institution network that spans regions and countries and allows for centralised management and control. The new type of cheap VSAT solutions that are based on DVB allows not only for a certain degree of bandwidth control between hub server and client server but adds to that the multicasting functionality that allows for just-in-time distribution of all kinds of materials including streaming media, software updates, applications etc.

Content

Content can take many shapes as already indicated in the previous part: educational content comes in the form of direct communication (lecturing, tutoring, conversation, discussion) between many actors involved (authors, teachers, tutors, learners, researchers, peers) and supported in many different ways (multimedia, laboratories, exercises, assessments).

In this section we would like to demonstrate how content choice and treatment can influence the selection of telecommunications support.

Linear multimedia content

The production quality of the teaching and learning support materials is a decisive factor contributing to the overall cost of an ICT-based education system. Some types of content require high-quality production standards (eg medical subjects mostly require high resolution and colour images, be it video, graphics or print materials), resulting in a relatively high production cost at the side from where the materials originate, and requiring equally high-quality transmission systems.

We do not intend to go into detail about the requirements for a certain level of production standard, but we would like to point out that the use of TV and TV-like distribution as the medium means that viewers have a certain level of expectation regarding the overall quality: the reference frame of educational TV programmes and videos is not very different from what the viewer is used to seeing on the TV screen and therefore he/she will not so easily tolerate poor quality material.

To give an impression of what bandwidth means in the case of TV and video: a VHS tape of 240 min colour video with an English and a French language channel represents an uncompressed storage of 23.4 Gb on a cassette costing less than US$ 4, weighing less than 300 g, fitting into almost any pocket, playing on an estimated 980 million VHS players installed worldwide5. MPEG-2 digital compression will allow you to store the same video in higher quality on 2 DVD disks, occupying about half the storage space expressed in bytes. Materials cost about the same. DVD players however are only now becoming a standard multimedia device. Shipping to a large number of users will increase the total cost quickly to unacceptable levels: depending on the urgency and the distance of shipment, the cost per item will be from 1 to a few hundred US$.

The worldwide TV household penetration of VCRs is more than 50%. Remarks by Charles Van Horn, President of the International Recording Media Association (IRMA), at REPLItech North America, 20 February, 2001, Los Angeles.

Broadcasting video content via a terrestrial or satellite TV channel will enable large quantities of users to receive the materials in high quality instantly and with very little cost per item, with satellite having the advantage of reaching audiences way beyond national borders. These distribution technologies however, require access to the originating broadcast site: the satellite uplink or the terrestrial transmission centres. The cost is relatively high but when spread over a vast number of potential receivers this cost becomes almost irrelevant. On average this type of cost is in the order of about US$ 2,000 per hour for analogue TV or less than half of that price for digital TV. Adding additional viewers to the receive target group does not affect the cost of the distribution as opposed to distribution by mail. The main distribution cost is entirely imposed on the content provider (who in turn, of course, can opt for schedules such as Pay TV or subscription fees to recover his/her costs). The cost to the end-user is limited to investment in receive equipment and recording and display equipment (depending on the location the receiving antenna may cost from a few hundred to a few thousand US$, the receiver is again a few hundred US$).

The example above illustrates the costing principle for traditional video distribution, which should not be underestimated in terms of outreach, acceptability and impact.

IP over Everything, Everything over IP

In the mid 90s, the ability of the Internet Protocol (IP) to run across virtually any network transmission media and communicate between virtually any system platforms, led to IP's phenomenal success and to its ubiquitous presence on all communication means. The enormous growth of the Internet led to the convergence of the worlds of telecommunications (telephone), broadcast (TV) and multimedia (PC). The ability now exists to author and publish ones own content almost completely digitally (on the desktop) and to compress and package the content in such a way that it can be transported over the Internet. Such a system has a strong appeal to all types of content providers, commercial as well as institutional or educational.

A wide variety of new applications now make use of IP as the packaging protocol, to allow transport over any network, and many applications seem set to become available including telephony or voice over IP. However, this entire overload has made the Internet a victim of its own success.

Increasing the bandwidth - the data carrying capacity of the network - is not sufficient to accommodate the increasing demand. Internet traffic has not only grown dramatically, but it has also changed in character. New applications have new service requirements and, as a result, the Internet needs to change as well.

It all comes down to bandwidth. In an ideal world, all users, content providers as well as end-users, would have unlimited bandwidth at their disposal, wherever they are. Unfortunately, that is not the case and it does not look like it will happen soon. The available capacity now is insufficient (hence the disappointing performance of many broadband delivery systems). Telecommunication and connectivity providers are struggling to keep up with the pace of the demand while being forced by the end-user to decrease the price continuously. To give an idea, it is estimated that in Europe the bandwidth demand increases by almost 10% every month, while the price to the end-user for Cable, ADSL or Leased Line Internet access has dropped more than 50% over the last 3 years.





Continental Interconnectivity Bandwidth Growth

This increase of demand is fuelled not only by the continuously increasing number of users but also by the bandwidth demand that applications are putting on all networks. This is true in all sectors of telecommunications: from mobile telephone to satellite communications. More and more users want to be able to do more while paying the same or preferably even less. Many of the new Internet applications are multimedia and require significant bandwidth. Others have strict timing requirements, or function on a one-to-many or many-to-many (multicast) basis. These require network services beyond the simple 'best-effort' service that IP delivers. In effect, they require that the now 'dumb' IP networks gain some 'intelligence' or that the user chooses an alternative networking solution.

It is most important to select the right type of communications mode that supports the application under consideration most efficiently and cost effectively. Some simple examples:

• There is no point in trying to do videoconferencing over cellular phone networks because of the low bandwidth (9.6 Kbps), and even given future versions of mobile wireless telephony it may not be the appropriate medium.
• ISDN is much better suited for videoconferencing because it is a standard that is accepted and that allows for reliable connections with sufficient transmission quality to allow for low-quality images and sound to be transferred live.
• For high-quality images however, ISDN videoconferencing falls short. When the images need to be sent and received live (synchronously) a distribution medium with high guaranteed bandwidth may be required. This could be satellite when multiple sites in geographically dispersed regions are targeted, or this could be a point to point connection (via microwave link, terrestrial landline, or satellite) when only one receiver is targeted.

When, on the other hand, the images do not have to be displayed live, it is again possible to use a lower capacity and asynchronous network for the transfer.

Network costs

Not only the amount of bandwidth (more bandwidth costs more) but also the chosen frequency band will have an effect on the overall cost: when the system requires the use of C-band satellite communications, it is understood that the cost for installation of Earth terminals will be higher. The choice of satellite position (and its inherent coverage and power) will also influence the cost of the Earth station (eg, size of antenna). Therefore it is necessary to calculate carefully the consequence of selecting the right satellite: paying higher costs for the lease of a more powerful or better located satellite, may pay back in savings made on the ground when installing smaller and cheaper Earth stations.

Operational costs

Choosing a pedagogical model, the technology to support it and a communications medium, always brings a number of related costs that should not be underestimated. Computers in the classroom require for example, maintenance, upgrades, security provisions, initial training, additional peripherals etc. and are more expensive than TV receivers to install and operate. TV receivers have a much longer depreciation time than computers. A videoconferencing room is more than just the videoconferencing terminal: the room itself always requires adaptations that have an important cost consequence as well. It is not unusual to estimate the cost of a technology-supported classroom installation and adaptation to be equal to the hardware technology costs.

Contribution

By contribution we mean the way by which the original content is brought to the place within reach of the end-user: the transmission point. Course providers and content creators need a priority access to the repository where the materials are kept and from where they are distributed. In the case where the content is distributed via IP-based transport, the producers need to have sufficient bandwidth to upload to the servers. In the case of a videoconferencing type application, the originating site may need adaptations in order to enhance the effectiveness of the teaching and tutoring. It is therefore necessary to take into account the specific requirements of the course contributors because they may affect the choice and architecture of the network.

Time

There are many time issues related to technology-supported applications for education. The application may require synchronous (instant) feedback (by telephone), or asynchronous feedback (by email), or no feedback at all. This influences the choice of network application. The use of a satellite service that relies on a geostationary satellite implies a certain latency in the communications path, caused by the distance between Earth stations and the satellite. This delay can be crucial in some applications (such as remote process control), cumbersome (in telephone conversations or videoconferencing) or unimportant (eg, in web browsing or in emailing).

More bandwidth over the Internet is becoming available at a rapidly growing rate. The Internet is still not always the medium of choice. As long as there are no procedures in place that allow for the occupation of a particular segment of the bandwidth for certain purposes, the medium is not the most suitable for the transfer of synchronous content in the shape of audio conferences, videoconferences or live broadcast programmes of sufficient quality. The protocols to allow for these applications are under development but it will take more time before sufficient bandwidth plus the required bandwidth allocation mechanisms are in place.

For mission critical content, that is content that needs to be supplied at a given moment with a guaranteed level of quality and success, the Internet is just not the best choice. The alternative is some kind of private network. Such a network can be provided via terrestrial telecommunication means such as leased line, ISDN, ADSL or other flavours of broadband distribution. Unfortunately, the roll out of terrestrial networks is highly dependent on costly infrastructure works that build incrementally on existing structures. The rather slow roll out of ADSL in the UK and in other technologically advanced countries such as Finland is a clear demonstration of this problem. Satellite communications and especially the different versions of VSAT and broadband via satellite may overcome the long wait and may deliver the user an almost instantly available solution to connect according to his/her requirements.

Satellite network time requirements vary widely. Fortunately the technology adapts well to a range of different requirements: connections that are always on; on at set time intervals; irregular and unpredictable; night-time connections; and more. While connection time may be much cheaper at night, it may not suit the application. However, for data distribution that is not time-sensitive it may be a cost-effective option.

Location

So far we have been demonstrating that satellite communication is ideal to overcome distances. ICT-based education gives the user the impression that time and place are becoming irrelevant. However, careful consideration of the geographical requirements of a network application can avoid high costs or inefficiency. For example, the German Mediadesign Akademie (www.mediadesign.de) in collaboration with a German employment office, uses satellite broadcast to deliver ongoing training courses to over 500 small offices and home offices all over Germany. The teacher/trainer lectures on camera for the largest part of the broadcast, and interacts during the lecture with his/her learners via email and chat. Because the images of the lecture have to be of sufficient quality, because some of the support materials have to be prepared carefully (slides, graphics etc) and because the interaction returning from the remote students requires a certain amount of infrastructure and support, a minimal level of resources is required. These cannot be found in every location of the teaching staff, so a central 'studio/classroom' based in Munich is being used, requiring the teachers to travel from all corners of the country to that location to deliver their teaching contribution.

The design of a distance education network can indeed involve the use of a central location from where the teaching contents are distributed. It may be that for reasons of investment only one classroom is sufficiently well equipped with all the tools and equipment to support the teacher/tutor in his/her work. This can be because the architecture of the network is such that the output of the central location is better than the output of the individual remote sites (eg, because of the use of an asymmetric network such as broadcast TV or radio). In that case, the operational cost of bringing teachers to the central location can be considerable and should be evaluated against the investment cost of installing additional remote contribution sites, or using completely symmetric networks where all sites are equal.

Similarly, when providing web-based learning materials, it may be important to select the physical location of the content server in such a way that all end-users are served with sufficient quality. It may be worthwhile considering the porting of the content to a new server location that is better suited to service the whole user community. This may be a location that is closer to the backbone of the Internet. Economic factors here are rather the availability of the server to all users on the network rather than to the content provider itself. For example, if the content provider has only limited bandwidth to the Internet, it may be better to use this bandwidth to transfer the content even over longer periods, towards a central server that has sufficient access bandwidth in order to serve many clients at the same time. Furthermore, it may be necessary to investigate the need for a Content Delivery Network, that pushes the content to distributed servers that are within an easier reach to the end-user.

Whenever a networked application such as distance education or telemedicine is planned, a geographical plan should be part of the total project, because of the close relationship between costs of investment, network architecture, communications and operation.

Quality

Satellite communication can be tailored exactly to the needs of the users. Bandwidth and coverage are the most important factors of interest. Different levels of service can provide scalable levels of quality of service: fixed and privately allocated amounts of bandwidth or flexible bandwidth allocation whether always-on or available upon request. In addition, because there are no intermediate parties in the communications chain, monitoring, control and management can be done effectively and easily. Compare, however, the costs for VSAT networks with the cost estimate for a two-way satellite Internet connection solution, and it becomes clear very quickly how quality of service (such as guaranteed bandwidth) affects the overall cost.

Quantity

For traditional communications technologies such as TV and radio broadcast, numbers are extremely important: the larger the audience, the more cost effective the broadcast can be considered. This argument is true in public broadcasting where the government pays the cost for all to receive free. It is also the case for commercial broadcasting: viewers pay for a subscription or to receive certain programmes, or advertisement pays for the broadcast so that the audience can watch free of charge.

Quantity has a different significance for educators. It means economies of scale when they can produce once and reuse many times (eg, with good quality multimedia materials) or when they can distribute content (and teach) to many at the same time (eg, via broadcast TV or via the Internet). Communications technologies make it easier to reach more users. Technologies in themselves however, do not always improve the quality of communication.

Here is a simple example to illustrate this: videoconferencing interaction works very well between two sites when there are limited numbers of participants at each side. When the number of participants at each site increases, it becomes impossible to interact effectively with each of them, very much like in a 500-seater university lecture hall where little or no interaction happens between the lecturer and the student at the back row. When more than two sites need to be connected via videoconferencing, costs increase while interactive effectiveness decreases.

In a similar way one should not underestimate the cost of support and tutoring required with web-based learning environments. These can be set up relatively easily and then made accessible to many. Support, tutoring, assessment and follow-up however, increase with the numbers of users. While the basic principle is sound (improving access to learning materials and education) the practical elaboration may require scalability consideration.

Availability

When the satellite system is properly designed at all sides (uplink dimension and redundancy; downlink specification; transponder availability), satellite is able to guarantee an almost unparalleled availability. An important factor is the balance between required availability (expressed in % per year) and installation margins. Availability of 99.5% and more is easily achieved. The weakest link in the chain here is the end-user Earth station which may suffer from adverse meteorological conditions (storms), accidents, lack of maintenance or interference. The availability of satellite communication parallels easily the availability of terrestrial Internet connectivity.

Energy

Power and access to telecommunications are considered readily available commodities in regions like Europe and North America. This is not the case in many other regions. While satellite can bring access to telecommunications surpassing territorial and geographical barriers, energy will still be needed on the ground to receive, process and apply the contents.

Satellite and terrestrial TV are equally accepted and important, but they require power for receiver and TV monitor. Satellite broadband and Internet are no different: their power consumption has become low but is still important enough to cause problems at some locations (for small VSAT-type installations including multimedia PC an average 500 W is needed).

Security

Because satellite communications avoid borders and physical infrastructures such as nodes and exchanges, it is a highly secure system. The only part of the transmission chain that is exposed and is in some way vulnerable is the outdoor antenna. The rest of the communication provision is not dependent on any intermediating elements that can delay, curtail or hinder the communications. The management and control system that is inherent to satellite communications, also allows the network provider to contain or restrict the communications, to set up a completely 'walled garden' or private network infrastructure that keeps the user community safe and secure. All levels of security and privacy can be managed easily and centrally, allowing for content filtering, conditional access, even pay-per-use and accounting.

Hardware installation and maintenance

The use of advanced technologies has a certain element of risk in it because of the lack of experienced support people in the field. In many cases where satellite communications are used for the first time, more initial support is required. Networks based on terrestrial communications on the other hand, are built more incrementally and allow both service-provider and end-user support services to become fully acquainted with all service and support issues.

The selection of a particular communications service (eg broadcast TV vs VSAT vs Internet via satellite) brings a different support requirement. Support for the course and content provider in the case of broadcast TV is almost always extensively organised as an inherent part of the professional TV broadcast and satellite transmission services. The support service at the end-user side is also commonly dealt with in the commercial sector by professional satellite installers and resellers. Worldwide-distributed brands of consumer equipment using common standards make interchangeability and support easy even for the end-user themselves. Because there are no security issues around the use of receive-only satellite installations, installers and technicians do not require high levels of training, and consequently there is no shortage of this expertise and service to the end-user is cheap.

VSAT, at the other end of the scale, requires a lot more specialised support that is network specific. Service engineers and installers are not as widely available as for satellite TV and in some cases they may not even be based in your locality but cover a wider region. Although the technology is very reliable and tested, support and maintenance will be needed and may be more expensive than in the previous example.

Internet via satellite again, is relatively new to either group of support staff. It will probably require the set-up of a new and quite extensive support structure (including remote assistance, help-desk etc). Because this system is aimed at end-users, support structures close to the end-users (local and regional sales and support services) are required. Customers have learnt to expect and appreciate a certain level of support from ICT providers, and they will not compromise for a new service such as Internet via satellite.

Both these last examples require more than ordinary installation skills, because transmission implies security and safety risks on the ground as well as for the transmission path itself. Finding trained staff is more difficult than for the first example.

Depending on the scale and the typical requirements of each application, it may be that the course provider sets up his/her own support network, possibly linked to remote departments, facilities or affiliates. This may represent a considerable cost factor.

Support

An efficient service relies on quality support mechanisms. End-users need to be able to call for help with different aspects of the educational service. The first level of help is directly related to the content or the methodology of the provision and is ideally organised by the course provider and involves the person responsible for that particular course. This is the default support level for the end-user. Calls at this level can be dispatched to the appropriate destination: to the course content specialists, to the tutor or teacher, to the administration or elsewhere. In the case of technology support calls, the call needs to be transferred to the appropriate technical help service: this can be either a local or network specific support service, or the technical support service that is related to the satellite service provider. For example, let us take the case of a student participating in a web-based course delivered via two-way satellite Internet where the terminal is located at a remote campus of a university. If the student encounters a problem with his/her connection to the web site with course materials, he/she first calls the course providers' support service. This service checks if the problems are related to the content, if not the service will pass on the call to a technical support service that is located as close as possible to the end-user and that will diagnose first of all the technical problem. If the problem is related to the satellite network, the local support service will pass on the call to the satellite service provider.

Training

Depending on the selection of the technology and the pedagogical model, the parties involved may require a certain amount of training. It should be clear that very little training is required for the specific satellite part of the chain.

Regardless of the transmission path (terrestrial or satellite), most applications of technology-supported education require some training or at least some raising of awareness. Teaching and learning by means of a web-based learning environment requires some basic skills that are not related to the telecommunications technology but rather to the methodology. It is much the same for producing quality video for distribution on tape or via terrestrial and satellite transmission. Videoconferencing requires some training in order to be effective, and in which there may be some specific reference to the network technology used: every network technology under consideration here will affect the performance in some way, whether it is ISDN, Internet or satellite.

Using satellite communication technology may necessitate specific training for those working at the side of the originating Earth station or the hub server where it may be necessary to make the satellite service provider's staff aware of specific educational issues. At the end-user side there may be a requirement for training in specific hardware-related topics such as the set-up of the Earth station (antenna pointing or software updates). Training can be provided to either the end-user him/herself (eg pointing of a satellite receive dish), or to local or regional support people when the technology requires a higher level of competency or specialised tools that cannot be made available to all end-users.

Sustainability

When considering the total cost of a technology application, all too often only the initial investment cost is taken into account. However, when setting up an ongoing service solution, operational costs including support, network communications costs, replacement scalability and upgrade costs are all important factors. Satellite communications services are perceived as being expensive. Satellite communications however, in general score very well with regard to operational costs: VSAT networks, for example, have a lifetime of more than 5 years before the network requires extensive upgrades or changes. This is because the network is independent of intermediate nodes, and switches and technologies can easily be implemented network wide.

For IP services over satellite, where the competition comes from terrestrial broadband networks such as fibre, a shorter lifetime expectancy is reasonable: with the evolution of fibre technology and the increasing installation of new backbone links to, as well as within, large areas (for example the construction of the Metropolitan Area Networks), satellite connectivity for backbone connectivity to the ISP will become less attractive within a few years. ISPs therefore take only medium- to short-term leases (up to 2 years) for this type of satellite service.

Internet connectivity for the end-user (using small one-way or two-way VSATs) may also have a medium-term life expectancy (3 to 5 years), at least in the areas where, because of the economic and demographic situation, deployment of terrestrial solutions is viable. In remote areas where access technologies will not become available within the short to medium term, satellite Internet may be a solution for at least 10 years. The problem with this type of technology however, is that many operators and service suppliers are relatively unfamiliar with supplying services to the end-user, and the sustainability of this has to be proven. The good news for the end-user is that costs keep coming down, both for hardware and for communications costs.

Revenue

We rarely think in terms of revenue when referring to education solutions. It is probably more appropriate to speak about potential cost savings. This can be achieved by combining the usage of technology solutions.

On the one hand, it is good practice to use the same network and technologies that are already in use (by the course provider itself or by a related institution or course provider). By joining a group or institution that is already using some service, a large part of the initial set-up costs (including piloting and trial costs) can be saved by sharing experiences and enlarging the existing network. Some of the services described in Chapter 5 are certainly open to this type of proposition.

On the other hand it is increasingly common for educational providers to make their technology solution available to others: for example by opening Internet access facilities to other community or even commercial services. A school or a remote campus can, for example, share the infrastructure with a medical service, or make the access available to businesses and individuals outside campus hours. Although this approach requires careful planning and maybe some additional investment (security, accountancy), it may be worthwhile.

Pricing trends

Costs in the ICT world have come down dramatically over the last 10 years. Personal computers are costing less than 15 years ago and can come down much further. Competition in the telecommunications sector has brought the price of a telephone call down. The cost of Internet access has gone down significantly in the decade that it has become widely available to the public.

While this trend towards lower costs will probably continue, end-users are recognising more and more that there is no such thing as a 'free lunch'. The decline of DotComs has clearly demonstrated that electronic services need to adopt sustainable business models. Although free Internet access was very appealing to users, many providers have gone out of business. For professional or mission-critical applications a price has to be paid for a certain level of quality and service.

The same applies to satellite communications. While objectively speaking the costs have come down, a threshold seems to have been reached. Prices in the satellite communications industry have reached a level that can be compared to terrestrial services such as ADSL or leased lines. Service suppliers in any of these technologies will watch each other's price settings in order to stay competitive. When new technologies or players enter the market, they will again set their prices according to the service levels and quality guarantees they are able to deliver.

Regulatory issues

Satellite communications make use of a part of the radio spectrum, which is a valuable resource, shared by many different types of applications and users. In order to avoid conflicts, abuse and interference between users, but also to watch over public safety and health issues, some form of regulation is required. This regulation is handled by national or regional institutions that apply international recommendations, agreed in organisations such as the International Telecommunications Union (ITU). Some freedom does exist however within ITU recommendations, for each regulatory body to adopt the procedures according to local requirements. For the sake of simplicity we will in this section address VSATs in the generic meaning of the word: 'Very Small Aperture Terminal', any fixed satellite terminal that is used to provide interactive or receive-only communications.

Licensing problems

While satellite communications offer immediate cost-effective solutions, some countries' policies, rather than facilitating satellite communications, hinder or prevent their deployment. In many countries regulation procedures are outdated, expensive or cumbersome.

See a full overview of licensing in Europe
In some countries, the national public telephone operator is the only entity that may install and service, or even own, operate and maintain VSATs. In other countries a local commercial presence is required by administrations as a precondition for licensing.

Licensing fees also remain too high in many countries. Furthermore, some regulatory bodies apply additional taxes, annual operator fees, high customs tariffs restricting importation of VSAT equipment, and tariffs be paid to the incumbent Telecom Operator - even if they do not participate in the service chain. The accumulation of too many fees tends to be prohibitive for many VSAT applications.

Furthermore, many countries still apply unnecessary burdensome licence application processes resulting in unnecessary delays issuing regulatory licences. Sometimes, existing Earth-station regulations are geared to the broadcast industry and do not contemplate current uses such as data, Internet, and private voice networks. Some countries enforce zoning restrictions that prevent the installation of rooftop VSATs.

Some administrations require type approvals for antennas, even though the antenna type is already being used for the particular satellite system being requested.

While there are more than 500,000 VSATs operating in most of the world's countries, many of them are precluded from international applications. This is an unfortunate waste of resources, because VSATs are ideally suited not only to provide domestic connectivity, but also to offer trans-border communications for wide-area networks. On the national level, VSAT rules are often neither transparent nor accessible to the general public. Further, these rules are often difficult to interpret. On the regional level, service providers are required to seek out a multiplicity of application forms - as well as contact details for the officials responsible for processing them - among the jurisdictions where they provide services.

In general, it has become apparent that the more regulations, fees and other requirements that are imposed on providers of VSAT systems and services, the fewer communications options will be provided in the individual country.

Licensing perspectives

The Global VSAT Forum (GVF), which acts as a representative body for the VSAT industries, has developed regulatory recommendations and guidelines as a tool for regulators and policy makers, who are interested in modifying regulation to facilitate the use of VSAT-based services. Regulators around the world are already taking these factors into account and are implementing new policies that facilitate the use of VSAT systems and services. For example, the regulator in the UK has adopted a one-stop- shop procedure that guarantees short turn-around times for applications.

In order to advance furthermore the uptake of satellite communications, the GVF encourages the elimination of licensing and monopolies to wipe out sub-standard services offered at above-market prices.

Conclusion

It should be clear from this chapter that a network solution for education, be it via a satellite communications network or via any other distribution means, has many different cost elements. Very often, decision-makers find themselves blinded by the cost of distribution. It is essential, however, that the contribution of these costs should not be overestimated. The first and foremost cost is always the cost of creation, collection and adaptation of the content and the cost of tuition and support. It is commonly observed that the cost of production of the content is at least five times the cost of distributing the content. The latter cost is only marginally affected by the medium itself.

The value of the distribution medium should be evaluated correctly by looking at the direct costs (and every education provider will agree that these are costs that are to be avoided as much as possible) as well as looking at the opportunities the technology brings to the education system. The possibility to widen the audience, to reuse materials, to better use staff resources and to improve the teaching and learning experience are factors that cannot easily be expressed in financial terms.

Back to contents