MARIA - A2123

Multimedia Applications for Regional and International Access




DATE: 1 September 1994






This report is the first deliverable of the study of available cancer information services (WP11). It focusses on publically available internet services in cancer and describes not only the information services, but also the tools to access the information. This study has so far resulted in 18 cancer information services to be included in TeleSCAN.

This report has been compiled by the EORTC/MARIA group - NKI - Amsterdam


P-1                    10-Aug-94       First draft                             
P-2                    14-Aug-94       comments Visser & Dalesio               
P-3                    22-Aug-94       comments Jenkin                         
V-1                    1-Sep-94        Submission to AIM                       

This sheet provides a record of changes made to the document. The final document has been approved by the president of the EORTC, and sent to the CEC as deliverable 9 of Work Package 11.

All deliverables of the EORTC partner are subject to improvement and subsequent modification during the course of the project. The final EORTC project document produced at the end of the project will be a collection of all reports written by the EORTC partner during the MARIA project.

All documents produced by EORTC in the MARIA project can be accessed over Internet with (X)mosaic as of October 1. Choose `file', `Open URL' and type `'.

Any correspondence concerning this document should be sent to:

Emile van der Donk

Biometrics Department

Antoni van Leeuwenhoek huis

Plesmanlaan 121

NL-1066-CX Amsterdam

The Netherlands

Tel no: +31-20-5122665

Fax no: +31-20-5122679


This document has been written by:

Emile van der Donk - EORTC

Tobias Kuipers - University of Amsterdam

This document was reviewed by (to date):

Otilia Dalesio, Head, Coordinator EORTC MARIA team

Sjoerd Visser, EORTC - MARIA team, WP11

Linda Jenkin, EORTC - MARIA team, WP13


Prof. J.G. McVie, president of the EORTC


1. Summary 4

2. Background 5

2.1 Introduction 5

2.2 What is MARIA 6

2.3 EORTC and its role in MARIA 7

2.4 EuroCODE 8

2.5 The European demonstrator: TeleSCAN 8

2.6 Relevance to TeleSCAN 9

2.7 About this report 9

3. Technology 10

3.1 Networking basics 10

3.2 Workstation basics 10

3.3 Networked services v.s. non-networked services 11

3.4 The Internet services 12

3.4.1 Gopher 12

3.4.2 The World Wide Web 15

3.4.3 Usenet 19

3.4.4 Electronic mail 21

3.4.5 Other services 22

3.4.6 New developments 23

4. Internet services relevant to TeleSCAN 24

4.1 Introduction 24

4.2 Evaluation criteria 24

4.3 Cancer related gopher services on Internet 25

4.4 Cancer related World Wide Web services on Internet 38

4.5 Relevant Usenet newsgroups 40

5. Conclusions 41

1. Summary

2. Background

2.1 Introduction

Much excellent work has been undertaken in recent years in the use of information technology (IT) to help improve the provision of health care services and to promote the sharing of knowledge and information. However, it is still true that the uptake of IT as an effective tool is still not widespread amongst medical practitioners and staff.

Several factors now combine to present an opportunity to develop telematics systems useful for the busy medical professional:

- the emergence of cost effective workstations that support multi-media,

- more widespread availability and use of advanced communications infrastructures such as ISDN,

- the development of more user friendly software (Windows based),

- the worldwide use of standard tools for exchange and dissemination of information over International networks (Gopher, Mosaic).

There is much interest in the Oncological community in extending the available computer services.The MARIA project is intended as the beginning of a major network serving cancer centres throughout Europe. It builds on the existing EuroCODE network, and incorporates horizontal links into other medical disciplines of paediatrics and molecular biology.

2.2 What is MARIA

MARIA is an acronym for Multimedia Applications for Regional and International Access. The MARIA project is a DG.XIII / AIM funded project which aims to assess the current exploitation of Telematics by medical professionals and to demonstrate how multimedia applications can contribute to the daily work of the medical professional.

The project is two-fold: a regional part and a European part. The regional part consists of a demonstrator in Paediatrics. The European part consists of a series of studies on existing Telematics services and infrastructures, on ongoing projects of potential interest, organisational and security aspects of a European Cancer telematics network, and a large study to determine the requirements of the Oncological community, leading to a demonstrator of Telematics services build on the existing EuroCODE network.

The present report is prepared by the EORTC partner, which, through its Telematics team in Amsterdam, Netherlands, is responsible for the European part of the project.

The MARIA project started in December 1993 for a duration of 18 months. The EORTC partner has been allocated 166 KECU, which are mainly used for labour.

More specifically, the EORTC is responsible for the following workpackages.

WP     WORK PACKAGE TITLE                                                     
1      Review of AIM and related project results                              
11     Study of Available Cancer Information Services                         
13     Review of Target user groups for European Cancer Telematics Services   
14     Network security and privacy legislation                               
15     Analysis of the technical, organisational and financial aspects of     
       the Multinational implementation of a European Cancver information     
       telematics service                                                     
16     preparation of EuroCODE to function as a platform for the European     
17     Implement and Evaluate a European Demonstrator for Telematics Cancer   
       information services                                                   
18     Software development for a register of trials                          
19     Technical coordination of the European Demonstrator                    

2.3 EORTC and its role in MARIA

The EORTC, European Organisation for Research and Treatment of Cancer, conducts, develops, coordinates and stimulates research in Europe on cancer. The structure and organisational setup of the EORTC, being a European wide non-profit organisation with an extensive structure of collaborative study groups in practically all fields of Cancer, and with strong links to National groups, leagues and institutions, is ideal performing the tasks of the MARIA project on a European scale.

The EORTC and OECI (Organisation of European Cancer Institutes) have already agreed to closely collaborate and take a leading role in mobilising the oncological user communities in moving towards the development of a European `Cancer Centre without Walls'

In the EORTC part of the MARIA project, the main goals are to obtain a clear view of what is already available and what the Oncology community in Europe would expect from an International Oncology information network, and to demonstrate to the actual medical professional what kind of applications could be developed and / or implemented to support his or her daily work.

2.4 EuroCODE

EuroCODE (European Computerised Oncological Data Exchange) is a project launched in 1988 by the EORTC under the auspices of EC. It is a European wide oncology information network that currently serves a population of 500 physicians and staff across Europe and offers general communication facilities, such as electronic mail, and specific software for cancer clinical trials. It is supported by 8 National nodes, based at centres of excellence in cancer research in member states.

As a data collection tool for clinical trials, the EuroCODE network has been very successful, and currently, over 40% of patients entered into EORTC trials are registered through EuroCODE. Currently users of EuroCODE services are mainly data managers and junior oncologists.

The current project will build on EuroCODE to provide a basis for a comprehensive European Oncology Telematics network to the service of a broad range of potential user groups, including oncologists, general physicians, general practicioners, medical trainees, nurses, and eventually patients, families and the lay public.

2.5 The European demonstrator: TeleSCAN

Many valuable computer tools for cancer research, treatment, education, screening and prevention have been developed in the past decade. Parallel to this, the Internet community has produced generic software tools for the exchange and dissemination of information in a worldwide accepted and technically outstanding fashion.

TeleSCAN, the European demonstrator within the MARIA project, aims at the synthesis of these achievements, and by providing these tools to the oncology community, mobilising a critical mass of users for a future European telematics oncology network.

In TeleSCAN, 4 types of telematics services will be demonstrated:

- public access to the Internet through standard tools: Electronic mail, remote login, file transfer, Usenet, and gopher- and Mosaic-client,

- public access Gopher and Mosaic servers of EuroCODE, offering information specific to the EuroCODE node,

- Special European developments, such as the Randomisation program, Address Book for Cancer, List of Meetings and the Register of Trials,

- Commercial products, such as Video conferencing.

TeleSCAN could be used as a showcase of telematics services in Cancer conferences, meetings and workshops. Commercial and non-commercial telematics service providers could be targetted to use TeleSCAN as an entry point for marketing their oncology related Telematics product. Attempts to organise activities within this framework are currently being undertaken by the president of the EORTC.

2.6 Relevance to TeleSCAN

TeleSCAN should show the benefits and possibilities of multi-media Telematics to the European oncologist. Multi-media means incorporation of live video, sound, images, text and databases in a common application. In addition, the information presented within these applications should be complete and relevant as to speak to the imagination of the European Oncologist.

The aim of this survey is not only to highlight services which may be of benefit but also to present them in a way that makes them easily accessible to the users of TeleSCAN. The work for this workpackage has been organised as to allow the incorporation of services already accessible through the Internet in a first demonstrator setup. Indeed, the services described in this document can be accessed through the Dutch TeleSCAN / EuroCODE node ( as of October 1.

2.7 About this report

This report is the deliverable resulting from the first part of workpackage 11 of MARIA: Study of available Cancer information services. The main product of the EORTC effort in MARIA is a demonstrator of Telematics services for the Oncology community. Many of these services are actually available over International networks, in particular the Internet. These services make use of off the shelve public domain software to access publically available information on cancer. The Internet services in general, and those related to Oncology in particular, are the main subjects of this document.

The Internet is the playground for information technology novelties. Although information technology is a fast evolving field, the Internet, despite its enormous growth, only 1% of the worldwide population can access it. Therefore, chapter 3 gives an explanation of networking and an introduction to 'cyberspace' to the newcomer.

A single search on the Internet of the word `cancer' resulted in over 4,500 different hits, which have all been reviewed for quality and interest. Both good and bad publically accessible information resources were found. A selection of 18 well organised telematics services related to Oncology available on the Internet are described in chapter 4. The conclusions in chapter 5 provides a list of services that will be actually be incorporated into TeleSCAN.

In addition to the publically accessible services on the Internet, many Telematics services have been developed to run in a stand alone or system specific environment, for instance CD-rom applications, special Multi-media systems and DOS (PC) based programs. This type of services will be addressed in the second report within this workpackage, due in May 95.

The present document has been setup not as a dry report but as a guide that can be distributed to potential users, namely the physicians. It should be able to serve the experienced Internet users through the oncological services, as well as to take by the hand those not familiar with Internet services and guide them through `cyberspace'.

3. Technology

3.1 Networking basics

A computer network consists of a number (larger than 1) of computers, connected with each other by communication lines. These communication lines can be local, for instance within a single office, which is referred to as `Local Area Network'. As an extension to this, Local Area Networks can be interconnected over a long distance using for instance sattelite links, leased lines or ordinary telephone lines provided by Telecommunication industry.

The main purpose of computer networks is sharing of resources, such as information, programs and printers. Kathalysed by the worldwide adoption of a vendor independant communication protocol for computers, over the past two decades many local area networks have been created and interconnected into Wide Area Networks, which in turn have been interconnected to form a global network: the Internet.

A connection to the Internet requires a data carrier, which is in general a public telecommunication service such as ISDN, X.25 or leased lines. These topics are outside the scope of this document. The most important characteristics of the data carrier is speed and price: high speed means high price.

Through the generic tools developed for the global internet, information that physically resides on a system at one end of the world becomes transparantly available to the desktop of anyone connected to Internet at the other end of the world. Being mainly text based facilities in the late eighties and early nineties, these tools have been extended to include Multi-media facilities over the last years.

3.2 Workstation basics

Because hardware keeps getting cheaper, the use of multimedia computing is no longer reserved for large research organizations. At the moment more and more microcomputers incorporate some form of multimedia.

Multimedia means here, the integrated use of video, audio and text, preferrably over a network. In practice this means that a microcomputer needs some form of a graphical user interface (GUI). The three most common GUI's are MS-Windows, X-Windows and Macintosh. The computer needs to be sufficiently fast to perform all (complex) multimedia tasks at the same time. This means that the CPU must be either a Intel 80486 or a Motorola 68040 or faster. For the use of both sound input and output, a PC will need a SoundBlaster compatible soundcard, a Macintosh will need a builtin microphone.

Because of the enormous amounts of data involved in multimedia computing, a computer will need a CD-ROM player. For the use of multimedia over a network a high speed link is needed, capable of running one of the well established protocols serving multimedia applications.

Non PC-based UNIX workstations usually have all these facilities builtin, but are in general quite expensive. Since most companies develop their UNIX workstations for research environments, the emphasis is on speed rather than on user-friendliness. The UNIX operating system has a long tradition in the reasearch world, but has a very steep learning curve. Examples of such workstations are Sun Microsystems's SPARC, Sillicon Graphics's Iris and Digital Equipment Corporation's Alpha. The development of the PowerPC by IBM, Motorola and Apple Computer is a development to watch in the multimedia world. Right now the cheapest PowerPC based computer offers the userfriendliness of a microcomputer at twice the speed of the cheapest computer by Sun, at half it's price.

3.3 Networked services v.s. non-networked services

Telematics is the combined use of Telecommunication and informatics. Thus, Multi-media telematics could be described as applications combining live video, images, sound and data running on telecommunication networks.

Many cancer information services have been and are being developed to support the clinicians in their daily work. From the point of view of TeleSCAN, these services can be divided in two categories: networked services, which can be accessed over infrastructures such as Internet, and those running on specific, stand alone or only locally networked systems.

Most of the cancer information services are available for specific systems, in particular the DOS and Windows environment. The advantage of these systems are fast access, short software development time and ease of use. Major drawback of services based on these platforms is that they cannot easily be ported to so-called multi-user network environments. In effect, these services only be included in TeleSCAN when they are rewritten to run on Unix or Vax, which is often not feasible or desired.

TeleSCAN will include a directory of these non-networked services. This directory will provide information on the service, and give the possibilty to contact the responsible persons of these services through electronic mail.

3.4 The Internet services

3.4.1 Gopher

The gopher information service is a client-server based menu driven protocol for information delivery. Gopher is a powerful system that allows a user to access many of the resources of the Internet using one consistent manner. Everything within Gopher is done in a menu. Whatever happens when a user selects a certain menu-item, representing a text-file, is transparent to the user. That is, the user will just see the text file appear on his screen, he will not be bothered by a command language and cryptical keywords.

Using a gopher client, a user can access a gopher server. All gopher servers combined are referred to as "Gopherspace" (analogous to the omnipresent "Cyberspace"). The entry point on a gopher server is called the root-menu. From this menu he can access all the information on the server, and on other servers. Depending on the type of gopher client he can either "click" his mouse on a certain menu item, or use the keyboard to navigate through all the items.

A menu item is a "link" to another information source. This is either a file, a searchable index, a telnet session or another menu. If it is a file, it could be either a text file, a sound file or a picture (graphics) file. If a user selects such a link, the client program will try to get the appropriate file to the user, either by displaying the text file, playing the sound through the computer speaker, or displaying the picture on the screen. If for some reason it is not possible to use the file in the appropriate way (e.g. because the user's computer does not have a speaker, or cannot display sound files) it depends on the gopher client program what action is taken. Usually the program will save the file to local disk (to the user's computer) for the user to decide what to do with it.

If a menu item is a link to another menu, the new menu will be displayed in exactly the same way as the previous menu. (Same font, same colours, same layout). Selection of menu items is handled the same way as in the previous menu.

If the menu item is a telnet session, the telnet session will be opened, after displaying a warning that the user is about to leave Gopherspace. For most telnet session a user needs a special accountname to login to the telnet host. Most gopher client programs will give the user a message telling what accountname needs to be used.

An example Gopher session. First we start our local gopher client. The client usually makes a connection to a default server. Here it is our own server (which was under construction at the time of this example). This is the first screen we see:

(The layout and colour configuration is largely dependent on the type of client program that is used. We used Hgopher for MS-Windows.)

From this menu we can select any item by clicking on it with the mouse. We choose for the Cancer Information Services. We can see it is a link, because there is a small arrow in front of the menu item. Again, in other client programs this arrow may be some other symbol. The next screen we see is:

The menu items with a small computer monitor are Telnet session, as discussed above. We now choose OncoLink. It is a link, hence the arrow. We then see:


The second menu item here is a "search" item. On this server that means a we can enter a word (or a combination of words) and the server will then retrieve all documents containing that word (or combination of words). We, however, choose What's NEW on OncoLink.


The items with the small pair of glasses in front of them are text files. We choose the sixth menu-item: Meeting Update.


This is the text file. The gopher client calls another program to display the text. In this case, the program "Kladblok" (or scrapbook) is called (We are running a Dutch version of Windows). We can now manipulate this text, we can print it, we can save it to disk, or we can mail it to someone. The file we see in "Kladblok" is actually a copy of the file that is on the OncoLink server. We cannot change the text file on the OncoLink server.

3.4.2 The World Wide Web

The World Wide Web has many names. It is also referred to as WWW or W3 or W3. The WWW project, started by CERN (the European Laboratory for Particle Physics) in Geneva, seeks to build a distributed hypermedia system.

To do this it uses the notion of hypertext. Hypertext is similar to normal text, but some words or sentences are marked as a link (usually by giving the words another colour, or underlining them) to another hypertext file.

The advantage of hypertext is that in a hypertext document, if a user wants more information about a particular subject mentioned, he can usually "just click on it" to read further detail. In fact, documents can be and often are linked to other documents by completely different authors. This is much like footnoting, but the referenced document is instantly accesible.

To access the web, a user will need a browser program or WWW client. The browser reads documents, and can fetch documents from other sources. It then shows this document in the appropriate way for both the type of document (this could be just about anything, ranging from plain text files, hypertext files, sound, or high resolution pictures) and the type of computer. The browser knows if the users computer can display pictures, can play sound, and so on. This is much like how a gopher client works. Information providers set up hypermedia servers which browsers can get documents from.

The browsers can, in addition, access files by FTP, NNTP (the Internet news protocol), gopher and an ever-increasing range of other methods. On top of these, if the server has search capabilities, the browsers will permit searches of documents and databases.

The documents that the browsers display are hypertext documents. Hypertext is text with pointers to other text. The browsers let a user deal with the pointers in a transparent way. If he selects the pointer, the document that is pointed to is presented.

Hypermedia is a superset of hypertext -- it is any medium with pointers to other media. This means that browsers might not display a text file, but might display images or sound or animations.

The browser navigates through the World Wide Web by means of so called URLs. URL stands for "Uniform Resource Locator". It is a draft standard for specifying an object on the Internet, such as a file or a Usenet newsgroup.

URLs look like this: (file: and ftp: URLs are synonymous.)

r file://



r news:alt.hypertext

r telnet://

The first part of the URL, before the colon, specifies the access method. The part of the URL after the colon is interpreted specific to the access method. In general, two slashes after the colon indicate a machine name.

How to acces a specific URL depends on the browser that is used. Browsers used on windowing systems (MS-Windows, X Windows, Macintosh) usually include a menu item in the file menu saying: "Open URL". Other text based browsers such as Lynx or the simple line mode browser use keywords or keystrokes to open a certain URL. The exact operation is documented with the program.

WWW is a superset of all existing Internet resources, and is designed to stay that way. This means that any new Internet resource that might be invented can and will be easily included in the World Wide Web and its client programs. Thus making sure that a consistent and uniform interface exists for all different sources of information on the internet.

This is an example of accessing the World Wide Web. We will use a browser called NCSA Mosaic for X Windows (or XMosaic for short) in this example. It is a browser for X window systems. It automatically jumps to a user-defined "home-page" or top level source. Ours is OncoLink(tm), The University of Pennsylvania Multimedia Oncology Resource.

The top part of the window shows all sorts of administrative data. The document title, the document URL and the menus. Only part of the actual document is now shown, hence the slider bar on the right. Next to the OncoLink logo is a small picture of a speaker. If we click on this we will hear a voice out of our computer's speaker welcoming us to OncoLink. All words that are underlined are so called "links" as discussed above. We choose "Search OncoLink using a keyword".

We can now enter a keyword in the text field and hit enter. We choose osteosarcoma as a keyword.

We can now select any of the underlined sentences to read that article. We choose NCI/PDQ Osteosarcoma - physician information.

This is an article from the National Cancer Institute's Physicians Data Query. It is a plain text file, not a hypertext file. That's why there are no words underlined. Some files do not adhere to the size of our window. This one does not, apparently. We can either use the slider bars or increase our window size to read the whole file. We can also click the Save As... button on the bottom of the window, and save the file to our disk, for further reference.

3.4.3 Usenet

Usenet or NetNews or just News, is a large collection of articles transferred over the Internet using NNTP (or Network News Transfer Protocol). These articles are grouped together in so called newsgroups.

A newsgroup is a collection of articles discussing the same subject. Examples of such newsgroups are and There are many groups that are of local interest only. The group nlnet.markt.banen for instance, discusses job offers in The Netherlands only. A lot of groups are of global interest. Just about every topic has its own newsgroup.

Group naming conventions follow certain guidelines. A certain hierarchy exists. The groups distributed worldwide are divided into seven broad classifications: "news", "soc", "talk", "misc", "sci", "comp" and "rec". Each of these classifications is organized into groups and subgroups according to topic.

"comp" Topics of interest to both computer professionals and hobbyists, including topics in computer science, software source, and information on hardware and software systems.

"sci" Discussions marked by special and usually practical knowledge, relating to research in or application of the established sciences.

"misc" Groups addressing themes not easily classified under any of the other headings or which incorporate themes from multiple categories.

"soc" Groups primarily addressing social issues and socializing.

"talk" Groups largely debate-oriented and tending to feature long discussions without resolution and without appreciable amounts of generally useful information.

"news" Groups concerned with the news network and software themselves.

"rec" Groups oriented towards the arts, hobbies and recreational activities.

These "world" newsgroups are (usually) circulated around the entire Usenet -- this implies world-wide distribution. Not all groups actually enjoy such wide distribution, however. Some sites take only a selected subset of the more "technical" groups, and controversial "noise" groups are often not carried by many sites (these groups are often under the "talk" and "soc" classifications). Many sites do not carry some or all of the comp.binaries groups. There are groups in other subcategories, but they are local: to institutions, to geographic regions, etc. and they are not listed here.

A typical newsgroup name consists of a number of words, seperated by dots, with each consecutive word decreasing the number of subjects suitable to this group. For instance means that it is computer related (comp), it is about systems (sys), but only systems manufactured by IBM (ibm) and specifally their Personal Computer (pc).

Two types of newsgroups exist. Moderated newsgroups and non-moderated newsgroups. Moderated newesgroups have a moderator who screens all articles before they appear. If he or she decides that an article is not suitable for this newsgroup, it will not appear, but will be resent to the author instead. This is done to reduce traffic to the newsgroup and to filter out junk articles. Groups that do not have a moderator are absolutely open to anyone. Everyone who can read articles in such a group can post an article to the group, and it will appear.

Not all groups are carried by all hosts. The news administrator can choose not to include a certain newsgroup in his distribution. Anyone can start a new newsgroup. Since there is no controlling authority on the Internet a new newsgroup will originate if a (sufficiently large) number of people think this new newsgroup is necessary. Proper procedures for starting a new newsgroup include a vote among the internet population. After a certain period of discussion (usually two weeks) a vote is held. If the yes-voters outnumber the no-voters by 100 or more, the new group is installed. All News administrators (in general) will include the group in their distribution.

Usenet was conceived to display technical notes and news items in 1979 at the University of North Carolina. Originally it involved just two computers. It outgrew its original purpose soon, and is now a huge collection of discussion groups, and, technically speaking has not much to do with News anymore. It is still referred to as the News or NetNews.

Some usenet statistics. In July of 1994 the usenet transported an estimated 72,755 articles per day. That amounted to 189 megabyte per day. Usenet consisted of 190,000 sites, with 22,920,000 user. Of those users, 7,138,000 have read at least one usenet article in that month.

To use Usenet a user needs a client program that is called a newsreader (not to be confused with the human newsreader). The newsreader sorts all the different articles to the right newsgroups and lets the user subscribe to the newsgroups of his liking. This way, he will never get any groups he is "unsubscribed" to. This saves a lot of time. If a new newsgroup is set up, it will automatically be included in the list of subscribed newsgroups, until the user specifically unsubscribes himself. This way, a user will never miss a new newsgroup.

Subscribing and unsubscribing are somewhat misleading terms here. There is no central subscription mechanism. Only the local "newsreader" knows about subscriptions. Apart from normal Internet connection charges, there is no extra fee for using Usenet, as there is no discount for not using Usenet.

Because Usenet is so large, to get the complete news distribution a machine needs a large amount of diskspace. Depending on how long the articles should be kept at the site, a machine now needs about 1 gigabyte harddisk space to keep all the articles on the machine for a month. It is, however, very workable to keep articles on the machine for only a week or ten days. Only 250 megabytes would then be needed. Further space reduction could come from only selecting part of the complete News distribution.

3.4.4 Electronic mail

Electronic Mail is by far the most used service on the Internet. Some people actually confuse the Internet with electronic mail. Internet electronic mail (or email for short) lets anyone on the Internet send a message to anyone on the Internet. To do so, a user needs to have a so called email address (or Internet address, hence the confusion). An email address is very much like a regular (Post Office) mail address. It consists of a user identification, usually the user's name, or an abbreviation of a name, an ``at'' character (@), and a more general machine address. For instance the email address of Tobias Kuipers at the University of Amsterdam is where kuipers is obviously the name, and is the machine address. The machine address is usually an abbreviation of an organization or company name. All email addresses of IBM employees end with, for instance. The last part of the address is well defined. Email addresses in all countries except the USA end with a two letter country code. nl is The Netherlands, be is Belgium, fr is France and so on. In the USA all email addresses (for historical reasons) end with one of .gov, .com, .edu, .org for government, commercial, educational or non-profit organizations. Then there is the generic .net extension for all higher internet authorities (such as main, so called backbone servers or other general services).

Email is transported over the Internet using SMTP or Simple Mail Transfer Protocol, a protocol describing how a mail message should be composed, what it should look like, and how it is sent. A user never addresses SMTP directly but instead uses a small program called a mail agent. Examples of these programs are mail, elm or pine. Users that are on a local area network connected to the Internet, and are using a PC or a Macintosh will mostly use another protocol called POP or Post Office Protocol. This is a protocol designed especially for local area networks. Very often when a user types a mail message at his PC it is transferred via POP to the nearest (Unix) server, and from there to the Internet address it is directed to. The main difference (from a user's point of view) between POP and SMTP is that it is not possible to use the same mail agent for both protocols. Thus, PC or Macintosh users cannot use any of the programs above, but will use programs such as Eudora (on the Mac), NuPop and Popmail/PC (on the PC).

With the advent of multimedia on the Internet, one service legged behind: Electronic Mail. Because of the many different types of binary data (The binary representation of, for instance, the same file in WordPerfect or MS Word is completely different) SMTP only knows how to send text charcters or ASCII text, to be more precise. ASCII is a very well defined, widely accepted encoding scheme for characters. It is the only encoding scheme that is the same on 99% percent of all computers. Therefor, to mail some form of binary data, such as graphics, sound, or video, a scheme is needed to encode this binary data as text. Many different encoding schemes exist, and no one system is standardized. This is a nigthmare for the casual user and has been slowing down the use of multimedia correspondence quite considerably. There now is a draft standard for transporting multimedia mail. It is called MIME for Multimedia Internet Mail Extensions. This a package for encoding binary data in a standard, transparent way. This means that a user does not have to know anything about how this binary file is encoded. Unfortunately, this system is not yet very widely used. It is yet unclear if MIME will become the de facto standard for sending mail on the Internet

3.4.5 Other services


Ftp (File Transfer Protocol) is a protocol for transferring files over the Internet. All machines on the internet should honour this protocol.

Ftp refers both to the actual protocol, and the program to transfer the files. Retrieving a file through ftp from another host is subject to user authenticization. Only users with a password on the remote system can transfer files. Another form of ftp is so called "anonymous" ftp, where a user can retrieve a file from a remote host, without having an account on that machine. This is the most widely used system for file exchange on the internet. Most anonymous ftp sites specialize on a certain subject. The number of anonymous ftp sites is unknown, but is estimated to be in the tens of thousands. Millions of files can be retrieved from these sites.

Files on anonymous ftp sites span all possible subjects. Examples are: pictures from weather-satellites, scientific articles, all sorts of software, government publications.

Because of the large number of files, and sites, it is often hard to find a certain file on the Internet. A service called "archie" more or less solves that problem. A user can log onto a so called archie site, and query it's database. This database consists of all the files that are "ftp'able" within a certain geographical region (Europe, for example). Archie scans this region each night, and makes sure it's database stays up to date. Thus, even if a user remembers only part of the filename, it is relatively easy to retrieve.

Internet Relay Chat

Internet Relay Chat (or IRC) is a protocol for conversation over the Internet. As its name suggests, it is a number of messages, relayed (asynchronously, or "real-time") over the Internet. This gives a user the opportunity to engage in a real discussion via his keyboard.

A IRC client program is needed to connect to an IRC server. When a user conncts, the server will ask him which one of the current sessions he would like to join. Most of the sessions on an IRC server at any one given time will be informal discussions about whatever comes up. There are some sessions (or channels, as they are called) that discuss specific topics. A user can, however, at any given time, create his own session. This can be very useful when working with a number of people over great distances. A group can agree to meet, say, every second friday at 4pm on channel #cancer (all channel names start with a #). One of the users in the group creates the channel, and as a result, will have permission to grant access to other users, thus being able to keep the discussion private.


Talk is a protocol for message delivery, very much like the Internet Relay Chat. The difference between Internet Relay Chat and Talk is that Talk is private by definition. The other difference is that two user do not have to agree on a time before hand.

When a user wants to "talk" with another user he invokes the talk program with the other users Internet address (usually the same as his email-address). If the receiving is logged on he will see a message on his screen saying: "User XXX@YYY wants to talk with you", and he can then choose to reply, or ignore this message. If he replies, everything he types after the reply message will be displayed on the screen of the sending user, and vice versa.

The original Talk program limited the number of people in a conversation to two, (hence the popularity of IRC, with an unlimited number of discussion members). A new implementation, called Ytalk, overcomes this limitation, as added a lot of new, very useful features. The most noteworthy is that a user now can decide to save the whole conversation (or parts thereof) to a file. This removes the need for someone to take the minutes.

3.4.6 New developments

A relative newcomer on the Internet is videoconferencing. So far only one application exists that makes use of the Internet network protocol for sending live video. The program is called CU-SeeMe and is developed at Cornell University (hence the CU). It is still in beta testing, but the first results have been encouraging. At the time of writing, we have not had any hands on experience with this system.

4. Internet services relevant to TeleSCAN

4.1 Introduction

This study has resulted in a large number of `demonstrators' of telematics, and a even some Multi-media telematics in Cancer. One of the advantages of gopher and mosaic services is, that a menu of services can be locally setup to provide access to services in remote places. The preliminary results of this study, although only viewed briefly and in an experimantal setup, very much appreciated by several clinicians that coincidently walked into the room with the test system.

This chapter presents the results from this study, and services that will be available in TeleSCAN.

4.2 Evaluation criteria

The purpose of the study was to find in the Internet all services that had anything to do with cancer. This included gopher and mosaic sites, Usenet newsgroups and other E-mail lists. The Internet has several tools to search in the Internet. For this study, `veronica' was used.

A search on the word `cancer' resolved 2645 hits, including gopher and mosaic sites, public E-mails from Newsnet, and telnet hosts. Of these hits, the E-mails were checked to see whether the mail was from a Usenet newsgroup or not. The gopher menus were first sorted to take out the double entries, and the rest was browsed through to determine the following characteristics:

- kind of information

- status (starting, fullgrown)

- scope

- quality of the information

- what we could learn from it

4.3 Cancer related gopher service on Internet

Site : Baylor College of Medicine, Texas Medical Center []

What kind of information does this site offer:

Tumor gene database. A searchable database of tumor genes. Future electronic databases produced by the Baylor College of Medicine.

What is its status?

Apparently this is a starting site. Altough the tumor gene database is of late 1993, all the documentation states that this is a site ``under construction'' and that more data will be added in the future.

What is its scope?

No apparent target audience. This site is globally accesible. With just this one database installed, it's scope will be mainly people in the molecular biology community. The language is english.

What is the quality (complete, well organized)

With just 1 database installed, there's not much to say about the overall quality. Further development of this site is necessary before we can say anything about it.

Site (: CancerNet, National Institute of Health []

What kind of information does this site offer:

The National Cancer Institute CancerNet service. A service containing treatment information (both for physicians and patients), NCI factsheets on Cancer (related) topics, and the Physicians Data Query (PDQ).

The PDQ has the following subjects:

PDQ Treatment Information for Patients

PDQ Cancer Screening Information

PDQ Supportive Care Information

PDQ Cancer Prevention Information

PDQ Drug Information

PDQ Database Information

Other services:

Other NCI information

Design of Clinical Trials

Fact Sheets from the NCI

Monthly Updates

Search CancerNet Database

CANCERLIT Citations and Abstracts

What is its status?

This is a very stable, well developed site. It is a prestigious project of the NCI, and, as such is very well maintained.

What is its scope?

This site is mainly targeted to an American audience of physicians and cancer patients. However, most of the data is useful to an European audience. The language is English. This site is globally accesible.

What is the quality (complete, well organized)

This is a very complete site. Everything is up to date. No stale data. All the data is updated every first of the month. It is well organized, and very well documented.

What can we learn from it?

This is a very good site, both in terms of efficiency and information quality. Something similar could be done here.

Site : Dana-Farber Cancer Institute, Harvard, Cambridge, MA.

What kind of information does this site offer:

Mainly an adressbook of local researchers and employees. All sorts of tidbits, campus information. Links to CancerNet, the National Institute of Health and other Harvard departments. Most services still under construction.

What is its status?

It is under construction. Some data is very old, but other data has been put there yesterday. It is not clear what direction this server will take, but it might be interesting to see what happens to it over the coming months.

What is its scope?

The scope (now) is fairly local. There is a lot of on-campus information. However, some menu-items suggest that a wider scope is under development. Again, it is still very experimental.

What is the quality (complete, well organized)

At the moment of writing this, the quality is poor. It is expected to improve over the coming months or so. We cannot say anything definite about it.

Site: Food and Drug Administration Bulletin Board System (FDA - BBS)

What kind of information does this site offer:

All sorts of FDA information. Up to date in formation on drug approval, news, archives on drug approval (or disapprovals). A list of topics:


NEWS News releases

ENFORCE Enforcement Report

APPROVALS Drug and Device Product Approvals list

CDRH Center for Devices and Radiological Health

BULLETIN Text from Drug Bulletin

AIDS Current Information on AIDS

CONSUMER FDA Consumer magazine index and selected articles

SUBJ-REG FDA Federal Register Summaries by Subject

ANSWERS Summaries of FDA information

INDEX Index of News Releases and Answers

DATE-REG FDA Federal Register Summaries by Publication Date

CONGRESS Text of Testimony at FDA Congressional Hearings

SPEECH Speeches Given by FDA Commissioner and Deputy

VETNEWS Veterinary Medicine News

MEETINGS Upcoming FDA Meetings

IMPORT Import Alerts

MANUAL On-Line User's Manual

What is its status?

This is a well developed site. It is not a gopher site, but a Telnet site. This means you have to log in to it, and obtain a user name. There is no fee, and anyone can obtain a user name. As soon as a user logs into the system it asks him for his name. If the user has never logged in before, it asks some questions, and it grants him an account on the system.

What is its scope?

This site is targeted to anyone who has an interest in the decisions of the FDA. It caters both to a general audience, and to professionals in the medical community.

What is the quality (complete, well organized)

FDABBS is very well organized, but a bit clumsy to use. It is not menu driven, but command-line driven. That is, a user has to type in certain keywords to get the information he wants. Altough those keywords are well documented, and on-line help is available, it is a long way away from the ease of use of a gopher system, and could be confusing to some users.

What can we learn form it?

It is indeed a lot better to use gopher for information delivery then telnet. This site could be a lot more useful if it used gopher.

Site: Fred Hutchinson Cancer Research Center (Seattle, WA.)

What kind of information does this site offer:

This information server provides Center employees with access to local and remote Internet information sources. It also provides the Internet community with a variety of information relating the the Center as well as access to a variety of local scientific services. This Center is a Non-profit organization.

The Mission of the FHCRC is to:

- Maintain facilities for cancer research.

- Conduct investigations into the nature and causes of cancer and related medical and public health problems.

- Investigate the methods of cancer prevention and treatment.

- Conduct education in all phases of cancer research.

- Perform research in all aspects of biomedical science that have a relationship to cancer.

- Disseminate all knowledge acquired.

The exact organization of the menus and the services provided are subject to change.

What is its status?

This is a probationary site. It is still developing.

What is its scope?

See above.

What is the quality (complete, well organized)

This is a good site, but with mainly local information. If they incorporate more real cancer information, this might become an interesting site.

Site: IST National Institute for Cancer Research, Genova, Italy

What kind of information does this site offer:

This Gopher is primarily intended for making IST data services available to the widest possible body of users and for facilitating access to various Internet services for IST researchers.

It includes data from the Interlab Project Databases (biological materials availability in European laboratories) and the Bio-Media Bulletin Board System (biotechnology researchers, projects, fundings and products). It enables IST researchers to access Gopher servers and WAIS sources, with particular reference to biology and biotechnology.

What is its status?

It is a fairly well developed site, and contains mostly biology, and microbiology databases. There is not much information on cancer, but a lot on molecular biology.

What is its scope?

It is located in Europe (Italy), but tries to make its information available to anyone interested. It offers various gateways to other medical information services.

What is the quality (complete, well organized)

The real information this site contains consists of a number of databases. These databases are created at the IST and are put on the server as soon as they are available.

Site: National Cancer Center, Tokyo, Japan

What kind of information does this site offer:

All sorts of cancer related information, databases, news and a gateway to the NCI's CancerNet. Also some general information about Japan.

What is its status?

It is a very well developed site, with consistent menus. The information is very good, altough the National Cancer Center does not assume any responsibility for the data on the server.

What is its scope?

Part of the server's data is in Japanese, and can only be viewed with specific (Japanese) hardware. Most of the information however is in English, and is targeted on an international audience, both within the oncological community as well as cancer patients.

What is the quality (complete, well organized)

It is rather well organized. It might be a bit difficult to find information on this server on the first visit, but after two or three times, a clear organization becomes visible.

Site: OncoLink, A Multimedia Oncology Information Resource on the Internet

What kind of information does this site offer:

This is a gopher gateway to OncoLink. OncoLink is actually a World Wide Web server, and is discussed in the section on World Wide Web servers.

Site: Springer Journals Preview Service

What kind of information does this site offer:

For selected journals in the life sciences and radiology, tables of contents and BiblioAbstracts (= bibliographic information plus the author's abstract) are distributed via Internet approx. 10 days before shipping of a new issue. The files are kept in strict standard ASCII code and the information is tagged according to BRS Medline conventions to enable both browsing and importation into a personal bibliographic database.

Users can obtain files either by subscribing to the appropriate e-mail list or by requesting individual items from our mailserver. New files will be sent to the subscribers automatically as they become available.

What is its status?

It is maintained and founded by Axel Springer Verlag GmbH, one of the worlds leading academic publishers. It is a very prestigious service, and, as such, is both very well maintained and very well organized.

What is its scope?

All individuals and institutes with access to Internet can subscribe. Tables of contents are supplied free of charge; BiblioAbstracts for ALL journals (currently about 90) cost US $20,- per calendar year.

What is the quality (complete, well organized)

It is fully operational, and very easy to use.

Site: Texas Cancer Data Center

What kind of information does this site offer:

This is mainly a telnet site, which means a user needs to log in and register as a new user. (The procedure is explained when a user selects this service.)

The Texas Cancer Data Center is funded by the Texas Cancer Council as a component of the Texas Cancer Plan to provide computerized information on cancer demographics, resources, services and programs to all who plan, develop, fund, provide, need, and/or use cancer resources in Texas.

The information provided consists of information for or about:

Health Professionals

Health Facilities and Services

Continuing Care Services (Discharge Planning)

Population and Cancer Deaths

Bulletin Board (Read Only)

Texas Cancer Council Projects

Statistical Topics

Community Resources

TCDC Assistance *** New Fact Sheet Information ***

What is its status?

It is a reasonably well developed site. They are in the proces of setting up a gopher server, but since most of their userbase still logs in through a modem this may take a while. This means that the userinterface they are using is not consistent with that of other gopher servers. This makes it somewhat harder to use than a normal gopher server.

What is its scope?

The scope of this server is mainly the state of Texas, USA. We included this server in our review as an example of how a small regional server can be built up.

What is the quality (complete, well organized)

It is a reasonably well organized server. It has a few rather impressive databases.

Site: The Jackson Cancer Laboratory (University Of Maine)

What kind of information does this site offer:

Mostly links to databases. This is because the server is still under construction. It now links to Johns Hopkins University's GDB database and other genome databases. It offers very little (real) information of their own.

What is its status?

It is starting. It is still very much under construction.

What is its scope?

According to documentation on this site, they are aiming to create a server for both the molecular biology and cancer research community.

What is the quality (complete, well organized)

We will have to wait and see. Right now there is not much to say about the quality of this server.

Site: University of Texas M. D. Anderson Cancer Center

What kind of information does this site offer:

A lot of local information, and links to other cancer related information services.

What is its status?

It is a starting site. It has been running for a couple of months, but apparently not a lot of development effort has geon in.

What is its scope?

As of the writing of this article, its scope was limited to local users. There is a lot of specific local information. We got the impression, however, that they plan to include more data for wider use on the system.

What is the quality (complete, well organized)

It is still rather experimental. As there is not much real information on this site, not much can be said about quality.

4.4 Cancer related World Wide Web services on Internet

Site: National Cancer Center, Tokyo, Japan

What kind of information does this site offer:

It offers the same information as the National Cancer Center's gopher server. In fact it is merely a gateway to the gopher server. This server has not been up and running for a long time. Apparently this is still a local experiment so far.

What is its status?

It is starting. Apart from the aforementioned gateway to gopher, there is nothing on it but links to other servers. There is no multimedia use whatsoever.

What is its scope?

The scope of the server is mainly Japan, (there are a lot of documents in Japanese) but part of the server is global, hence the gateway to the NCI's CancerNet.

What is the quality (complete, well organized)

Not organized at all. This is all very new, and should not have been operational yet, we think, other then as an experiment in how to set up a WWW server.

Site: OncoLink, The university of Pennsylvania Multimedia Oncology Resource

What kind of information does this site offer:

[extracted from: "A Short Paper on OncoLink", available on the site]

OncoLink's purposes are:

1. the rapid dissemination of information relevant to treatment of cancer and concomitant problems;

2. education of health care personnel (at all levels) in the field;

3. education of patients and families of patients who have cancer;

4. posting of clinical trials and eligibility criteria;

5. the rapid collection and dissemination of quality, peer-reviewed information pertinent to oncology in general and specific subspecialities;

6. provide a well-organized, frequently updated hypertext system to access other quality cancer information resources on the Internet.

OncoLink attempts to provide one-stop shopping for the patient, healthcare provider, researcher or Internet browser searching for cancer-related information.

Since its inception on March 7, 1994, OncoLink has averaged more than 36,000 accesses per month from around the world. While also accessible by text-based gopher servers, preliminary observations infer increased use of multimedia and hypertext documents over traditional text-only resources. From the large following of users, it is clear that electronic dissemination of high quality, peer-reviewed cancer information is very popular.

We conclude OncoLink is both useful and has wide interest in the international community. We believe, in the future, such systems will become common media for the international dissemination of cancer and other medical information.


OncoLink was released on the Internet as a WWW and gopher server in the evening of March 7, 1994. Between March and July 31, 1994, a total of 233,784 accesses were recorded from more than 62 identifiable countries around the world. Approximately 80% of these access were from WWW clients. On a month by month analysis, March yielded 34,972 accesses, increasing steadily to 63,956 access in July of 1994. The average weekday access rate is generally three times heavier than weekend use. Within a given day, OncoLink is accessed heaviest between 8am to 11pm EST, though use between 1-6am EST typically is typically heaviest from Australia, New Zealand, and Japan, reflecting their working hours.

From the Internet names, we have determined the scope of OncoLink users to be international, with accesses from countries originate from essentially all settled continents., even if English is not the primary language of the country (e.g. Japan, Germany, Poland, Israel, Brazil, etc.). OncoLink has only been advertised in the electronic environment, with minimal conventional print-media announcements.

4.5 Relevant Usenet newsgroups

Despite all the gopher and world wide web sites found on the Internet, no newsgroup specifically for cancer exists within Usenet. It would probably be a good moment to propose a number of cancer related newsgroups, as to allow incorporation of such newsgroups into TeleSCAN. However, it should be investigated if there is a sufficiently large number of potential users for such newsgroups.

The only related resolve was a large Frequently Asked Questions (FAQ) file of a social newgroup discussion on Powerlines and Cancer.


This study has been very useful for TeleSCAN in 2 respects. First, a large number of services already exist, including real multimedia ones, and will be accesible through TeleSCAN. Second, many lessons can be learnt from the gopher and world wide web sites currently publically available on the Internet.

Many gopher sites were found which did not include any structure at all, or a completely irrelevant menu. Often this is due to, probably innocent, system managers who register their site as a server, while it is only supposed to be accessed by local users. Setting up a gopher or world wide web server requires much organisational work, that should be caried out carefully, and not too hastily.

Thus, the TeleSCAN nodes and EORTC have to plan carefully the information to be put online.

Further, it should be investigated if there is a sufficiently large number of potential users for cancer related newsgroups.

These topics will be worked out in detail in the `setup of EuroCODE to function as a platform for the demonstrator workpackage.


Copyright 1994-1996. Netherlands Cancer Institute.
Date last modified: