Common databases and the cooperation between institutes

Gerd Hagmeyer-Gaverus

Why should we hold such a workshop?

To make my remarks brief, I will give you only two of the many reasons:

The first reason - from the 'information-producer perspective' - is that cooperation in information production, collection and dissemination saves money and other resources. In times of general budget cuts for institutes, this is one way to cope to some extent with that problem. Resource cuts are being made in many areas.

Where can we make savings?

a.) in information production We need to avoid duplication of work; we should not be producing and publishing the same information. We need to organize work-sharing in, for example, our indexing efforts in the libraries. As I mentioned at the conference in Geneva, an institute like SIPRI can save at least $80 000 a year by sharing its indexing task with another institute.

b.) in information collection Institutes do not need to collect the same information; rather, they should each specialize and thereby increase the total spectrum of available information.

c.) in information dissemination Not every institute needs a powerful Internet server system. Server systems could be shared by offering storage place and access to other institutes. On-line feeding of information can be made possible through simple browser interfaces. We should also start to mirror our information on other servers in order to increase accessibility. As we all know, intercontinental Internet connection lines are the bottlenecks of global data traffic. In Europe, for example, it is often very difficult to access any US server after 12.00 am, when America wakes up and starts working.

d.) in information retrieval I wonder how many hours in total our research community spends in searching for information on the Internet and in databases and how much of the same information is looked for. Assigning this search time a value in terms of working-hour costs, we can see the enormous savings we should be able to make. Letís take a fictitious example: 100 researchers, who each cost institutes $25 an hour, spend 1 hour a week searching for information on the Internet, which with better organization they could find in 5 minutes. Those 100 researchers thus waste a minum of $2500 a week or $10 000 ($9667 to be exact, because they would use the 5 minutes) a month in resources which otherwise could be used for better and faster output, for example.

The second reason - seen from the 'information-consumer perspective' - is that cooperation and specialization in the collection of information broaden the access to information and therefore help researchers to achieve better results, help politicians to make better decisions, and give the media and the public a better understanding. Today's information technology eradicates the concept of distant locations. Nowadays a location is not distant as long as there is an Internet connection. For the first time in history we have the chance to access information vertically and horizontally, at (in the best of all circumstances) almost the speed of light. Research on the security problems of Malaysia, for example, can be conducted from an office on Greenland if the information is available on-line. With all the other tools - e-mail, video conferencing etc. - some of the important elements constituting the "personal contacts business" can also be covered, although personal meetings can never be entirely replaced. (You still cannot drink a beer with your colleague on-line.) However, if institutes in Malaysia make the relevant local information available on-line - journal articles, newspaper clippings, documents etc. - the researcher in Greenland has all she or he needs. Regional networks of cooperation are therefore an essential part of our daily work. Regional networks, in their cooperative approach, do away with work duplication and coordinate resources in the most efficient way. Regional networks are a major element of transparency in the study of regional problems.

As I mentioned there are many more reasons but I donít need to convince you, as you wouldn't be here otherwise.

But why should we start now?

We need a concept and a policy to establish our own structure before economic or other structures are taking our place. Many institutes are only now starting with their electronic publishing business. If we don't organize ourselves and our information, commercial enterprises will take up this task and offer - for money, of course - information packages which they collect from our servers, and we will have no control over the standards of such information packages. Commercialization of our information at this point, by the way, is most questionable because users from countries in the developing world or the new evolving democracies in Central and Eastern Europe, for example, who are in the greatest need of this information, would be excluded sisnce they cannot afford it.

The rapidly growing Internet hosts more and more servers everyday and is becoming an impenetrable jungle. If we don't succeed in sticking out as the highest trees in this jungle we are going to be lost and invisible. To be invisible is the last thing we want because we want our information and messages to be widely spread. Only a strong cooperative effort makes us information providers which cannot be overlooked.

What is the future vision for such a cooperation?

The vision is a comprehensive information system in the field of International Relations and Security that is user-friendly, organized, well-structured and global - a system not only for specialists but for the general public as well, a system with the relevant information stored electronically and made accessible, a system with a comprehensive search engine allowing the user to access information on the subject in all the forms of our multi-media world, including video and sound.

However, we cannot talk about a final design unless we have identified the elements and functions of such a system, a task we will achieve by having this kind of workshop.

This is why we have to look into the questions of distributed database systems, intelligent information retrieval systems, user interfaces and so forth. But the greatest task is to get our information published in an efficient - meaning resource-saving - and rational, logical way. However, we need to bring technical experts together with subject experts in order to make this happen.

What should be our goal for the workshop?

According to the mandate from the conference "The Institutes and the Security Dialogue"in Geneva last fall, at least one workshop should be organized, but possibly more. This first workshop should set roadmarks, should point out a direction. We will not be able to cover all the problems of database cooperation because it is a complex area, but we should be able to identify the first main elements.

From the huge area of database cooperation I hope we will be able to pick a few which we can immediately transfer into concrete projects. I have one example in mind - a common database on hypertext links. Two institutes represented here have already done major work in this area: the International Relations and Security Network (ISN), an effort which evolved from the 1994 conference "The Institutes and the Security Dialogue" in Zürich; and IanWeb of the University of Pittsburgh. Both projects have been in place for some years now and connecting them together would create a powerful search engine in our field. The questions to be solved are: What work-sharing scheme must be applied, and what standards and procedures should be adopted?

Defining standards is crucial for every sector of our efforts. We need to define information quality standards, information reliability standards, and information accuracy standards. We also need standards for labelling and processing information such as official information, reports, analysis, raw data, compiled data. The list is long.

To make this workshop a success we should be result-oriented and aim to launch some pilot-projects. The working groups session should discuss such projects and present them to the plenary.

Let me give you some ideas of projects which are already under discussion.

SIPRI will for UNIDIR publish their DATARIS database system as an on-line interactive system. DATARIS is a collection of databases on institutes, research, conferences, experts and so on. The database system, which remains the full responsibility of UNIDIR and for which SIPRI serves as a host, allows for on-line changes from the record owners, the individual institutes and organizations. This set-up will allow for instant up-dates of information whereby UNIDIR keeps the contacts and reminds the institute to make an on-line update or UNIDIR itself makes the on-line updates if the institutes do not have the possibility to do so.

Another idea is to develop an information system - (if financing can be found) a SIPRI and University of Brussels cooperation - on conventional arms, chemical and dual-use technology export control issues. This system will be designed for both the expert and the lay-person user. The expert user enters a search engine for explicit expert search concepts. The layperson will with the help of educational modules be guided through the subject to finally end up in the expert system for his specific detailed questions.

Additional ideas are to further develop a comprehensive and cooperative literature database and databases on full text documents. What comes to mind is documents of international organizations, especially the OSCE, but also NATO, the WEU, the Council of Europe and others. If locally produced and hosted, these databases need an interface which allows users to search in all of them. Searching among those different sites is still the most time-consuming task of researchers and other users.

The list of ideas can be developed in the working groups.

What is the impact of our efforts?

What is it good for? Accurate, high-quality information over a broad spectrum helps us to make better assessments. The demand is there. Looking at our server accesses, say, those of SIPRI, NATO or any other organization, shows that we are getting an exponential increase in the number of accesses. What is not surprising is that it is the professionals who access the information most often, not the curious one-time Internet surfers. For example, when I looked more closely at one of the recent month's server statistics I found that SIPRI had more accesses from US government agencies than from any other country but the US in the world.

The change in working procedures is certainly one of the reasons. Working from the desktop computer and collecting information directly from the Net makes the working process faster and more effective. Visits to the library or archives are made much less often. Information at our fingertips is becoming a reality faster than many expected it would. (We are becoming truly 'Internetional'!)

Our task is to provide qualitative information. In return this will help us and others to prepare better decisions. If our information penetrates the administrations of governments, there is a good chance that our information will influence the decision-making processes.

I believe there are a number of studies that will look into the question of how the Internet influences the political decision processes, a subject we should be interested in. A conference in Washington held by the United States Institute of Peace will deal with the question of whether the Internet has an impact on peace processes. From the knowledge we have about the many accesses to our information servers, I have no doubt that we already make some difference in these processes. We make a difference when it comes to transparency, thereby contributing to confidence among states. We, the institutes and organizations, are as a total entity in fact a confidence-building measure.