The evolution of modern medicine towards its present status as a discipline with a well defined scientific base followed a parallel way to the recognition of the importance of docu-mental and bibliographic sources, their regular consultation, and the feedback to those same sources thus a knowledge socializing process of the new data obtained through research. In this sense, public health as a multi/transdisciplinary field has not yet reached the same matureness, remaining within health care administrators and health authorities suboptimal routines in infor-mation managing and decision making (Weis 1986). In spite of this adequate information is a must, from public health research (both as starting and as reaching point), to decision making at all levels: political authorities, planners, and managers.

2. INFORMATION: REQUISITES, DIFFICULTIES, CONSTRAINT

In the era of information explosion, every effort undertaken towards its contention and control appears to be self justified although they may be of limited success. An adequate iden-tification of the different information typologies and sources is very important. From appro-priate information we mean the one that is pertinent, complete, correct, self-contained, up-dated, verifiable, accessible in time, place and cost. It's main varieties are textual, factual, and graphic information. The universe of pertinent information is broad and complex, encompassing demo-graphic, socio-economic, political, cultural, educational, environmental, and legal aspects, as well as those related to the health services system and the medical sciences. This causes the information necessary for public health investigators and managers will be found disperse and under heterogeneous gathering, indexing, editing, distributing and consulting systems, making its obtaintion very difficult. Necessary information will be contained in books and mono-graphs, in journals, newsletters and serials arising from scientific societies, official or technical agencies, and private companies; in publications of almost all government departments; in annual reports, directories, and legal norms compilations; in documental sources of municipal authorities, autonomic institutions, teaching and research centers of different classes, etcetera. The kind of documents may vary from normal editions accessible through regular channels, to the so called gray literature, represented by documents of limited circulation but of perhaps great interest. The dominant physical media may be paper, but a certain amount of relevant data may be in magnetic tapes, magnetic or optic discs, microfilms collections, films and slides, and others.

3. DATA BASES: THE OLD AND THE NEW

Traditional data bases have contribute largely to information and documentation diffusion, but their disadvantages are important too: hierarchic structure and rigid organization make un-avoidable a linear-lecture, and require lots of technical skills for being able to perform produc-tive searches. From a general point of view we could talk of two different types of databases, the textual-bibliographic ones on one side, and the numerical-factual ones on the other.

In the international field the most important bibliographic databases in biomedicine are produced by the National Library of Medicine in the USA, Excerpta Medica in the Netherlands, BIREME in Sao Paulo, Brazil, and in the Spanish field by the Institute of Medical Documen-tation in Valencia, and by the Institute of Scientific and Technical Information in Madrid. The printed or electronic products of outstanding importance are: Index Medicus, Index to Dental Literature, International Nursing Index (contained in MEDLINE), Health Planning and Administration, International Pharmaceuticals, Current Contents, EMBASE, Index Medicus Latinoamericano-LILACS; and in Spain the Indice Médico Español (Spanish Medical Index), now in CD-ROM altogether with the databases from CSIC (Superior Council of Scientific Research).

Factual or factual and textual information is processed by the National Statistics Institute (INE), the Consumers Affairs Institute (INC), the General Pharmaceutical Directorate, the Health Ministry Data Processing Center, the centers committed to epidemiologic surveillance, and in the international field for a large number of governmental agencies, international organi-zations, and private institutions. Undoubtedly a wide sea of information where it may be easy to get lost or drown oneself rather than find what one is needing. The new database systems under hypertext and multimedia concepts, organized in a similar way as the human mind functioning by association, trend to break down barriers between man, machine and informa-tion. Under this light a public health hypertext would appear of great interest for three reasons: its great storage capacity; the possibility of free navigation, and free data and ideas linkage; and the multimedia support, mainly of text, numeric data, graphics and images. Hypertexts require an appropriate informatic system, well integrated by a physical ware (computer) and a logical ware (set of programs). We found that the HyperCard™ program for the Apple Macintosh was a quite interesting tool, worth being explored for this objective. This was suggested to us for the large number of works done around the world within the past 2-3 years (Hardman 1989, Landow 1989, Nunn 1989), and by some initial experiences done in Spain in the fields of history, sports, and arts (Rodriguez de las Heras 1989, Canals 1990).

4. THE EXPECTED PROMISE OF HYPERTEXT AND HYPERMEDIA

It seems like if hypertext would have been a brilliant idea waiting for the external condi-tions to make it mature. When in 1945 an American government scientific adviser imagined a virtual machine called "Memex" (Bush 1945) that could prolong the capacity of human brain acting by automatic association, and finding instantly any required data, the information technologies where still far from being able to produce such sort of apparatus.

Twenty years later Ted Nelson achieved the term "hypertext" (Nelson 1965), and although his long standing "Project Xanadu" a sort of "whole world knowledge distributed database" remains utopian, he surely inspired the people that another twenty years later where going to develop real hypertext and hypermedia systems (Goodman 1987, Conklin 1987, Ambron 1988, Barret 1988, Cambell 1988, Chen 1985-91, Baird 1989). At the present we are seeing in the world of information technologies a real explosion and glamour of multimedia, making com-panies and developers move in this direction, aiming to offer new products, impose fact stan-dards, etc... For the first time in nearly thirty years of technological independent evolution, the separated world of information-data processing and the video-images industry are beginning to converge.

If we add to this picture the continuing price dropping of information equipment, and the increasingly availability of programs for end users ("make it yourself" is more than ever the era motto) we can expect a wider use of this technologies and resources in all the fields of society, like education, libraries and health, in none of which this tools and possibilities have been previously abundant.

5. OBJECTIVES AND PHASES

Defining our project as a R&D one, it should be noted what we mean for each of these components. Research does not refer here to the acquisition of any new or basic knowledge, it rather focuses on the investigation of how appropriate may result the hypertext-hypermedia concepts and models for the organization and handling of the information pertinent to Spanish public health, and how well do the hardware and software systems selected perform to accom-plish this task. The development component imply that the expected result is to obtain a product, a mass of useful, organized information, publicly available to the scope of potential users.

The users of our system should be a variety of health students and professionals, from the undergraduate and specialist in training field, up to officials working in health planning, hospital care, primary care services managing, teachers and researchers in the schools of medicine, public health, nursing and allied disciplines, and even professionals from outside the health field (journalists, sociologists).

The functioning of the system should be friendly enough to permit people with no pre-vious training in computer use to deal with it productively. The system should be able to give answers to specific, occasional questions, as well as in depth studies of complete parts, or wide thematic research. The user should be able to obtain products to take away "especially printed reports" from a working session. Finally, and although a hypertext system may still be a corps of data where the possible connections between its components is predefined by their authors, users should have a lot of freedom to navigate it, resulting in each session from each user being different from others.

The practical course of the project encompasses two phases. The first one has as the objective the development of a prototype. This should have small amounts of information of as many as possible types of data. In the same way the links and connections should be present in all its possible complexity. The prototype is then available to be tested with different types of real users, and to gather their opinions and suggestions.

In the second phase, and after all the necessary corrections have been done, the prototype is able to be transformed in a definite system, through the inclusion of large amounts of data in each of the designed categories. When this process ends the product is ready to pass the stan-dard industrial procedures (premastering, mastering, duplication) and enter predefined ways of distribution and circulation.

6. CURRENT DEVELOP

In the layout of our project we have tried to simplify its components in three areas:

• The prime matter: information;

• The tool: an integrated informatic platform; and

• The final product: an information public resource (prototype in hard disk first, definite system in optical disc finally).

In the following lines we will comment regarding some specific considerations about the tools and the prime matter. We are currently working in the develop of the prototype, for which we have organized a small informatic laboratory, integrated by a Macintosh with processor 68030, working in a sort of network with two compatible PCs, with processors 80386 for one of them, and 80286 for the other. Altogether these devices have 16 MB of RAM, and approxi-mately 400 MB in file systems. They are networking through Apple resources, especially Flash cards in the PCs and the Tops software, and utilities of the MacLink+PC program. Attached to this equipment there are two CD-ROM players (Mac and PC), a color scanner, and two laser printers. It was planned, but it has not yet be implemented, that this equipment would connect to a departmental network served by a minicomputer running Unix operating system, which would also be integrated in a host consortium supported by the Ministry of Health Data Pro-cessing Center, and other official institutions of this area, offering an important catalog of on-line databases and other resources.

HyperCard™ was the program selected for our system; we knew of other options, in different computing environments (Guide, Intermedia, NoteCards, SuperCard, and others) but for an institution like ours, with limited resources in equipment, money, knowledge and com-puting staff, HyperCard™ seemed to be the easiest and more extended tool used in similar projects. We planned to "put around" this core engine other pieces of software, necessary for information loading, converting, treating, and preparing before putting it into the hypertext. There where all standard software for word processing (WordPerfect Mac 2.01, and PC 5.1), utilities for format conversions between PC and Mac files (MacLink+PC), a image processing software (PhotoShop), an optical character recognition software (OmniPage), a spreadsheet (Wingz), and other miscellaneous programs for PCs.

We are especially concerned with the issue of database management systems (DBMS) as complementary programs, or maybe subsets of the hypertext, mainly because of the limitations that HyperCard™ may present in tasks like indexing and global and efficient processing of a large number of records. An additional problem commented by some authors (Canals 1990) for hypertexts systems intended for duplication and distribution is that if a standard DBMS is linked to it, the producer should distribute as much run times as copies, with the copyright royalties obligation that this may imply. At the moment we have limited our work to the use of documen-tal DBMS as intermediate tools, for containing temporary "record tanks" that will be lately selected, treated, and loaded to the hypertext, as well as for managing the bibliography and references for the project realization it self; the programs we are currently using are PC versions of the BRS software, and CDS/ISIS from Unesco.

7. THE IDENTIFIED PRIME MATTER

The main sources and types of information with which we are currently working are bibliographic databases in electronic media, and printed matter from different kinds. Due to constraints in the telecommunications infrastructure and services available in Spain, that make difficult and quite expensive remote connections, we strongly prefer the use of databases in CD-ROM.

In respect to international products we are exploring the data contained in MEDLINE (Compact Cambridge) from 1966 to 1991, in HEALTH (Compact Cambridge) which contains public health and health administration information in one disc covering the 1980-91 period, and in LILACS (PAHO/WHO) a disc containing Latin-American biomedical information in a set similar to Index Medicus. From all of this information (in MEDLINE more than 5,000,000 records) we have to sort by terms that will relate to Spain, and refine some search profiles in order to exclude information strictly referred to clinical or laboratory data, or any other basic information, which would not be pertinent for us.

In respect to Spanish products we are working with the information contained in the CSIC CD-ROM, containing in one disc several databases produced by the Superior Council of Scientific Research institutes, one of which is the IME (Spanish Medical Index), with references published in about 300 medical journals from 1980, and with de CompuLey disc on legislative information (General Laws, Royal Decrees, etcetera), from 1981 to the present. From this products data is sorted in order to its pertinence to public health.

In general the exploration of CD-ROM products is followed by outputs of groups of records in their native formats, to text files. This files are then processed and converted to a standard record format database, acting as an intermediate source (the records tanks mentioned earlier), and from here, through different criteria, to the corresponding stacks of HyperCard™.

Referring now to printed matter we will only mention some of their types and processing. References from medical periodicals published usually weekly, where information and news about public health are obtained, and catalogued manually in the standard record format data-base. This also applies to some newspapers articles and news related to health care issues. With both of this references, a copy of the original article is stored, and may be loaded if necessary as a full text document through the scanner and OCR software. Documents of different length, produced by our own institution of others, or by individuals may also be stored as full-text through this same procedure. Other relevant information, for example the complete text of the General Public Health Law, will be incorporated. When scanning and using character recog-nition, technical problems may occur due to the quality of original media, typing the data manually is the alternative.

Finally other group of printed matters of interest are the ones integrated by graphics, photo-graphs, maps, etc... Very typical are the photographs of hospitals and primary health care centers (that may constitute by itself a visual catalog); the maps of the Health Areas and Sectors with the allocation of health care facilities; and then photographs of relevant persons, authorities, and scientists in the health field with attached information about their background, position, works (integrating a potential directory by itself).

8. PERSPECTIVES AND CONCLUSIONS

Working with this enormous and heterogeneous mass of information, analyzing it, classi-fying it, cutting it and pasting it, on behalf of constructing some form of intelligible superior order renders results many times astonishing. We still have a lot of problems to solve, in the different stages of the information acquisition, processing, handling and loading; one of the more critical issues being the possible limitations that may arise in the functioning of Hyper-Card™ with a critical mass of data, and the ones derived from its scarce possibilities to openly connect with other applications. There is a main technology problem to solve when a PC version of the product may be intended. In any case, we believe that the "HyperCard like interface" is a rather optimal one for an information platform including its navigational resources based on mouse pointing and clicking principles (may be complemented or substituted by sensitive screen touching), even if other application software may be supporting the whole system in a lower layer.

Nevertheless, while all the technologies we are working with are charming and glamor-ous, we try to keep in mind that they are only instruments in order to accomplish the final objectives, which are, offering health professionals a powerful tool to improve their knowledge and work, and, specifically in the public health field, to better serve the people we are committed to.

REFERENCES

Ambron S & Hooper, K. Interactive multimedia.. 1988.

Baird P., "Hypermedia," Hypermedia, 1 (1) Spring 1989.

Barret E., ed. Text, context, and Hypertext: Writing with and for the computer. Cambridge, MA: MIT Press, 1988.

Bush V., "As we may think," Atlantic Monthly, July 1945.

Campell B. & Goodman J. M. "HAM - A general purpose hypertext abstract machine," Communications of the Association of Computing Machinery, 31 (7): 856-861 (1988). (Also Appeared in Hypertext 87).

Canals I. & Llacer, E. (Institut de Estad stica de Catalunya): Por una metodolog í1 a de dise 9 o de interfaces "navegacionales" para los sistemas hipertexto basada en el an á lisis por facetas. INSYS 90. Jornades de Tecnologies Interactives. Barcelona, 29-30 Mayo 1990.

Chen, C. PROJECT EMPEROR-I: China Treasure Revealed via Videodisc Technology. Boston, MA: Simmons College, 1985-present. Videodisc and multimedia software published by both Simmons College and The Voyager Company (1991).

Conklin J., "Hypertext: an introduction and survey," Computer, 20 (9): 1741 (September 1987).

Dumais S.T., Furnas G.W., Landauá T.K., Deerwster S., & Harshman R., "Using latent semantic indexing to improve access to textual information," In Proceedings of the ACM (Association of Computing Machinery) CHI'88 (Washington DC, 15-19 May 1988). pp. 281-285.

European Conference on Hypertext ECHT'90. Paris, November 27-30, 1990.

Frisse M.E., "Searching for information in a hypertext medical handbook," Communications of the ACM (Association of Computing Machinery), 31: 880-886 (1988).

Frisse M.E., "From text to hypertext," Byte, October 1988.

Glushko R.B., "Transforming text into hypertext for a compact disc encyclopedia," In CHI'89 Proceedings, New York, May 1989. New York: ACM (Association of Computing Machinery), 293-298.

Goodman Daniel, "Coping with the information revolution," In The complete HyperCard handbook. New York: Bantam Books, 1987.

Goodman Daniel. HyperCard Developers Guide. New York: Bantam Books, 1988.

Haedman L., "Evaluating the usability of the Glasgow online Hypertext," Hypermedia, 1(1), 1989.

Hypertext' 89. Proceedings of the Second ACM (Association of Computing Machinery) Hypertext Conference. New York: ACM (Association of Computing Machinery), 1989.

McAleese R. ed. Hypertext: Theory into practice. Oxford: Blackwell Scientific Publications, 1989. 175 pp.

McKnight, C., Dillon, A., & Richardson J., "Problems in Hyperland? A human factors perspective," Hypermedia, 1(2): 167-178 (1989).

McKnight, C., Richardson, J., & Dillon, A., "A comparison of linear and hypertext formats in information retrieval," In Proceedings of Hypertext 2, University of York, UK, 1989.

Nelson T. H., "The Hypertext," Proceedings of the International Documentation Annual Conference, 1965.

Nunn, D., Leggett, J., Boyle, C., & HickS. D., "The REXX Project: A case study of automatic hypertext construction," Technical Report, Department of Computer Science, Texas A&M University, April,1989.

Oren, Tim, "The Apple Learning Disc: A HyperCard collection for education, project design and management, including selections from The Whole Earth Learning Disc, The Grolier US History Demonstration, Project Jefferson (USC), Perseus Project (Harvard and Bowdoin), and The Electric Cadaver (Stanford)".

Rodriguez de las Heras, "Los Hipertextos: el moderno Aleph. Iris," Boletín del Programa de Interconexión de Recursos Informáticos (Fundesco), N0, Madrid, Abril 1989. p. 20.

Shneiderman, B., "Reflections on authoring, editing and managing hypertext," In E. Barrett, The Society of Text. ed. by E. Barrett. Cambridge, MA: MIT Press, 1989.

Vaccaro, William (Conrad Sulzer Regional Library, Chicago), "Culture 1.0: Creating a Cultural Workstation. HyperCard Applications," Computers in Libraries pp. 42-5 (September 1989).

Weis, P., "Health and biomedical information in Europe," In Public Health in Europe. Geneva: World Health Organization, 1986. p. 27.