[VIROR][ULI] [Institut für Informatik] [Fakultät für Angewandte Wissenschaften] [Universität Freiburg]

[Home]  [Ebene höher]  [Seitenende]  [Suchen]  [Kontakt]

Bruno Mannoni: Bringing Museums Online


Bringing Museums On Line

Bruno Mannoni (mannoni@culture.fr)


Copyright ACM 0002-0782/96/0600 Permission to make digital/hard copy of part or all of this work for personnal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage, the copyright notice, the title of the publication (Communications of the ACM) and its date appear (June 1996), and notice is given that copying is by permission of ACM, Inc. To copy otherwise, to republish, to post on servers, or to redistribute to lists requires prior specific permission and/or a fee.

The author provides an overview of the organizational aspects of publishing and distributing large collections of materials on the Internet.


In most European countries an important effort has been done, often during centuries, to preserve, describe and index the cultural heritage. More recently, many public authorities or private entities have undertaken to digitalise a more or less important part of this heritage. Digitalisation means creating databases containing images, reference documents, factual data, sounds, describing as precisely as possible the items constituing the cultural heritage. Paintings, sculptures, monuments, ancient manuscripts, music instruments, historical furniture, and photographs can be described by digital information, thus offering a large population of users easy acces to our cultural heritage. The rôle of cultural institutions is not only to collect, preserve and display the collections but also to organize the information thus collected. This was clearly stated at the february 1995 G7 meeting in Brussels, where it was specified that "culture is a key dimension of the Information Society. The exchange of information on the world's cultural heritage will help people from different cultures around the world to understand each other better". Among the eleven pilot projects launched by the G7 initiative, one is dedicated to "Multimedia access to World Cultural Heritage ". The preparation of this project is entrusted by the Italian and French governments and its purpose is "to accelerate the multimedia digitalisation of collections, and to ensure their accessibility to the public and as a learning resource for schools and universities".

At that point, it is necessary to stress that there is a major risk of impoverishment in doing so. If we deal with cultural heritage as a market driven product, what will be the interest of a private entitie in digitalizing little-known, unpopular or difficult to comprehend works of art ? I am personally convinced that the risk of " pauperization " exists, and there is a need for state intervention in the cultural field. I know that many people do not agree with that non-liberal (from an economic point of view) statement, but I fear that the " global village " concept could induce a " global lowering " of the quality level of cultural information (due to economic pressure). That is one of the reason why France raised the " cultural exception " issue during General Agreement on Trade and Tariffs discussions, and why we have quota on TV tansmission and radio programs, a law to defend the French language, and a ministry of culture (whose role is to describe and make known to the world France's cultural heritage).

When we consider " information highways ", it is very important to ditinguish clearly between the pipes (the hardware) and the contents that these " highways " will carry. The word " highway " is somehow misleading: building a conventional highway rquires a major financial investment . Since the lobbying capabilities of the Telcos (the knights of the pipes) are very efficient, politicians are primarily speaking of the necessary investments to set-up the hardware, instead of considering the content that will be carried, which is the most important matter. The support at the highest political level of the "  information highways " is both an opportunity and a challenge for the European Union. In the US, some major industrial companies have understood that " cultural heritage " can be used to attract customers to commercial networks. Europe has to be proactive to ensure its fair share of this new market. Convergence of electronic, communication and computer industries has created a large alliance, which could be detrimental to Europe whose contents industry is scattered.

Nevertheless, in spite of these somewhat pessimistic considerations, this digitalization effort must be carried-on and increased. It will take many years to complete this effort (for example, the ministry of culture in France has 22 million photographic negatives waiting for digitalization and indexation).

We must bear in mind that digitalization is of no use if indexing is not done previously, or at the same time. Digitalization is an addition to indexing. Since indexation can only be done by highly educated human beings, it is much more expensive that digitalization, but also it adds a lot of value taht can be invoiced to the customer.

We are going to be faced with the creation of fairly large information bases containing millions of multi-media documents. Users will need easy access to this information wherever it is located. It is therefore important to design information systems that will make possible effective management of these huge information banks, allow easy retrieval of the relevant information through a multi-lingual interface, and adress the needs of experienced users, which will facilitate the realization of interactive presentations for the general public.

It is now clear that the World Wide Web paradigm is a good starting point concept to define such an information system. It lacks many features, however:

  • The HTML DTD is too weak to reflect complex hypermedia structures
  • The HTTP protocol establishes a connection at every HTML page exchange.
  • No ISBN like capabilities (hard coding of links necessary).

I am quite confident that these features will be available in a few years, and we will incorporate them as soon as they are released, as we have done in the past. For the last 25 years, France's cultural heritage has been indexed by professionals, and we now have important textual databases. Clearly the fact that there is a state administration dealing with cultural issues helped this process, particularly in creating rich (from a scientific point of view) and common thesauri for the different fields. The information system of the ministry of culture in France is organised around a set of major databases. Each database contains approximately 80 fields (keyword, numeric data, and free text). The size of an average record is approximately 1,200 bytes. Lexicons and thesauri contains four categories of relations (generic terms, specific terms, synonyms and see also) and ten levels of hierarchy. The three major databases are :

  • MERIMEE (130 000 documents) which describes the entire collection of monuments
  • PALISYY (170 000 documents) for the furniture heritage
  • JOCONDE (120 000 documents) for the paintings, drawings and sculptures located in French museums

We offer general public access to these databases, with the exception of the PALISSY database -there is a risk of stealing the described objects, which are often in churches or private homes. In fact France has been the first country in the world to set up a large general online information system with the minitel in the early 1980s. Nearly 7 millions minitels are installed in offices and at homes, generating a service revenue of 16 0 millions US $ per year, and 1,2 million of on-line shopping transactions in 1994 (The figure in the U.S. for 1994 was 800,000 online shopping transactions.)

Unfortunately we were not able to export the minitel technology, and it did not spread out of France as the Internet spread out of the U.S. We are seing a (small) decrease in the minitel usage in France since last year, but the minitel system is not dead yet. We do not need multimedia capability to query a phone directory or an airplane timetable. Gateways between the Internet and the minitel network are now available through Intelmatique (a subsidiary of France-Telecom, the national operator). For the past 4 years, anyone in France can use the minitel to access the databases on France's cultural heritage through the " 3614 JOCONDE " service. It cost one telephone tax every three minutes to the caller -this is roughly 20 US cents every three minutes, which the caller pays to France-Telecom (no money goes to the ministry of culture). The deregulation of telecom operators will take place in France on January 1, 1998 -we will then see if the local rates, and the leased lines tariffs will decrease. The cost of telecommunication, especially leased lines is far more expensive in France than in the U.S. For example, a T1 connection to the Internet is 30 times more expensive in France than in the U.S according to published tariffs.

The telephone infrastructure is very correct in France nowdays: digital from end to end, with ISDN capabilities everywhere in France. In 1992 we began to set up a corporate network (culture.fr) based on TCP/IP - linking together 4 000 micro-computers in 80 differents locations all over France. The initial purpose of setting up this network was to facilitate access to internal computer applications, but in the summer of 1992 we also had an Internet accees (mostly used by the computers professionals within the ministry of culture). From a technical point of view, we set up in every site a 10 base T (cable 5) ethernet network on which we connected the Pcs, a Unix machine running Lan Manager Unix for sharing ressources and we interconnected these 80 Local Area Networks using TCP/IP over X.25. Potentially, everyone in the organization could access the Internet network. Fortunately, we set-up a network compatible with the Internet suite of protocols prior to the growth of Internet in France. (In July 1991 10,000 computers were connected to the Intenet in France ; 100,000were connected in April 1995. The last DNS count (May 31, 1995) indicated 120,000 computers connected to the Internet in France, compared with more than 400,000 in the U.K. and Germany and more than 150,000 in the Netherlands).

When Jacques Toubon became the minister of culture in 1993, he made dealing with cultural issues brought on by new technologies (on-line or off-line) his number-one priority. When we requested his staff to officially connect the culture.fr network to the Internet to provide on-line information on France's cultural heritage to the Internet community, the answer was: " go on ", which was not, in the context of 1993, an obvious decision. There was at that time a powerfull lobbying against Internet, this so-called American network, not invented here, and giving no revenue to French industry. The first service that was set-up was to provide electronic mail capabilities for the minister - ministre@culture.fr - an address that received a lot of trafic when France decided to resume nuclear testing in the Pacific.

Times have changed, France-Telecom is now clearly in the market with the RAIN offer (which we use) and the RENATER offer for the universities and research institutes. Sother indicators of changes include the sixty-two Internet Service Providers now listed at http://www.urec.fr and the fact that sixteen cyber cafes opened last year in France.

Many things happened in those last three years, technically and politically. On the technical side, I would say that the Mosaic browser technology applied to the World Wide Web was the most important development The first time I saw the WWW and Mosaic was at the first Interop meeting in Paris, at a conference sponsored by CERN in December 1993. At that time I was very proud of a gopher service we had set up, but the World-Wide Web technology with the multimedia capabilities that Mosaic enabled was a real shock. Since we knew nothing anything about HTML technology at the ministry, we asked the French institute in computer science, INRIA , to help us in setting up an imaginary exhibition on the World Wide Web (as did the library of congress with the vatican exhibit).

The head of French museums (whose managing director was a specialist of the 18th century) chose the subject: the age of enlightenment in the paintings of France's national museums. We used Kodak photo CD technology for digitalization, CERN World-Wide Web technology for the HTTP daemon linked to a WAIS research engine to query the database. This service was officially launched at a meeting on the September 26, 1994, by Jacques Toubon. You can visit this exposition (in French or English) at the following URL:

http://www.culture.fr/lumiere/documents/files/imaginary_exhibition.html

A panorama of eighteenth-century French paintings is presented in this exhibition through the works of 100 selected artists. In addition to the pictorial world, the history, music, litterature and science of the period are recalled in the background to these occasionally little-known works from 18 museums all over France. It is really an imaginary exhibition, not a virtual one, because in real life we would not have been able to put all these paintings in the same place for legal reasons or for the poor state of physical conservation for some of the, paintings.

The experience was very conclusive, but questions are still pending on the legal and commercial side. Nevertheless it has been decided to setu-up all our World-Wide Web pages in French first (and eventually to translate some of them in other languages), to build more exhibitions, and to speed up the digitalization process in France.

On legal issues, Pierre Sirinelli, a well-known French jurist, has been charged by Jacques Toubon to study legal issues concerning new technologies and culture. His work has been published and is very interesting. In France, you can go into every state museum, take a photography of any painting (you are not allowed to use a flash) that is more than 50 years old, and reproduce them on postcards, tee-shirts, and multimedia products without having to pay any fee to the state. Professor Sirinelli suggested that the museums should become producers of multimedia products, because they will then have rights on them.

We made the decision to put on Internet only medium-resolution pictures, which is not enough for an art quality publication in case of " hacking " of the pictures. We are also working with INRIA on water-marking technology to protect our work.

The World-Wide Web technology in its present form is not adequate for important databases. Human creation (hard coding) of links for 100 or 1,000 documents is feasible, but is not realistic when you deal with more than 100,000 documents. So we have proposed to the Europen Union a joint research program with INRIA, Grif SA, O2 technologies, BULL SA, Greece, Italy, and the UK, called AQUARELLE, to work on additions to the existing technology for providing cultural multi-media services in the years to come, adding object oriented database technology to the WWWservice.

The French prime minister in 1994 asked for a report on the information highways and asked the minister of industry to set up a call for proposals on experimental services on the future of information highways. More than 600 projects have been proposed (in a very short period of time), and 49 have been accepted from immediate beginning and about 100 have received the label " service of public interest ". One of them is a proposal from the ministry of culture and groupe Bull, to set up an information service on the cultural heritage. A beta version (for the time being in French) of this service is located at the following URL:

http://www.culture.fr/spic.html

The idea is fairly simple : we have databases on the cultural heritage that are running with a proprietary product from Bull called " mistral ". It is a powerfull documentary information retrieval system, which our conservation professionals are used to. The user interface is not at all suited for general public use on the Internet. So we decided to incorporate some of the Internet technology, namely World-Wide Web queries capabilities and Z 39.50 compliance.

The first step has been achieved, and with a forms capable browser userscan query our databases, which now have much-improved interfaces. As we know that the general public wishes to see an iconography linked with the textual part of the database, we are in the process of digitalizing thousand of documents, which will be incorporated in our database in two formats (icon and full screen). This was not too difficult to achieve (6 months/man effort) because application programming interfaces were available with mistral, which is now capable through a cgi-bin C program of generating HTML pages on the fly, answering the query users ask.

Another project that began in October 1995, and in which we participate as a content service provider, is very promising. It has been set-up by " La Lyonnaise Communication " which utilizes the cable TV network in Paris and its suburbs. Here again the idea is simple: 45% of the cable TV subscribers in Paris have a computer. Multicable is introducing the TCP/IP technology on the cable TV network, thus offering applications that will run on the computer (through an Intel CablePort modem connected to cable TV), and Internet access is also available. Two hundred people in Paris are beta testing the service , which provides access to radio (Europe 1), travel agencies (Nouvelles Frontieres, Dégriftour), on-line market (3 Suisses, La Redoute), publishers (Larousse, Bayard), banking services (crédit Lyonnais, CCF, Banque Sofinco, Banque Populaire, Crédit Mutuel), news agencies (Agence France Presse, sygma), newspapers (l'auto-journal, pariscope, Le revenu Français, Le Monde, La Tribune) and the ministry of culture. This service is operated by VTCOM a subsidiary of France Telecom. All the products are written with the HTML technology and most of them are running on Linux boxes hosted by VTCOM. The Internet access is provided by Transpac/RAIN, a subsidiary of France-Telecom. The cost is of 15 US dollars per month for the basic services plus 30 US dollars per month to have Internet access.

Pilot Projects

Politically, at the international level, things have also progressed. The G7 meeting in February 1995 in Brussels , dedicated to the Information Society was a succes and many services were demonstrated there to high-level decisons makers.Eleven pilot projects were launched by the G7 meeting (and will probably be demonstrated at the next summit in May 1996 in South Africa), two of which involve France and French cultural heritage:

Bibliotheca Universalis

The objective of the project is to constitute from existing digitization programs a large distributed virtual collection of the humankind's knowledge, available to a large public through networking. Bibliotheca universalis should strengthen the role of libraries and improve availability at international level. The digitized resources include both the bibliographic records and the information content itself (text, graphics, still image, sound and video information). It will promote the large-scale digitization of documents and encourage the definition and adoption of international standards. Furthermore, it should demonstrate how digitization might improve long-term preservation and allow easy access to such documents. The project will rely on major libraries, and will be extended to countries other than those participating in theG7. The works selected will be displayed in their original languages, possibly in a multimedia environment. Specified documents would belong to the public domain.

As mentined earlier in this article, many countries have undertaken projects, often for centuries, to preserve, describe, and index cultural heritage. More recently, many libraries and information centres have begun to digitize documents. Digitization is a first step towards the creation of multimedia databases. Precious manuscripts and literary works, music scores and sound recordings, photographs and motion pictures, can be highlighted by digital information.

Bibliotheca universalis should provide a practical framework for international cooperation. Creating an open environment, it should attract a large participation beyond G7 countries. It should develop new trends in sharing cultural and knowledge resources, giving the end user the opportunity of working in a friendly environment. It will address the needs of the general public, researchers, scholars and students.

Bibliotheca universalis provides the guidelines for a distributed multimedia information system. All the information bases,managed by local entities and national authorities in charge of digitizing and indexing the contents, will be available to G7 countries and world citizens throughout existing and interoperable networks and simple terminals.

Multimedia access to world cultural heritage

This projectdesigned to ensure the interoperation of networks for open multimedia access to major museums and galleries. The goal of the project is to accelerate the digitization of collections and ensure their accessibility to the public (e.g. schools and universities). This include virtual access to buildings and artifacts through a natural language access system based on Internet/Mosaic; the creation of a virtual gallery that would allow the public to interact in real time with museums and galleries; and an "electronic neighbourhood" network system allowing citizens to communicate with museums and galleries.

Technically it was not difficult to include into proprietary tools multi-media capabilities using the Internet technology and the World-Wide-Web pardigm Hopefully we do not have to through away our proprietary products : we then protect our investment, enhance them at low-cost, and provide global acces to our cultural heritage. The industry and the commercial structures are moving on to the network, and politically there is a will at the national and international level to participate in the information age.

We are now facing some challenges :

  • The Internet is becoming a media, like others. What will it be in the years to come ? Will it still have this " educational " and " cordial " goal ? Will it still help people to co-operate or will it be the mirror of more intense economic competition ?
  • Will countries in development,especially those countries in Africa, have real access to the network ?

We also need the necessary hardware and software for servicing hundred of thousands of images, and a way to navigate efficiently . At present, the Interenet is like a library, a huge library, with no complete catalog capabilities. We will need something like ISBN for books and link servers.

We discovered the need for link servers in an unexpected way. An important archaeological discovery of a 30,000 years old painted cave (the oldest painted cave in the world) was made in December 1994 in France and announced in January 1995. Since this site will not be open to the public, we decided to put some of the drawings on our World-Wide Web server The next day our Internet traffic increased 22 timles : we had to change our 64k leased line to a T1 leased line and our Internet service provider had to upgrade its router and its access to the E-bone network. We had also to move our WWW server to a more powerful hardware. (That was a pain. Not really for us, but for the people who had hard-coded in their servers links to our server. ) We must find a way to have a dynamic process here, not just hard-code. Links servers seems an interesting concept to achieve that goal.

We were astonished by the publicity we have had just by putting four pictures on the network: articles in the Washington Post, Newsweek, International Herald Tribune. Archaelogy seems very popular and we have now set-up a virtual exhibition on under-water archaeolgy that receives a lot of traffic.

Conclusion

It is clear that these new tools will help us in achieving one of our goals: making known France's cultural heritage. The launching of this service should also help industry in providing value added multimedia products, and encourage people to choose France as a travel destination. Although we face many challenges on the technical, legal and political side, and also some risks, the promotion of cultural heritage is worth it.

All the services here described, and many more, are available at the URL:

http://www.culture.fr


[Seitenanfang]