Section I: Description of the overall project

1. Marketing aspects

1.1 Description of the services
1.2 Market analysis
1.3 Customers

2.1 Costs
2.2 Preliminary financing plan

4. Technical aspects

4.1 Overall technical description
4.2 Technology and standards used

5. Project maturity

6. Socio-economic benefits

7. Justification and eligibility for Community aid

The places of culture (museums, monuments, archeological sites, libraries and historic archives, etc.), the main goal of this type of tourism, are certainly the heart of the system which, however, has to also include hotels, restaurants, welcoming places, transport, flow organization and management mechanisms, etc., in general all those components which improve the quality-price ratio of the offered system.

The Heritage UK market analysis below gives a rough indication on forecast and likely funds available to Internet, multimedia and VR technology projects. Source information is from the National Lottery Annual Report & Accounts 1995/6 they have been used to estimate the percentage of funds available to technology based projects.

• Millennium Fund - £800m 10% of Fund Approx. spend =£80m - Growth
• Heritage Fund - £800m 4% of Fund Approx. spend = £32m - Rapid
• Arts Fund - £800m 8% of Fund Approx. spend = £64m - Emerging
• Sports Fund - £800m 2% of Fund Approx. spend = £16m - Restricted

The figures sketch the potential of the Lottery market (total of £192 million) calculated over a three-year period, June 1997 to 2001. The majority of technology spend is likely to occur during late 1998 through to 2001 once the infrastructure and building projects are completed.
MOSAIC’s main objective is to make art available to the public. It will encourage cultural exchange between countries in Europe and the rest of the world. This is fundamental because recent studies (Eurobarometer 09/97) have shown that the number of visitors from Japan and America interested in our art is far greater than the 28% of Europeans interested in getting ‘on line’ to view art.
So, MOSAIC can be seen as an operative branch of the "Memorandum of Understanding" backed by the EEC. Already around one hundred museums and other operators in the field (around 400) have joined to create the biggest possible multi-media communication system.
In the feasibility study phase an analysis of the target user has been done. In fact, before to introduce a new service on the culture is very important to better outline the users expectation and needs. A preliminary list of users is, in any case, take into account in our commercial analysis (specialists, researchers, students, tourists, tourist operators, adept employees,..).

Considering the fact that the first step of the CSC services is manly based on the network dissemination of existing cultural application, we have put our attention to the Internet world of users. The following figures illustrates how the Internet can offer an easy way to get wired with the system.

The major considerations that can be resumed are the following:

• Access Systems
• 28% use PCs at Home (22% interested)
• 15% have CD-Roms (24% interested)
• Internet usage low (4%), but interested high (24%)
• Scandinavian & Benelux countries have highest use (and interest) in the Internet

• Internet / WWW users in the EU are predominantly
• Male
• Young
• Highly educated & still studying
• Managers & White Collar Workers
• & have high incomes

• On-Line Museum Visits
• 28% are interested. 19 million people
• Interest is highest in Southern Europe & Scandinavia
• Most willing to pay are :
• 25-54 years olds
• Highly educated & still studying
• Managers & self-employed
• & have high incomes
• Possible new revenue stream of 1 bn/ECU/year

The following table illustrates the number of existing Web pages in EU countries :

• 34% of Web sites are in the local language only
• 29% of Web sites are in English only
• 30% of Web sites in the local language and English
• 7% of Web sites are in multiple languages in addition to English and the local language
• only two Web sites in two languages, neither of which was in English

• 64% of Web sites have some English pages
• 19% of web sites have some French pages
• 8% of Web sites have some Spanish pages
• 7% of Web sites have some Italian pages
• 3% of Web sites have some German pages

So far, MOSAIC has been the only project, financed by the EEC in the TEN-Telecom program (Trans European Networks Telecommunications), that is actively involved in and dedicated to our cultural heritage. The TEN-Telecom framework will, in this case, promote and support the creation of International networks for viewing art through multi-media.
So, MOSAIC can be seen as an operative branch of the "Memorandum of Understanding" backed by the EEC. Already around one hundred museums and other operators in the field (around 400) have joined to create the biggest possible multi-media communication system.
Mosaic represents a particular initiative as far as we didn’t see any other global project concerning both services and products on the field of cultural heritage already activated in a organized structure as this one.
We could consider important projects as AMICO and CIMI but as far as we have already reach an agreement with them to cooperate in the field of cultural heritage we must consider them more as an integration of our project than a real competitor.

MOSAIC network is linked to advanced technology ISDN and ATM and its nodule points in technological and service management partners are: Infobyte of Roma, Politecnico of Milano, CINECA of Bologna (GARR and the University network, a consortium of colleagues of 13 Italian universities: Ancona, Bologna, Catania, Ferrara, Firenze, Modena, Padova, Parma, Siena, Trento, Udine, Venezia) ZGDV of Darmstadt (the network of the universities of Germany, Portugal and South America), Arenotech of Paris, JCR of Graz, Japan (the private and university network) USA, Canada, Egypt and other countries. The broadcasting Centres can rely on the Banca di Roma distribution network (1200 branches open to the pubic) and 300.000 consumers of the experimental network ATM by Bell. Both in the ABI network (Italian banking Association) and with the board for Cultural Relations of the Italian Ministry of Foreign Affairs have shown interest in becoming members in order to spread Italian cultural heritage through a diplomatic network abroad. Firstly through Italian Cultural Institutes and then through Italian Embassies and Consulates throughout the world. A technological agreement is going to be defined between MOSAIC and CHIN (Canadian Heritage Information Network).
An interesting experiment will take place by the end of ’98 at the Shoenbrunn castle in Austria. The first MOSAIC Service Centres will be started so the products and services can be tested by the public. There are numerous industrial partners involved in the MOSAIC project, here are some of them: Hewlett Packard, Silicon Graphics Apple, Telecom Imation, Microsoft and Italtel. Hewlett Packard, in particular, will take part in the star up of the Image Server and the relative techniques of the management of images. Imation is taking part with its years of experience in the field of photography (Ferrania brand) and medical images whereas Italtel is participating with the setting up of network technology and user points.
In particular, for the trial validation and verification phase the CSC network, will be enlarged to the Sweden, Portugal and Spain. And will be reinforced the CSC distribution on the already present countries: France, Italy , Germany and Austria.

A relevant part of the interlocutors, contacted during the workshops, have already demonstrated a concrete attention to the MOSAIC project. The list of the additional partners is the following :

• Schloss Schoenbrunn
• Österrechische National Bibliothek
• AIT - Angewandte Informationstechnik - Graz
• Associated of Hofburg
• Albertina – Wien

• Virtual Presence Limited - Sale Cheshire
• VASARI Enterprise ltd – London
• Cambridge Management group

• Cité des Sciences et de l’Industrie - Paris
• SIM TEAM - Paris
• Fondation Sophia Antipolis
• Club MITSA - Sophia Antipolis
• Cryo-Networks – Paris

• ART+COM
• Museum fur Antike Schiffahrt - Mainz
• Romisch-Germanisches Zentralmuseum - Mainz
• Bildarchiv Preussicher Kulturbesitz - Mainz

• Accademia Massimo - Roma
• AIACE - International Association of Computing in Archaeology
• Apple Italia – Milano
>
• Archivio Centro di Catalogazione Beni Culturali Villa Manin - Regione Friuli Venezia Giulia
• Archivio Storico Bulgarini d'Elci
• Archivio Storico Diocesano di Milano
• Associazione Bancaria Italiana – Roma*
• Bancaria Editrice – Roma*
• Biblioteca Capitolare di Verona
• Canadian Embassy of Rome*
• Cap Gemini Italia S.p.A.
• Capware – Ottaviano, Napoli
• CNR - Consiglio Nazionale delle Ricerche
• Cooperativa archeologia, s.c.r.l. per la valorizzazione del patrimonio culturale - Firenze
• Ente Interregionale - Progetto Europeo "Herity" - Roma
• Fondazione Antonio Mazzotta – Milano
• Fondazione Longhi - Firenze
• Fondazione Ricci Oddi, Piacenza
• Fondazione San Paolo di Torino
• Galleria Nazionale d’Arte Moderna - Roma
• Istituto Italiano per l'Industria Culturale - Roma
• ITALTEL S.p.A. – Milano
• Luiss - Libera Università Internazionale degli Studi Sociali
• Marsilio Editore - Venezia
• Mediatech, Roma
• Ministero Affari Esteri – Roma*
• Ministero per i Beni Culturali e Ambientali - Sopraintendenza per i beni ambientali e storici per l'Abruzzo, L'Aquila.
• Musei Capitolini - Antiquarium Comunale - Roma
• Museo dell’Orto Botanico - Roma
• Nautilus – Capri
• Progetti Museali S.p.A.
• Progetto: "I Musei Virtuali della Certosa di Calci". Progetto concordato tra lo Studio ZYX, la Soprintendenza B.A.A.A.S. di Pisa
• Provincia di Milano, Assessorato alla Cultura e Tempo Libero
• Regione Lombardia, Assessorato alla Cultura e Trasparenza – Milano
• Scuola Normale Superiore di Pisa - Centro Ricerche Informatiche per i Beni Culturali
• Silicon Graphics - Milano
• Sovraintendenza delle Antichità e Belle Arti del Comune di Roma
• TELECOM ITALIA
• Università degli studi di Pisa. Supportato dal Comune di Calci e dalla Provincia di Pisa
• Università degli Studi di Trieste - Dipartimento EI – Trieste
• Università di Padova - Dipartimento di Elettronica e Informatica

• Centro de Computacao Grafica – Coimbra
• Universidade Do Minho – Retoria
• Museu Regional De Arqueuologia D. Diogo De Sousa – Braga
• Universidade de Tras-os-montes e Alto Douro

• Göteborg University - Department of Environmental Science and Conservation - Institute of Conservation
• Metimur Company (technical Partner)

• Centro de Estudios del Patrimonio – Madrid
• University of Jaén
• Universidad Autonoma de Barcelona
• Universidad Autonoma de Palma de Mallorca
• Ministerio de la Cultura – Madrid
• Museo Nacional de la Ciencia y de la Tecnologia – Madrid*

• Hewlett Packard Europe (Belgium)
• Imation S.p.A (borne of 3M Innovation)
• Microsoft Europe
• National Technical Museum - Prague

• R.I.T.S.E.C. - Regional Information Technology and Software Engineering Center - Egypt
• CRC Research Institute Inc.
• SUMS (System for Universal Media Searching) project - G7 - Toronto - Canada *
• Guggenheim Museum - NY (USA)
• CAD Center Corporation
• CALTECH - CA (USA -G7 partner)
• Capital Normal University - Beijing - PR of China*
• CHIN Canadian Heritage Information Network*
• Xerox Palo Alto Research Center - Palo Alto CA – USA*
• Lighscape Inc. - San Jose CA – USA*
• Multigen Inc. - Palo Alto – USA
• Microsoft Reseach Centre - Redmond WA – USA
• AMICO – USA*

The distribution of information between the CIS and its relative subsystems will enable the management at a centralized level of the following data:
MOSAIC Node definition;
Artwork Definition and Classification;
Events.
The MOSAIC Nodes will be given the task of managing specific information of each single institution, i.e. actual data regarding the artworks, such as the Definition, Classification, Repertoires and Specimens. Interaction between the CIS and its will necessarily be made possible by the use of unique identifiers for each artwork.
This organization, deliberately distributed and potentially heterogeneous, will be adopted to reflect what seem to be the most common situations in the target structures (museums and galleries). In the aims of implementing a project of general value, the need to integrate already functioning specialized systems or subsystems must be taken into consideration, so as to make the implementation of the proposed structure as feasible as possible.
The dividing of nodes in a hierarchical structure and the splitting of information between the MOSAIC CIS and the local Nodes also will offer a number of inherent advantages with respect to centralized implementations on a single server. In the structure here presented, operations entailing the maximum exploitation of resources, i.e. retrieval of specimens and repertoire information, are carried out at a local level, without requiring a transfer in bulk of information towards the MOSAIC CIS: such a solution could also lead to the optimization of the infrastructures and a large flexibility in the choice of the appropriate network connections for each Node.
The Information system’s Database
The choice of the database management system underpins the general philosophy of the project, and inspires a thoroughgoing analysis of the main features of the Database Management System (DBMS).
The considerable improvements regarding services and reliability of relational database management systems (RDBMS) have brought, in recent years, to their increasing use in a number of different sectors. The benefits deriving from the use of RDBMS in the MOSAIC project could indeed be remarkable, especially considering the dimensions of the managed archives and the complexity of data structures. The techniques and platforms adopted to implement the many different and locally customized databases could be extremely diversified. Starting from small, Stand-Alone, systems on a personal computer, then reaching the large Client-Server system used for the management of the entire Mosaic Information System (MIS). Moreover, in the particular case of large structures, implementation techniques should be based on Client-Server architectures, guaranteeing a better, more flexible distribution of information.
Thus, notwithstanding the specific characteristics of each local structure, information management within the Mosaic framework will be formed by a great number of independent applications capable of satisfying local needs and, at the same time, capable of maintaining a sufficient level of communication between each other. The existence of a wide variety of local management systems, already available and used for artwork classification in several museums and galleries, suggests the need for a solution capable of integrating such systems as independent nodes of the Mosaic network, rather than replacing or duplicating the databases with a standard database model.
From this point of view, the use of relational database servers in each single implementation today guarantees a substantial level of interoperability, in line with the philosophy of the distributed databases.
The philosophy of the Mosaic network should therefore take into account the existence of independent subsystems, which anticipated the choice of a global policy encompassing the entire structure - a choice which was facilitated by great technological progress and by the availability of several solutions for the implementation of databases.
The distributed system resulting from the integration of the single implementations can no doubt be defined as heterogeneous: the problems stemming from the use of different RDBMSs, typically installed on different platforms, are many and not always solvable. Considering the state of current day technology, one may say that integration is possible in general, and that the considered systems are able to exchange information transparently.
The Communication Protocol
The adoption of a standard communication protocol for the exchange of data between the MOSAIC CIS and its Nodes is evidently one of the major concerns of this project. Several consolidated protocols are today available, but most of them can not be easily be used in the above-mentioned heterogeneous network. In many cases, in fact, the standards simply provide an efficient means for transmitting information from one node to the other with an "all or nothing" solution for the meaning of the information itself. In other words, depending on the standard adopted, the receiver has a priori knowledge about any token of information, or simply receives a block of unknown information. The definition of a complete "data dictionary" for the classification of the information used in the context of the MOSAIC project is evidently the cornerstone of the system itself, and should be integrated with the definition of a communication protocol capable of a precise transfer syntax, to ensure that information is correctly transmitted and received, and of flexible semantic rules, grouping the many tokens of information defined by the dictionary into several modules, each having their own rules. For example, the protocol should define the attributes "Title" and "Location" for an artwork, but should also require the presence of the "Author" attribute in a "Artwork Identification Module".
The use of an appropriate transfer syntax alone can not be considered satisfactory, since the different institutions could transmit heterogeneous information. On the other hand, the adoption of fixed semantic rules could be too rigid for the management of a large number of nodes, since the data dictionary could be progressively expanded and the organization of the data modules could be changed. According to these considerations, the solution proposed is based on the use of DICOM, the leading standard for the exchange of heterogeneous data (text, images, sequences, audio, etc.) in the medical field. This standard largely proved to be efficient and robust in many mission critical applications, and is suitable for any customization to the specific needs of the MOSAIC network.
The DICOM standard was developed by the American College of Radiology (ACR) and the National Electrical Manufacturers Association (NEMA), and is targeted to the exchange of digital images between different devices, the development and diffusion of Picture Archiving and Communication Systems (PACS) and the interfacing of different systems. DICOM is based on a relational model of the real world, and does not constrain the implementation and the management of data on each node archives. The standard is Object Oriented, and is therefore based on the definition of abstract data modules, each including several attributes, which can be grouped together to create the Information Objects (Information Object Definition, IDO). Each actual object of the real world is therefore described by the attributes of an Instance of the related Information Object. The definition of the data dictionary (i.e. of the many attributes available and of their characteristics), of the Data Modules and of the Information Objects is considerably flexible, and can be easily customized to fit the requirements of any application. The set of actions or operations that can be applied to the Information Object is defined through a number of service primitives, called DICOM Service Elements (DIMSE). These elements are basically divided in two categories: operations and notifications. Any communication through the DICOM protocol is therefore based on the coupling of an Information Object and a Service Element, thus defining a Service-Object Pair (SOP). The association of the same Object with a number of suitable Services defines a SOP Class. Each node of a DICOM network could implement one or more SOP Classes, playing the role of user of the service (Service Class User, SCU) or provider (Service Class Provider, SCP). The protocol is structured in nine parts, as follows: Introduction, Conformance, Information Object Definitions, Service Classes Definition, Data Structures and Semantics, Data Dictionary, Message Exchange, Network communication, Point-to-point communication. The DICOM standard is ISO/OSI compliant, and can therefore be efficiently adopted in the MOSAIC framework by simply defining the needed Information Object, together with their attributes, and the coupling of these objects with the Service Elements, i.e. the SOP Classes.
The Information Model: The exchange of data will be based on an information model describing the objects of the real world to be managed in the MOSAIC framework. The Artwork can be considered the main object of the information model, since it describes the root information to be managed in the MOSAIC system. An artwork belongs to a MOSAIC Node, which could represent a museum, a gallery, etc., and could also be part of a collection of artworks. For each artwork, several Repertoires can be created, each describing a particular situation in the life of the artwork, as for example, in the case of restorations. Each repertoire can then group together several Specimens, each including homogeneous multimedia representations of the artwork.

To allow a flexible definition of the IODs, all the available attributes could be grouped into several independent Data Modules, as shown in the following table.