The rich cultural heritage preserved in the Belliniano Civic Museum of Catania has been studied and promoted in the last years mainly thanks to the BellinInRete project. It includes collections of objects (or resources) of a very different nature: paintings, photos, pianos, autograph scores, manuscript leaves, books preserved in the Museum's library, etc. In order to make the Belliniano Museum's heritage interoperable and reusable by scholars and cultural operators, we propose to semantically organize it in a unique homogeneous container, the OntoBellini ontology, designed and developed according to the Linked Open Data and Semantic Web paradigms. The wide variety of the involved museum resources, not even fully digitalised and catalogued, led us to the idea of experimenting with the RDA (Resource Description and Access) standard for creating library and cultural heritage resource metadata. In this paper we describe the ongoing work towards the realization of the OntoBellini ontology.
The cultural heritage preserved in the Belliniano Civic Museum of Catania includes collections of objects (or resources) of a very different nature; in fact, the currently identified resources consist of about: – 250 objects among paintings, photos, pianos, brooches, watches, furnishings, posters, medals, textiles, etc.; – 4; 500 manuscript leaves of documents and letters; – 9; 300 sheets of manuscript scores; – 1; 900 printed scores; – 50 printed music booklets; – 280 books of the museum library; – 60 vinyl records of various musical compositions.
In the last years, this rich cultural heritage has been promoted particularly by
BellinInRete project [
In order to make the Belliniano Museum’s heritage interoperable and reusable by
scholars and cultural operators, we propose to semantically organize it in a unique
homogeneous container, the OntoBellini ontology, designed and developed according to
the Linked Open Data and Semantic Web paradigms. The wide variety of the involved
museum resources, not even fully digitalised and catalogued, led us to the idea of
experimenting with the RDA (Resource Description and Access)1 standard for creating library
and cultural heritage resource metadata [
In this paper we describe the ongoing work towards the realization of the OntoBellini ontology.
The paper is organized as follows. In Section 2 we briefly review some related works. Then, in Section 3 we describe the work carried out and in progress concerning the analysis and the representation of the Belliniano Museum’s resources, motivating the exploration and exploitation of RDA to build up the OntoBellini ontology. Finally, in Section 4 we draw our conclusions and discuss some hints for future works.
Various proposals concerning the semantic organization of museums cultural heritage
have been presented over the years. A lot of them relies on the CIDOC Conceptual
Reference Model (CRM)2, which is the international standard for the controlled exchange of
cultural heritage information since 2006. CRM provides a general core ontology which
can be adopted in cultural heritage contexts to develop semantic web-based information
systems, and to improve information sharing. Based upon CRM, semantic web models
have been devised aiming at enhancing semantic expressivity in the cultural heritage
domain and to address specific issues not completely covered by other existing models.
This is the case, for instance, of the OA entry and the F entry ontologies presented in
[
Recently, the RDA standard has emerged in the context of cultural heritage promotion. The main objectives of RDA are to identify and to relate entities at an abstract level. Originally, RDA implements the Functional Requirements for Bibliographic Records (FRBR) data model, classifying information resources in terms of a four-level entity hierarchy called WEMI (Work, Expression, Manifestation and Item)3. In fact, in its Novem1http://www.rda-rsc.org/ 2http://www.cidoc-crm.org/ 3https://www.ifla.org/best-practice-for-national-bibliographic-agencies-in-a-digital-age/ node/8915 ber 2016 meeting, the RDA Steering Committee subsequently agreed in adopting the IFLA Library Reference Model (LRM)4 as a conceptual model for the development of RDA5, superseding FRBR.
RDA aspires to provide a universal standard for data recording, a unique code for representing heterogeneous resources that can be found in: (A) libraries (manuscripts, books, music and movies); (B) archives (institutional documents, personal and family papers and business documentation); (C) museums (works of art, costumes, artifacts and natural objects and photos). To be highlighted that within the Italian context, the resources related to the union Libraries, Archives and Museums are managed, through the use of sound and acknowledged standards, by the respective institutions ICCU, ICAR and ICCD, coordinated by MAB (Musei, Archivi, Biblioteche) Italian Association6.
Within the BellinInRete project, the Belliniano Civic Museum’s heritage has been partially studied and analysed by musicologists, and by other museum personnel with archival and library skills, aiming at retrieving information about the museum resources. The museum resources have then been formally represented as data records comprising different fields of information (see below). The collection of these records forms the basis which our proposed semantic organization of the Belliniano’s heritage relies upon7.
Records for the museum resources have been created following the ICCD and ICCU Italian standards for cataloguing and documentation (Scheda OA, Scheda F and SBN cards)8. The number of fields in each record, and their meaning, depends on the type of resource represented by the record itself. There have been identified 14 different basic types of museum resources, namely9: – Manoscritti (Manuscripts) – Testi a stampa (Printed texts) – Musica manoscritta (Manuscript music) – Musica a stampa (Printed music) – Materiale grafico (Graphic material) – Arredi (Furniture) – Dipinti (Paintings) – Documenti (Documents) – Foto (Photo) – Medaglie (Medals) 4https://www.ifla.org/publications/node/11412 5http://www.rda-rsc.org/ImplementationLRMinRDA 6http://www.mab-italia.org/ 7Notice that, actually, the number of the created records correspond to about 70% of the total number of estimated museum resources. The involvement of the remaining resources is planned for the early future. 8http://www.iccd.beniculturali.it/, https://www.iccu.sbn.it/it/ 9Notice that the italian language has been used to record resource information.
– Statue (Statues) – Strumenti musicali (Musical instruments) – Tessuti (Textiles) – Oggetti generici (Generic objects) (Notice that, for completeness, in the list above we have reported the English translations (enclosed in parentheses) of the Italian terms (written in boldface) originally adopted to name basic types.) Within each basic type, the resources are subdivided, in turn, into more restricted, specialized subtypes. For instance, Manoscritti includes: letters, draft letters, copies of letters, baptismal certificates, death certificates, marriage certificates, expense reports, medical bulletins, etc. The Materiale grafico includes the posters, while the scores fall within the Musica manoscritta. Oggetti generici involve items owned by Vincenzo Bellini, such as watches and brooches, and other every-day life objects like spoons, knives, cups, etc. Figure 1 reports a selection of record fields in tabular form. Note that the information block stored within some record fields, presents a high level of granularity (coarse-granularity) which could be further refined by splitting the block across additional (atomic) data fields. This is the case, for instance, of the formato field (see Figure 1: the second to last column of the top most table) which is used to composedly describe certain physical characteristics of a manuscript such as dimensions, number of pages, foliation, writing direction, etc. It is worth mentioning that certain record fields are specific to the particular type of museum resource described by these fields. For instance, the lingua field (cfr. Figure 1) has been specifically designed to represent the language(s) of written resources. Similarly, it makes no sense to talk about the foliation of a table or a chair (cfr. the field formato). We mention further that the Belliniano Museum’s heritage even involves certain compound physical objects (such as medal containers and photo frameworks) which require a hierarchical record structure to be reasonably described10. Additionally, the Belliniano preserves also some music booklets which have not been catalogued yet. Also, we underline that several archival documents (like the various certificates), have received a rough analysis: it is planned to create detailed metadata for them by following the standards adopted by the National Archiving System of Italy (Sistema Archivistico Nazionale)11.
As it emerges from the considerations above, the created representations of the Belliniano Museum’s resources present a coarse, heterogenous character which makes it difficult to effectively translate them into an homogenous knowledge base. (Note that this partly derives from the particular representation criteria adopted for creating the data records corresponding to the museum resources.)
In order to improve such representations it would be useful to first clean, refine and then possibly merge the recorded data, so as to obtain a more uniform record collection. Then, RDA guidelines could be fruitfully exploited to get a more effective data (re)organization. In fact, RDA, through the WEMI classification mechanism (which inherits from FRBR), lends itself better than (e.g.) CRM to the description of both museum collections (composed of unique objects, and for which CRM has been primarily designed) and bibliographic resources held by libraries, as well as archival documents12.
In very general terms, the main activities involved in the (future) development of the OntoBellini ontology, can thus be schematized as follows. After a first phase of data restructuring, aiming at creating more homogeneous, finer-grained record collections (as described above), it is planned to identify the main concepts and properties underlying the OntoBellini ontology in compliance with RDA’s entity-relationship framework.
In this paper we have proposed to semantically organize the rich cultural heritage preserved in the Belliniano Civic Museum of Catania into a shared ontology–the OntoBellini 10Actually, such compound objects have not yet been disassembled for conservation reasons, and therefore, at the present, it was possible to retrieve only poor descriptive information for them. The created records for these resources are in fact poorly informative.
11http://san.beniculturali.it/SAN 12We mention that, basing on the IFLA FR models, an extension of CRM has been developed, namely FRBRoo (http://www.cidoc-crm.org/frbroo/home-0), which intends to represent the semantics of bibliographic information and to facilitate the integration and interchange of bibliographic and museum resources. However IFLA LRM, as an underlying conceptual model of RDA, allows for a greater level of generality than FRBRoo, as it includes less details compared to this latter. ontology–, relying upon the large amount of currently recorded data for the museum resources. The heterogeneous character of the recorded museum data, requires however a preliminary reorganization of the data itself in order to make it more homogeneous and easily encoded into the ontology. To this end we plan to exploit the RDA guidelines for creating library and cultural heritage resource metadata.
This work was partially supported by the “Patto per Catania” under the “Fondo Sviluppo e Coesione 2014-2020: Piano per il Mezzogiorno”.