Open CarlaVS opened 2 years ago
Some follow-up on the FRBRoo/BIBFRAME discussion:
Representing biography texts as part of their publication:
"If an instance of F2 Expression is of a specific form, such as text, image, etc., it may be simultaneously instantiated in the specific classes representing these forms in CIDOC CRM. Thereby one can make use of the more specific properties of these classes, such as language (which is applicable to instances of E33 Linguistic Object only)."
Overall, I think either approach would work, and which one is more suitable will depend on what kind of information we would want to express with these. If BIBFRAME is enough, it's probably a better choice due to it's simplicity. Whereas, if the bibliographic items themselves are of interest, then FRBRoo might be a better choice due to it's expressiveness.
We could also declare the article (modeled with Bibframe)) as an instance of crm:E73 Information object and add a P3 has note property to store the original text in.
Can we pick up the discussion regarding the modelling of the bibliographic information again. @razz0 and @CarlaVS . What would you suggest to use? We are mainly interested in the authors of the biographies and the publication information of the volumes these biographies were published in. Any other suggestions @biktorrr ?
Actually, in D3.4 we have stated that we use BIBFRAME:
"In the InTaVia bibliographic provenance graph, BIBFRAME is used to model the data about the source where the biographic information comes from originally, up to the level of an article. A relation between the idm:Person_Proxy (which represents the perspective of this source on a persons biography) and the biography article makes this data available for querying. BIBFRAME was developed by the Library of Congress since 2011 and is updated regularly. It was decided to use BIBFRAME for this module of IDM-RDF because it allows to cover all relevant aspects of the bibliographic sources and it includes the property bf:references which enables the documentation of sources of an article, which are available for three of the four biography sources (BiographySampo, Slovenska Biografia and APIS). For the InTaVia bibliographic graph the classes of bf:Work (conceptual level) and bf:Instance(individual, material embodiments of the work) are relevant. There is no need to represent an actual copy of a biography lexicon (bf:Item) in the InTaVia Data. The class bf:Work is used to store metadata like the subject of the article. A biography lexicon(bf:work) and an article (bf:work) have the relationship bf:hasPart, which documents that the article is “included either physically or logically in the described resource” (see Fig. 11)."
Besides the database and technical provenance we have to model the bibiographic provenance. For the OEBL/ APIS e.g. we have the number of the volume of the biographic lexicon and additional bibliographic information. The suggestion would be to link every proxy to one or more sources with Cidoc CRM (P70 documents E31 Document) which are stored in an own bibliographic named graph. The obvious idea would be to model these information in frbroo.
Are there other suggestions for the bibliographic source modeling? Or suggestions for other good bibliographic ontologies which are coherent with Cidoc CRM?