This doctoral thesis is about the possibility to put archival data (for example archival descriptions, authority records and so on) directly into semantic web, particularly using a technology called “Topic Maps”. Topic Maps are an ISO standard quite similar to RDF.To show that it's possible to express archival data directly into semantic web, I translated all archival standards (such ISAD, ISAAR, ISDIAH, ISDF) into Topic Maps Constraint Language schemas. TMCL is a standard of the Topic Maps family. What it's important is that you can declare constraints and inference rules to ensure that a topic map holding archival data is compliant to archival standards.So, once shown that is possible expressing archival data into semantic web, which is the main advantage?I think that we can find three advantages in this approach:you can build flexible and extensible information systemsyou can create import and export as linked datayou can merge data of different areas (for example you can merge archival authority records and librarian authority records) For that aim I also created crosswalks between different standards. For example I mapped FRAD with ISAAR.For import and export I created some XSL-T stylesheets to convert EAC and EAD to Topic Maps and I released them as opensource on google code.Lastly I proposed a strong idea to develop flexible information systems (for example a software to manage a digital library or an archival information system such the archival system for national archives). In those systems data wouldn't be stored in a database, but directly in RDF triples or, why not, in a topic map.This allows to extend or change the system without a database change (so you can add or change a descriptive field without touch the database).
展开▼
