Abstract
Cultural content and cultural assets are increasingly seen as invaluable drivers for education, entertainment, tourism, knowledge, and shared memory initiatives. As such, cultural information, as stored in cultural institutions such as museums, archives, and specialized libraries is increasingly demanded by several actors, from public policy, tourism, app developers and software companies, schools and the citizen at large. To satisfy the needs of such a large spectrum of consumers and end users, the information must be format independent, multi-lingual, multi-purpose, and make use of standards, norms and national and international recommendations. In this is paper we describe our past and current efforts to develop cultural information systems that can satisfy those requirements and those of the specialized public, such as scientists and curators. We describe an inventory and content management system and the conceptual and architectural choices that were made to allow its evolution, standards' compliance and multi-purpose use.
TopIntroduction
Cultural content and cultural assets are increasingly seen as invaluable drivers for education, entertainment, tourism, knowledge, and shared memory initiatives. As such, cultural information, as stored in cultural institutions such as museums, archives, and specialized libraries is increasingly demanded by several actors, from public policy, tourism, mobile app developers and software companies, schools and the citizen at large. To satisfy the needs of such a large spectrum of consumers and end users, the information must, as much as possible, be format independent, multi-lingual, multi-purpose, and make use of standards, norms and national and international recommendations. As Libraries, Archives and Museums are arguably the oldest examples of knowledge based organization (Hedstrom & King, 2003) it is natural that they are at the forefront of the web evolution. Cultural institutions have been using several inventory management systems, and investing in cleaning, formatting and adapting data to be shown on public portals and mobile applications. Some institutions have been exporting their data, encoded in some metadata format, to aggregators with national or international dimension such as Europeana.
After data is published on the web there is a growing need for linking it from different places, themes, languages, and culture so that it can be part of the linked open data world, and be used to its full extent. It becomes virtually impossible for an institution to cope with so many requests for information, in different formats, encoding standards, languages and level of detail. Requests can come under the form of sharing in social networking sites, supplying data to content aggregators, providing programming interfaces to app developers and other end users, participating in joint portals, and virtual exhibits, for example. Additionally, published data should be correctly found by search engines, and made accessible to the general public. The use of right metadata, together with search engine optimization techniques can help disseminate a cultural institution’s content. The web is an interlinked collection of resources, identified by what is called a URI (Uniform Resource Identifier). A kind of URI is what people generally type in a web browser to access a given site. To be able to make available their data so that it is found, either by searching or by looking from other data, institutions need to take a systematic approach to web publishing by providing as much context information as possible.
The so called semantic web is a web with context information, organized in a way to make search and navigation easy and accurate by providing only relevant information. The semantic web is designed to make search and discovery more efficient, and to navigate from one resource to the other as one would normally do in a web page. For example, when searching for the painter Picasso there would be irrelevant results, like those related to a car model with that same name, software packages or generic pages mentioning Picasso. Discovery is also an important behavior, as people would normally move from one page to the other by following links. The semantic web should keep and encourage that same behavior, mainly if computer programs are browsing the data as this is the only way they can move from one page to another.
The linked data movement proposed the Resource Description Framework (RDF) syntax to represent information to be exchanged (Bizer, Heath & Berners-Lee, 2009; W3C, 2014). There were several RDF vocabularies available then, and their number and availability continues to increase. Anyone can define and publish a RDF vocabulary, making it available such that other people can reuse it (see for example http://lov.okfn.org/dataset/lov/ for a database of more than 600 available RDF vocabularies).
Tim Berners-Lee (2006) summarized and defined the following characteristics for the semantic web, a web of linked data:
- •
Use URIs as names for things.
- •
Use HTTP URIs so that people can look up those names.
- •
When someone looks up a URI, provide useful information, using the standards (RDF, SPARQL).
- •
Include links to other URIs so that they can discover more things.
Key Terms in this Chapter
Crosswalk: Process using rules allowing transformation of data from one given metadata format to another.
Cultural Open Data: Data pertaining to the cultural domain that is published in the web according to the Open Data principles.
Semantic Portal: A web portal allowing navigation by following semantic relationships between data.
Linked Data: Data that is published on the web according to the principles set by T. Berners-Lee.
CIDOC CRM: The conceptual reference model proposed by the CIDOC for the cultural heritage domain.
Metadata: Data about data, in practice the set of properties characterizing data in a given domain.
Conceptual Model: An information model providing definitions and the structure for representing information in a given domain.