2006 Klaus E. Werner The CollectioOrg system is an Open Framework for Interconnectivity of Museum Knowledge Domains, defining a XML hyperdocument scheme for museum records with a mechanism of hyper-linking via URNs for interconnectivity between museums and external providers using a webservice in REST. hyperdocuments records interoperability webservice open source XML

The HyperRecord system is a framework for the interconnectivity of museums: it connects different museum’s knowledge domains and provides access to their information resources over the Web, defining common formats and protocols for storage, identification and retrieval. HyperRecord is born from EC project 0760/001001 CLT CA12 (Culture2000). At the moment, the following institutes are participating: Capitoline Museums Biblioteca Hertziana / Max Planck Institute Theological Library, Maribor The following are our main goals: Capturing all Types of Information Resources We capture (1) text documents (scientific articles, books, manuscripts &c.); (2) data (informations mostly organized in flat or tabular format); (3) mix of text and data (museum records, indexed archives &c.); (4) metadata of audio/video media (IPTC, XMP, RDF, &c.). Whereas normally only those information resources which are of the data type – and therefore easy to integrate in a traditional database – are preserved during digitization, the HyperRecord system aims to capture and preserve all information resources alike in order to preserve the entire information resource context. Information Storing them in Open Digitization Formats The digitization format of the various types of information resources is XML, which is able to store both document and data contents, and is an optimum solution for exchange across different platforms and for long-term storage, too. For the text documents, existing schemata like TEI, DocBook or DITA are used, for which a multitude of tools are available. A schema for museum object records has been created from scratch (CollectioML) and will be Free Software (GPL): museums, private and public collections can then use it freely to create their own digital records. Exposing them Using an Open Referencing System The different information resources are connected in a way that precludes any manual intervention. Instead of hard-wired links, Uniform Resource Names mark the individual items across the different document types: the actual link to a resource is no more than the result of a query to the repository. These URNs are made up of a museum identification code and the inventory number. Referencing Distributing them using a Federated Network Just as references are used to link up resources inside a domain, for use inside the museum LAN, they can also be made to link up to external repositories, whenever they are accessible over the web. This allowed to build a system where queries are forwarded across the network from the initial repository, which has only a reference, to another one which has actual ownership of the resource. Different domains thus communicate and exchange information by simply serving XML information in the requested form over a simple HTTP connection (the REST mechanism). Background Querying Here is an example for a cross-reference between various servers, in the form of a background-query on partner sites: Capitoline Museums, Statue of Giulio Caesare. Re-Using them in Multiple Channels Using a native XML repository as multipurpose container for storing, searching and retrieving all types of XML documents allows to expose the same information resources in multiple ways: using them for internal museum management, making them available to other museum collections and researchers, exposing parts or all of them to the general public in the way of a web portal. Finally, they can be exposed using a webservice for hooking up additional web publishing systems or allowing third parties to build additional services. Standards Based All technologies used are W3C, ISO or OASIS standards, such as: XML as source, exchange, and archival format all at the same time; RelaxNG for the schema language; XQuery as query language and for addressing and aggregating resources which can reside even in outside domains; XSL-T/FO for the transformations into other media; Unicode for the encoding; etc. FLOSS OpenSource Only open and non-proprietary standards can guarantee future compatibility between different platforms and systems, provide freedom of use, unlimited exchange and sharing and unmatched availability of tools. The full availability of the system without licensing costs and without fear of being trapped in proprietary technologies is a prerequisite for a collaboration between larger and smaller heritage organizations. For more information, contact kewerner@hyperrecord.net.