DiVAContents PDF version What is DiVA?
DiVA, the Academic Archive Online (Digitala Vetenskapliga Arkivet in Swedish) is a collaborative effort of a number of universities in Scandinavia which offers both publishing services and technical solutions for local repositories. The DiVA system, developed and maintained at the Electronic Publishing Centre at Uppsala University, Sweden, supports workflows for both electronic publishing and printing.
Through DiVA, fulltext documents from the participating universities are published and archived ? hence the name Academic Archive On-line. Today the archive contains mainly doctoral and undergraduate theses and research reports, but DiVA also supports the deposit of preprints and postprints of scientific articles. Monographs and chapters can also be published.
The DiVA system started as a project in 2000 and has been in full operation since January 2003. It is now used by 15 universities in Sweden and Norway, all of which are co-operating in further developments of the system. This site, the DiVA Portal, is a common interface for the local DiVA repositories ? a joint search and browse service.
The DiVA system is known for its focus on practical solutions to support the longevity of electronic documents and to ensure long-term access. It is based on Java and XML technologies and it also provides various metadata services, such as harvesting via OAI-PMH and the generation of catalogue records for library catalogues.
The Electronic Publishing Centre
The DiVA system is being developed at the Electronic Publishing Centre at Uppsala University Library, Sweden. The work of the Centre focus on the development of technical solutions and well-functioning workflows for electronic publishing of various types of academic publications. The Centre aims at supporting electronic publishing in general and of Open Access in particular. This is achieved through the development of useful tools and services which support the publishing process and through the assurance of long-term preservation and dissemination of published documents.
The Centre itself is the result of a survey of electronic publishing of academic texts which was carried out at Uppsala University in 1998/99. The Centre actively monitors the development of electronic publishing (metadata, long-term preservation and access to documents in the future, searching and indexing etc). Moreover, the Centre is involved in a number of projects, directly or indirectly related to electronic publishing.
The DiVA project - a backgroundThe DiVA project started in 2000. A workflow was established and technical solutions were found for the electronic posting and publishing of doctoral theses in fulltext at Uppsala University. Already at this point, there was a need to publish other types of documents, such as research reports or undergraduate theses. The goal was to create solutions which would allow for future modifications and for new technologies and standards and to simplify the workflow for both authors and staff at the university. A vision was that the structured information originally created by the authors should be reused in many other contexts. A decision was taken to develop a new system - the DiVA system. The first version of the DiVA Publishing System, was put into operation in January 2003. At this time, the Electronic Publishing Centre was divided into two groups: production and research and development. In 2005, a third group was established: systems management and operations. The production group is now responsible for the production of printed and electronic publications. The research and development group is involved in a number of new research projects and continues the work on a new version of the DiVA system. The systems management group is responsible for support, operation, information and maintainance of DiVA. A network of researchers and developers in Sweden and other countries has been created through DiVA and new projects have been established. Since DiVA was put into operation it has been given a lot of attention on both a national and international level. The DiVA Publishing SystemSupported WorkflowThe DiVA system is built on standards, recommendations, and new XML technologies. Many parts of the publication workflow and most parts of the archiving workflow are supported by XML. Data is entered only once by the author, who may use either an on-line submission form or templates for word processors of their choice. Templates are also used in order to capture the structure of document content and to facilitate the writing of the document. The data originally entered is used as a basis for creation, reuse and enhancement of all metadata. For instance, some of the metadata is used to produce the cover and title pages in PDF-format. All metadata and where possible also the document content, is stored in a uniform XML-based format ? the DiVA Document Format (DDF). The metadata can easily be transformed into other formats and be disseminated to other information services. A persistent identifier is assigned to each item as well as checksums to all files. All items are stored in a local repository as well as in a local archive and can be searched and browsed through various web services. A copy of the item, bundled into an information package, can be sent for long-term preservation at an external archive such as a national library archive. | 
|
PublishingDiVA metadata supports different types of documents, printed as well as electronic. If the same document appears in different physical formats (printed or on CD-ROM) or in different electronic formats, the formats are called manifestations (in accordance with FRBR - Functional requirements for bibliographic records). The DiVA system is used not only for creation of metadata for different objects but also for creation of the digital original which forms the basis for both the printed and the electronic publication. Hence, it is possible to guarantee the full correspondence of the printed and electronic version. The DiVA Document Format (DDF) is used as a format for storage of metadata and also of fulltext files where this is possible. Most of the fulltext files in DiVA are published and stored in PDF, however at Uppsala University fulltext files are created in an XML-based format, as a part of the DDF. Apart from XML and PDF other formats such as PostScript are also used. The system supports any file format. See poster The DiVA Publishing System Uncovered. Administrative Tools for management of contentDiVA Manager is the main administrative tool of the DiVA publishing system. The tool can be reached from a simple URL in a web browser. To run the tool on the client a java runtime environment supporting at least version 1.4.x must be installed and enabled. The documents created from the templates can be uploaded via the DiVA Manager. The tool also supports other ways of delivering metadata, abstracts and fulltext documents and can easily be integrated in different workflows. The DiVA Manager makes it possible to register new records and update and edit old records. It supports special characters from Unicode. Formatting (such as subscript or superscript) can also be preserved. Complex mathematical formulas are converted to MathML and displayed on the web by plug-ins. The import function makes it possible to directly import a file and store it in the system. This tool is integrated in the Java applet and conversions between different formats (for example from MS-Word to XML) is done on the server side. There is also an on-line registration/submission form and an easy-to-use on-line administrative interface. The on-line registration form is used by authors to submit their documents while the administrative interface is used by university staff to publish submitted documents. The DiVA Document FormatThe DiVA Document Format plays an important role in the DiVA workflow. The format defines the elements of the DiVA-documents and their relations. The DiVA Document Format combines metadata elements with elements for structural mark-up. Currently it is defined by an XML schema. The format contains 99 metadata elements. Elements according to the DocBook DTD are used for structural mark-up (i.e. fulltext contents). For more information about DiVA metadata and for a presentation of various existing metadata formats that DiVA metadata can be mapped to, see the DiVA application profile Metadata and Metadata DisseminationThe 99 elements of the DiVA Document Format makes it possible to create document descriptions, metadata, of high granularity. Other formats, such as MARC, METS, MODS, Dublin Core, Endnote, Reference Manager and TEI can easily be created. Services Based on MetadataMetadata is disseminated to other information services. For example: - Transmission of metadata to the national union library catalogue (LIBRIS). All records are transmitted to LIBRIS as soon as they are published See image.
- Metadata can be harvested the Open Archives Initiative Protocol for Metadata Harvesting (OAI-PMH). It is possible to harvest MARC or DC records and also to select the records by sets of subject areas, organisation or type of publication
- An RSS channel displaying new publications is also available.
Search and Browse - User InterfaceThe public user interface on the web is an important part of the DiVA system. The interface is built on XML technologies and can easily be suited for different layouts. A simple plain text version designed according to the Web Content Accessibility Guidelines (WAI) is also available. Different language versions can also be added, currently all pages are available in Swedish and English. Search options include a structured search form and freetext searching (including abstracts and keywords). Also, fulltext search is available. Apache Lucene is used as the text search engine in DiVA, which gives a high performance searching process. Bibliographical information about all records can be downloaded in different formats directly from the DiVA Document Format. Preservation and Access in the Long-TermThe DiVA system is known for its solutions that support long-term preservation of electronic documents. For instance, persistent identifiers are assigned to each document in DiVA and copies of the publications are stored at different geographical locations Persistent IdentifiersAll documents published through the DiVA system are assigned an unique identifier -an URN:NBN identifier. In cooperation with the Royal Library a URN:NBN resolution service has been set up and put into operation. The task of a resolution service is to keep track of a certain document and redirect the traffic to the right location. Today a document is redirected to the local DiVA sites but, in a scenario where the local archive is shut down, it will be directed to the archive copy at the Royal Library. Central archiveAn archive copy of the document, bundled into a package, is sent for long-time preservation at the national library archive. An archiving workflow has been set up between the DiVA system and the national archives at the Royal Library. Local archives A copy of the document is also stored at the local archive. The copy is separated from the DiVA Publishing System and there is only a one-way communication between the publishing system and the archive. The fact that DiVA documents are stored at different geographical locations minimises the risk of data loss. Each manifestation is stored as a package. A package includes a file in the DiVA Document Format and a file that defines this format - today an XML schema - as well as checksums for all files in the package. There are also specific files for each manifestation. For example, if the manifestation is published in PDF a PDF-file will also be included. If a manifestation is published in XML format stylesheets for transformation and images will be included. Since the DiVA Document Format is a mandatory part of the package, all metadata can be reconstructed by extracting them from the DiVA Document Format See image of package. See poster (PDF) Access Now and in the Future (Eva Müller, 13-15/9-2004, ECDL, Bath, England). Co-operation within DiVAThe DiVA consortium was initiated in 2002 at Uppsala university, Sweden. Today, in 2006, 15 Scandinavian universities co-operates within the DiVA project. The DiVA consortium offers collaboration on further developments of the system and sharing of the experiences and technical solutions developed within the project. Two major user-group meetings are held each year and introductions are available for new participants. A technical helpdesk, e-mail list and a wiki web site is also available. The participating universities have agreed upon a common document format, common subject terms (allows browsing by subject) and a joint interface to local repositories - the DiVA-portal. Co-operation within DiVA is open to all universities and publicly financed organizations. For more information, see paper The DiVA Publishing System. The Community's Collaborative Development Approach (2005) Ongoing development activities- Extended rights and preservation metadata.
- Enhanced and more flexible submission workflows.
- Sophisticated search and browsing at document level.
- Advanced module for print on demand services.
- Archiving of fulltext documents in XML.
- Tools for semantic markup.
The DiVA PortalIn the framework of the cooperation a common interface to publications has been created - the DiVA Portal Metadata is disseminated from the DiVA-portal for the documents published in fulltext. The records can be harvested by the Open Archives Initiative Protocol for Metadata Harvesting (OAI-PMH). The portal is also a good example of interoperability and builds on common agreements regarding organisation and technology as well as formats and the granularity of metadata. EvaluationA demo version of the DiVA-system is available at http://www.diva-portal.org/demo/ Contact DiVA Support at: diva-support@ub.uu.se | 
