Indexing Grey Resources: Considering Usual Behavior of Library Users and the Use of Dublin Core Metadata via a Database of Specialized Vocabulary

Grey resources spread over a large set of matters. In each field of interest, people have their own concepts, each group will use their own vocabulary. Most authors of grey resources were not able to determine the Decimal or Dewey classification themselves, nor are they able to fix themselves the appropriate terms to embed in the DC elements. It is then worth providing the web user with an interface that makes him feel at home, by adapting automatically the terms designing the fields of the search form to those which are in use in his domain of activity. The whole classification proposed cover about 110 different topics, each one having a set of terms chosen to entitle the fields of the interface of thematic search as well the submission forms. Moreover, seven entry points are proposed in the first step of the connection, driving the user towards the appropriate context in accordance with the purpose of the document to find: teaching and research; literature; technological documentation and data sheets; vulgarization and everyday life; citizen and social activities; arts, sports and spare time; bank of images. Indeed, the system must allow the flexibility that allow the replacement of any term by another if necessary. So the entire set of terms is put in a database. This system is also able to adapt the language chosen by the user. The submission forms filled by authors are parsed according to the topic selected and the Dublin Core metadata file––as an XML file––is generated and stored in the indexing server. It is delivered on simple metadata request, and is reachable by a hyperlink included in any page of search results. The experimental system has been built and evaluated over a period of five years, and was implemented in the site of the University of Poitiers since June 2005. The software is to be released under free software license (CECILL).