This poster gives an overview of the background and development of EXTRAMUEP [http://extramuep.mah.se], a stand-alone web application for DSpace data extraction, an application that aims at motivating content-owners to deposit more of their works as well as facilitating dissemination and administration of research. The main feature of this tool is the conversion of DSpace data into ready-to-use files in the user’s ‘Office’ applications.
EXTRAMUEP is developed for Malmö University Electronic Publishing [http://mah.se/muep], to ensure that data entered in DSpace is not only created and used, but more importantly re-used by the university itself. The workflow for the majority of entries in the system relies on authors registering their works themselves, and members of faculty reviewing the records before they are accepted. In in-depth interviews, we have received an explicit message from several researchers that, due to lack of time, parallel publishing in the institutional repository would only be prioritised if the organisation or funding agencies decided it to be mandatory or strongly encouraged. These responses made it clear to us that we needed to be pro-active and find other ways to ensure that research at Malmö University is made more visible and MUEP even more useful.
Hence we developed EXTRAMUEP. For the end-users EXTRAMUEP is a web-based interface where the user downloads a file that is opened in their favourite software used for spreadsheets and word processing.
The application itself can be adapted for other formats, layouts and field extractions. It is a stand-alone web-based PHP-solution, which calls the database directly and maps Rich Text Format and HTML 4.0 as the output formats. The application tells the browser to open the files in the software that is associated with the specific content types. The application for EXTRAMUEP will be documented and packaged for usage and further development by others in the DSpace community.
The current presentation format in EXTRAMUEP follows the university’s internal needs, which e.g. is used for statistical purposes and to determine keys for funding upcoming years. The information in the spreadsheets is similar to the one we have in DSpace, and we could see the benefits of getting the data extracted automatically for this purpose, and in a way that does not require end-users to use other software than the ones they are used to in their desktop environment.
In 2006 we made a manual copy and paste of the DSpace-data into spreadsheets, as an attempt to see if the metadata was good enough for the task. In the evaluations afterwards we found out that a majority of the researchers that had added their data in DSpace had a positive attitude to doing it again for 2007, and that the administrative staff could see a slight improvement of data quality this year. After this test in 2006, we find researchers that regularly add their publications to DSpace. We have also seen an increased amount of full-texts being uploaded into the repository and data added that is not mandatory, such as abstracts and links to full-texts etc.