A number of projects at the UVic TAPoR node (Katakana Database, Graves' Diary, 17th Century London Map, Internet Shakespeare Edition, MaClure House) use eXist. Our staff have identified certain specific modifications to features in eXist and in deploying eXist which would render it more suitable for a wider range of users and projects.

In this initiative, TAPoR developers work with the lead developer of eXist to:

• implement a more sophisticated indexing system to support larger and more complex xml documents
• provide documentation for administrators (installation and maintenance) and developers/users (building, xquery, invoking)
• produce 3 downloadable, executable versions of eXist:
  
  • a war file of cocoon with eXist as a block
  • a jar installer for a standalone administrative client
  • a standalone executable consisting of Jetty / minimal cocoon / eXist

Once these goals are met, developers of the TAPoR projects using eXist technologies can move forward with confidence in the stability of the underlying platform. A stable release with a fixed feature set is of value to anyone working with eXist search engines, including other TAPoR nodes. This initiative will have very concrete results: it will be easier and less time-consuming for anyone installing the eXist engine and that engine will be more powerful than previous versions, thus making projects based on xml-databases easier to create and more capable.

The potential for further development in the area of open source database technologies (especially in relation to the archiving, searching and retrieving of large objects such as multimedia) is inherent in this initiative. With a stable eXist platform, and greater understanding of its mechanics, the project team envisions application across a wide variety of projects, as well as development possibilities involving large amounts of data and complex individual files which were previously impossible. eXist's greatest asset is speed and efficiency of retrieval from a Document Object Model, so its development is of primary importance in any research involving XML-based datasets and high performance computing. Spring 2006

Image Mark-up Tool Initiative The image mark-up tool is a simple Windows application that any researcher can use to create valid P5 XML files consisting of annotations associated with specific areas of images. It enables the user to define a rectangular area by dragging the mouse, and then associate textual data (structured compliant with TEI P5) with that area. The user can zoom in and out of the image as required using the mouse wheel or a drop-down control, annotation areas can be added, moved, resized and re-ordered, areas can be added on top of other areas, and the teiHeader information that will go into the file can be edited directly. The application is fully Unicode-enabled, both in the data and interface GUI components. The image data and annotation areas are defined in a block of SVG code inside the XML document (making a multiple-namespace document). The image mark-up tool, developed exclusively using open source code libraries, will be released as an open source application. It has wide-spread use in a variety of Humanities Computing projects involving images of any kind: maps, historical manuscripts, floorplans and more. Locally, potential uses have been identified for the Internet Shakespeare Edition, the Vancouver Island History project, the 17th Century London Map project, and the Maclure House project.

Other significant technical developments associated with this initiative are: state-saving utilities, detection of installed browsers, application structure reflection for translation, and several other useful code blocks.

The initiative has further development possibility since the tool is highly extensible, and based on open source technologies. The project team has discussed the possibility of extending its features to further assist with the rendering and display of the end document. Currently in test release to public.

Copyright © 2012 University of Victoria TAPoR@UVic • Legal info • Contact us