Falvey’s Digital Library has just been upgraded with new backend software that will improve its ability to continue growing and improving the online collection. The Digital Library’s first incarnation was launched in August 2006. Over the course of 4 years, the DL’s collection grew to over 9,000 items, and a substantial software functionality wish-list.
- Add support for more file formats, so our collection can include a broader range of materials
- Incorporate an OCR process to facilitate full-text searching of collection content.
- Add support for inclusion of transcriptions with hand-written materials
Our initial software used a variety of technologies to achieve its goal of storing information about digital documents. Unfortunately, not all of these tools worked well together. While the new version of the software retains the METS metadata format and eXist-db XML database, it replaces nearly all of the other components with a suite of more closely-related technologies. The new, all-XML, all-Open-Source framework consists of the following components:
- METS XML schema – Library of Congress standard for describing digital objects.
- eXist-db – Native XML Database
- Orbeon Forms – Java-based XForms engine
- Tesseract – OCR Engine
- VuFind – Online Public Access Catalog
New Key Features:
Root level Document Attachment
Catalogers now have the ability to add document-level items to each object. The most relevant use of this feature is to attach a hand-transcribed, fully annotated companion document to a digitally scanned book. More information on this feature can be found here and a live example can be found by viewing the Lane Manuscript
AJAX-based metadata editor
Document layout and file attachment configurations
The system incorporates a batch-attach routine for adding multiple files (in our case the pages of a scanned book) to a digital object as a single operation. An interface is available to customize the arrangement and location of these files, as well as adding and deleting files when appropriate.
OAI/PMH is a standard for serving and harvesting metadata. The Digital Library is now fully harvestable using this standard.
In the coming months we will extend the software to include custom drivers for a VuFind front-end and modularize the metadata editor to support a wide-range of options including Dublin Core, MODS, EAD, and PREMIS support for preservation Metadata.
Our plan is to launch the software as a simple, open-source platform for preservation and presentation of digital collections. So stay tuned! We are targeting April 2011 for the Beta Release.
We are always looking for development partners! If you are interested, please contact us at firstname.lastname@example.org