Digitisation in the UK and beyond
All very obvious stuff when you think about it, but getting the details right make the difference between a good digitisation project and a great one.
Some basic advice that many projects starting up could still do with learning
A reminder that there are still places at the Digital Preservation Coalition, British Library workshop on JPEG 2000.
The workshop takes place on 25th June at the British Library, London.
More details on the conference are available from the DPC website.
As reported elsewhere, version 4 of the the metadata standard for digital images, VRA (Visual Resources Association) has been released.
Everyone knows the audio resources are enlightening, educational and entertaining, but it can be difficult to know how to integrate content such as that provided by the British Library’s Archival Sound Recordings.
The EASAIER Project: Enabling Access to Sound Archives through Integration, Enrichment and Retrieval is building tools that solve these problems. Project member Ceila Duffy spoke of the need to create tools that allows audio on the web to go beyond playback.
Some of the tools created by the project include web clients that allow audio files to be speed up or slowed down (useful for music practice), to locate emotions or accents without spoken word files (useful for resource discovery), to separate out different speakers or instruments or sounds, to represent music in visual form, or to do searching based not on keyword but on actual audio content.
This is all fantastic stuff and makes audio material so much more attractive. The next challenge will be to embed such tools into web interfaces that give end-users easy ways to tag, analyse and manipulate the content of audio archives.
An article on shareable metadata in First Monday points out the fact that if harvesting protocols like OAI are going to work, they need metadata which works when placed outside its local context.
The article suggests various stratgies for metadata providers to ensure their data makes sense when mixed with others’ metadata - consistency, content and coherence are three of the six key issues. Digital resource creators would do well to heed them.
But is this enough? With more and more metadata being created is it possible to rely on each and every metadata provider creating quality, consistent metadata at a local and national level?
Or do we need more intelligent, machine-driven ways of interpreting and mapping metadata in order to aggregate it and present it to the end user in a meaningful way?
