A.nnotate at the Edinburgh Repository Fringe
Just back from Edinburgh Repository Fringe - a gathering of people interested in making research data and publications more useful and easy to access. There were great keynote talks from Dorothea Salo ['the institutional repository is dead'] and David De Roure ['how repositories can avoid failing like the grid'], and lots of "performances" in the spirit of an unconference / BarCamp event.
A common theme seemed to be that setting up a university-wide database and expecting researchers to populate it by filling out forms just doesn't work; but providing easy to use web services for collaboration and improving visibility / citation counts of research is the way to go.
We've been thinking along these lines too - our A.nnotate.com service can be used to enhance existing document repositories with private / shared annotation features (using the A.nnotate APIs). As well as A.nnotate, we also run the PublicationsList.org service which makes it easy for researchers to maintain their list of publications on the web - and can be used as a front end service to institutional archives to make repository deposit as automated as possible.
Some of the interesting ideas from the meeting (see the repofringe08 wiki for the full programme):
- Ways for sharing/presenting research data - including Stuart Macdonald's DISC-UK DataShare project with interactive widgets, Simon Coles experiments with trackbacks / pings / ratings / comments for research data in chemistry, David De Roure's MyExperiment facebook-like social networking site for bioinformatics data which looks like it has great potential beyond its current emphasis on sharing glorified makefiles graphical workflows
- The dangers of "Bit Rot" [talk by Richard Wright, BBC archives] - should you archive the uncompressed version of files (e.g. TIFF rather than JPG?) given that all digital formats (tape / disk) degrade over time, and a few bits lost from a compressed file can render it useless, but uncompressed formats degrade much more gracefully. My immediate instinct would be to compress and store multiple copies for redundancy and error checking but for the longer term (10s / 100s of years) the uncompressed version is preferable.
Spotted in Edinburgh on the way to the meeting:-









