At least we now have some data from some public sector organisations.
The Centre for Information Leadership, City University London is working on a challenge paper to Government in trying to take forward this initiative. What occurs to me is that this site is only the first stage. What we need are a few more building blocks that need to be put in place before we can make the data really useful. Semantic Web ideas aside (apologies to Sir Tim), what we really need is-
- agree on a technical format (XML, CSV, ODS, EXCEL etc)
- agree on a metadata framework (what should and what should not be included)
- agree on a classification system (ontology, typology etc)
The above for each type of dataset provided.
I have just examined the expenditure Information for two London Councils. While both could be examined by people who know what they are doing with purchase invoices, it would be meaningless for the general public and impossible to process automatically. As stated by Sir Tim and Prof Shadbolt, we need to have standards from a technical, syntactical and semantic point of view. This will allow machine processing and also facilitate comparison.
Can anybody here tell me:-
- do we have Cabinet Office sponsored groups looking at defining the standards across each category of dataset submission?
- if they exist, how can they be contacted;?
- if they exist, what time-scales are they working to for publishing their standards.
I together with a team at the BBC put together the Standard Media Exchange Framework (SMEF) which won the RTS Innovation award in 1999/2000. This was most useful in classifying media content and informed the metadata required for media programmes and indirectly led to the BBCi-Player's classification system. It is a similar standard that is required for each type of dataset.
If anyone is interested in helping to create this, get in touch with me.
Director, Centre for Information Leadership
City University London