Open Data Camp 6 in Aberdeen opened with the ever popular (totally essential and very informative) ‘open data for newbies’ session.
And it started with a question: “Who knows what open data is?”
Participants decided the critical answers are that: it is data that can be made public (so not personal data); that is available free (or near free) for other people to use; and that is properly licensed as such.
Also, that it should be easily and available and consumable; although there is considerable debate about what that means.
The five stars of open data
Tim Berners Lee, the founder of the world wide web, has a five star scale for open data. Session leader Rebecca Stickland explained it as follows: 1: ‘It is on the internet – it might not be machine readable, but it is there’; 2. It is machine readable but in proprietary format; 3. It is machine readable in non-proprietary format (such as comma separated values or csv); 4. It has a link embedded to identify the record; 5. It is data in the web (linked data).
Any open data may be useful; but five tends to be the really exciting stuff. “[Linked data] uses a web architecture to link data together in a really cool way,” Rebecca said. “For example, if you go onto the BBC iPlayer, then you can see what song is playing, and the lyrics, and information about the singer – the BBC only holds one bit of this data; the rest is pulled in from other sources using linked data”.
Publish – you won’t be damned
The session moved on to discussing what data should be published as open data. Government departments, councils, and many other organisations can worry about what people will find useful and how to publish it; but participants advised the best thing to do is to “stick it on the internet”.
One participant from a government department said it had found that “if you publish it people will come to the good stuff”. Organisations can then maintain or improve the data-sets that prove popular, and just leave the data-sets that nobody is interested in.
Some companies are also interested in publishing their data as open data, but the session acknowledged this is harder, because they are committing to making an asset available for free, in what can be a very complex licensing environment.
Scream if you want to go faster
One participant asked if there was any training available in open data. xxx said there is a lot. “Start on Twitter,” she suggested, “the open data community is one of the friendliest you will find, and loads of people will help you.”
Then, there is the Open Data Institute, and its nodes, and events like Open Data Camp. “This is a great source of information,” she said, “and there are no stupid questions. Somebody uses a phrase or a three letter acronym that you don’t understand; challenge them.”