Culture Hack Day debrief event - the hacks

 by Martin Belam, 28 February 2011

Last Monday I attended a Culture Hack Day debrief session at the Royal Opera House in London’s Covent Garden. The evening featured some exposition of why the institutions involved had taken part in the event, and some presentations from “hackers” who had attended the event. They were not just showing off their hacks, but also talking about what people could do to make their datasets more easily available to be worked with.

Becky Stewart presented her hack of being able to jump between comments, using any word as a hyperlink, to another occurrence of the word in a different user generated comment. Her starting point had been an Excel spreadsheet of public contributions to the Crafts Council website, and you can read more about it on her blog.

She also introduced the audience to Tim Berners-Lee’s 5 star system for measuring data publication. “To go superstar is to make it linked data” she said.

Melinda Seckington showed her graphs made from UK box office data about films, but also confessed that she was having to scrub the data clean by hand. In the course of her hack, she’d had to cope with 9 different spelling variations on “Harry Potter and the Deathly Hallows - Part One” from the same set of data. Her plight pointed out the glaring lack of unique URIs for a lot of cultural entities.

Brendan Quinn spoke about being both a “hacker” and a representative of the BBC on the day. He used processing.js to do some data visualisation because, he said, “It was something I hadn’t used before, so I thought I’d just play around with it”. It is a common theme with developers, that a hackday isn’t just a chance to make new things, but also play with new toys. My colleague Robbie Clutton just published a post on our Guardian Developer blog about what he’d learned from trying out new tech during the recent Guardian Hacks SXSW event. Brendan also referenced the Guardian hackday, by directing the audience to Rewired State's seven simple rules for opening data:

  1. No PDFs
  2. CSVs not Excel
  3. Consistent formatting of documents
  4. Know what data you’ve got / can provide access to
  5. Appoint a decision maker and contact point
  6. Stick to standards
  7. Listen to feedback

There were a lot of ex-colleagues at the event - I worked with Brendan at the BBC, with Matthew Somerville on TheyWorkForYou, and with Chris Thorpe on the Guardian’s Open Platform. Chris was showing off a hack based around the paintings of Alfred Sisley, showing where they could be viewed. His premise is that a lot of great art and cultural artifacts are virtually hidden in small collections, and that there is no easy way of saying “I’m in a gallery and I like this painting - where is good to visit with more of this type of painting”. As he said, museums and galleries are not very well joined together. He is working on a start-up called Artfinder to try and solve that problem. His hack had been for the iPad because he sees them as great enabling devices - the app model and the lack of a visible filesystem gives you the power of a computer, without having to have any understanding of how computers work.


As well as featuring some lightning talks about the hacks themselves, the Culture Hack Debrief event featured an interesting overview of what the day had meant for cultural institutions from yet another ex-colleague, Rachel Coldicutt. Tomorrow I’ll have my notes on what she said, and look at why freeing our cultural data is so important to the future of the arts.

Keep up to date on my new blog