This weekend I attended Science Hack Day at the Guardian Offices, organised by Jeremy Keith (@adactio). I was unsure whether I could originally make it, loitering on the wiki maybe list for a while, and was relieved when I found I could get there. Long story short I was not disappointed and this was an excellent event with the atmosphere of other successful hack events (Mashed 2008, Open Hack London etc) and some awesome people and really interesting hacks.

I liked the idea of mixing a theme with a hacking event, that helped me draw inspiration, previously at other similar events, people would tell you I would rarely finish hacks after procrastination for far too long over what I wish to do.

The Guardian offices make for a good location and around 80+ geeks and scientists (yes some proper scientists) descended. The atmosphere was good, with the usual welcoming geeks and a band of familiar faces. I decided to start hacking a little early and was putting together a small statistics dashboard for the weekend on the way down. I had the idea I may put together some statistics on the weekend gathered through a variety of sources, gathering inspiration from a awesome product, the chirp dashboard I had seen demoed by Remy Sharp. I however had a slight twist in that the data I collected from a number of sources, Google Spreadsheets,Flickr,Github,Pachube and of course Twitter, was to be compared against research statistics from lots of animal species. This could allow the testing of a number of hypotheses, e.g. Is there a significant difference between the amount of sleep that a typical geek has at a hack event and say a Lesser short-tailed shrew (2.6hrs by the way). Basically it was just a bit of fun.

The talks were inspiring Stephen Friend greatly inspired and a group of sponsors explained the merits of their systems and API’s we could play with and there was a video message from Jill Tarter on SETIquest, in th style of some “your challenge if you choose to accept it”. As it is kind of familiar territory to me I was interested to see some of the API’s available from publication and research paper databases. So after more than adequate free lunchtime grazing I was ready to hack.

Rather than continue with the dashboard, in my usual fashion I started procrastinating about the options. My theory is hack days for geeks is equivalent to putting a small child in a sweetshop. So many tempting API’s,inspiration and enthusiam about you just don’t know which way to turn. I had a few other ideas inspired by the talks and got my machine set-up, suitably close to the seasoned hackers in case I needed some help or inspiration and started chatting with Mindy from Elsevier about the possibilities the newly released Scopus API could offer. After further chats with Jud and Michael from Elsevier B.V. this was the start of an hack. So the itch originally to be scratched was trying to help researchers solve the problem of locating citations amongst the many out there and especially try and handle cases where people may move around from institution to institution and there publications drop through the cracks. So demo API key in place I was away.

The API was interesting, I am used to REST interfaces or even SOAP when I have to but this was originally presented in a way I had seen but not really worked with. They exposed a Javascript object which made a number of calls to thier backend systems. It mainly returned HTML but you could overide that. I was mainly interested in working with JSON directly so I started to see if this was possible. Probing the Javascript object showed indeed it was handling JSON so it should be straight forward. IP restrictions incorrectly applied and a few initial hiccups about parameters to pass to the backend system and we were away just getting JSON returned, when the helpful Elsevier guys who had been communicating back with the office had the IP restrictions sorted.

So after all this I was away, but still not quite with an idea, however I did have a problem that I had discussed with some colleagues this API might help solve. So what did I end up building, a tool to help researchers who are writing papers keep alerted about publications made available, by author(s) or title keyword(s) or even affiliation (i.e University) right in the place where they are writing the paper, within a confluence wiki. I thought this may be useful. Then I finished off my dashboard – win win.

When it came down to presentations I presented both hacks, in a somewhat haphazhard fashion and was pleasantly surprised to win the Elesevier prize – a lego Mindstorm robot – thanks very much guys and also get an acknowledgement of the other three judges they thought the hack was useful.

So Science Hack Day on Jun 19,20 2010 in my opinion was excellent. Jeremy did a good job of organising it, he gives us an insight of how he organised it in his latest blog post. Obviously there were a number of key helpers too who deserve equal praise. The location, food and registration were what I had come to expect from attending other hack events at the Guardian and did not dissappoint. The inspiration was good, the depth of knowledge concentrated into the Guardian offices for the weekend was evidently outstanding – just see the list of hacks on the wiki. All in all lots of people did get excited and make things that were really cool. Of course if you want to know more there is a flickr pool, you can track the tweets from the weekend and beyond or one of the awesome side hacks Scilapse (hat tip to Carolina Ödman and her video editing skills as well as the really useful auth cloud hack she put together) lets you experience the whole weekend in a few minutes.

So it rated 5 out of 5 in my book. So much so it inspired me to blog about it. Something I haven’t done for a long while.