ODI Leeds

#AirHack Day 1

The hackathon began as all great endeavours do - in the aftermath of a storm. Doris had certainly made her presence felt the day before and travel was still being disrupted on the first day of #AirHack. The ODI Leeds team - Paul, Kathryn, Amy, Stuart, Neil, Giles, and (eventually) Tom - were ready regardless. Kettle on - check. Biscuits out - check. Some sunshine had even made an appearance just in time for Paul to make introductions and kick-off proceedings.

James Tate, from the University of Leeds, was first on the 'rug of truth'. James has been researching air quality and emissions in Leeds, and had some open tools to share. He first extolled the value of data repositories like Data Mill North (where a lot of Leeds air quality data can be found). He then demonstrated OpenAir - a way to process and visualise air quality data built on the language R. Broken down by month and displayed like an annual calendar, each day was coloured according to a scale. The darkest areas were the worst days for air quality, and were notably more numerous in the winter when weather has an adverse affect on levels of pollution.

He also demonstrated a polar plot showing concentration of air pollution (as measured) combined with wind speed and direction. Overlay an aerial photo of the measuring station and you can start spotting things like the affect of buildings on air pollution. Do areas with taller buildings have more problems with pollution? James then asked his colleague Jim McQuaid to walk through his journey that morning. From the University of Leeds campus on Woodhouse Lane to our front door at Munro House, Jim walked with a black carbon sensor. Black carbon (or soot) is a very fine particulate and often gets breathed in, sometimes with more toxic particles attached to it.

Jim did his best to match the route to where the spikes in particulates were recorded, and it makes for interesting analysis. There was a sudden spike just before Jim arrived at Munro House and he suspected that was the congregation of smokers outside the building's entrance. Ultimately, the most polluted place to be when the overall air pollution is high is actually sat in your car.

Elizabeth Bennet, on the UK government's Civil Service Fast Stream graduate programme, then spoke briefly about Defra, who kindly supported the #AirHack event and are looking to see more innovation done with their datasets.

After the introductions, it was time to cover the #AirHack challenges and how to use the communal HackPad. Everyone was encouraged to put their notes, code, ideas, musing, etc, in to the HackPad, allowing others to contribute and comment. It also acts as a 'recipe book' for the projects developed over the two days, allowing people outside of the event to recreate and improve in their own time.

The best ideas come from groups of people who know nothing about each other. So Paul instigated the infamous 'chair swap' - if you were sat next to or near a colleague, you had to move. Then the pens and post-its and reams of paper came out. Using the #AirHack challenges as inspiration, the different groups were tasked with coming up with as many potentially feasible ideas as they could and sticking them to the wall.

Once everyone was done, it was time to group similar ideas together. The resulting 'teams' were:

  • iWalk - how do you change user behaviour?
  • Leeds Air - a Leeds focused air quality display system, either physically around the city or as an app
  • Trumped (eventually renamed Shair) - releasing and sharing data
  • Senseair - how good are air quality sensors really?
  • Thunderclap - is there a link between noise pollution and air pollution?
  • Taxiderm - matching cars to emissions infor to see who chucks out more pollution

From here, people were free to work on the idea that interested them most. After lunch of course. Who can innovate on an empty stomach?

Throughout the afternoon, the ODI Leeds team were on hand to help guide the various groups - offering advice and some disruption to challenge the thought process. This is often the quietest yet busiest time of any hackathon event, where people are wrestling with concepts or code to start building their ideas. We stress that nothing needs to be 'finished' at the end. A polished sketch or series of flipcharts is equally as good as an app prototype. The potential for change is often found in the making of a thing, not the finished thing itself.

The end of the first day of any hackathon is a chance to catch up with the groups and see what they're missing or what they need to do next. Not everyone attending on the Friday could make it for the Saturday as well, so it was important for them to pass the torch on to their brave colleagues who were going to work through Saturday too. Amongst the round up included a potentially project-killing bug in a car-counting app; plans to create a wind-shield for a mobile phone to better record sound; and a brilliant journey planner app on paper but would it be too big to build?