This weekend my colleague Gavin and I decided it would be useful (and fun!) to head along to Culture Hack Scotland, a 24 hour hackday organised by the Edinburgh Festivals Innovation Lab and themed around both the festivals and the wider Scottish cultural scene.
The Edinburgh Festivals Innovation Lab is a new(ish) initiative which has emerged from Edinburgh Festivals, the organisation that is jointly funded by all 12 of the official Edinburgh Festivals to enable them to work together throughout the year, promote initiatives and festival content etc. The idea for the Innovation Lab apparently emerged out of discussions with all of the festivals about their use or interest in digital technology: there were lots of ideas and potential for projects but they didn’t necessarily have the time or skills to take these forward. Last year the Lab hired their inaugal geek-in-residence Ben Werdmuller (he of Elgg fame) and the Culture Hack Day was a significant outcome of the work he has been doing over the last few months.
The idea for Culture Hack Scotland was quite simple: gather cultural organisations and their data, add geeks and see what can be imagined and coded in 24 hours. As with other hack days a venue was provided, with wifi, power sockets, food, and all the other important resources one needs when hacking, coding, designing, creating. Announcements and invites triggered a rush of sign ups for free tickets (via Eventbrite).
An impressive and highly varied group of around 100 people from cultural organisations, tech companies, and all manner of freelance and student hackers and designers signed up but an even more impressive series of data sets were also assembled for the event. Some of the data was extremely specific to the festivals (e.g. 2010 programme data for the Fringe, the current International Festival programme); some has loose points of connection (Edinburgh Year-Round Footfall data, Skinny reviews, Guardian Content API, etc.); whilst others were complementary cultural sector data (National Museums, Galleries and Libraries’ of Scotland’s collections data, Culture Grid, National Theatre of Scotland audience and library data, BBC Radio Times Data). Formats varied a lot – Culture Hack Scotland had persuaded a number of the data owners to make their data available for the very first time for the event and many had provided CSV or similar files. However a full Summer Festivals API will be launched in time for August 2011 and it is hoped that some of the data providers will be making available API access to their data on ongoing rather than one-off basis. Indeed if there was a theme of the event it was that data can be useful and, as long as they can see this, the data owners will see value in sharing their data.
Of course other existing data sets could also be used as part of the event and I recommended both EDINA’s Unlock Text API and the AddressingHistory API as potentially useful. I believe Unlock was used by some hackers although sadly no-one decided to build my suggestion of a map of Fringe shows based upon locations mentioned in their description/in the plot of the show. Many of the hackers in attendance also brought in tools and data they already knew about – Google Maps APIs were widely used, some built hacks based on combining listings data with mobile phone software that allows you to speak to an automated menu system, whilst another used Festival data as a source for an impressive Arduino device he hacked over the weekend from an Arduino controller, LEDs, 6 ping pong balls and an egg box.
The programme for the event was well thought through as it allowed those with programming skills plenty of space to code and work on ideas but also ensured that those without technical backgrounds felt engaged and included in the event and were able to add their own ideas into the mix. In fact the “Culture Hack Express” session, a workshop for non technical hacking, generated fruitful discussions, ideas and ultimately a single well thought through concept (for a form of cultural passport to track events attended) that could be created digitally by others.
The event opened at 5.30pm on Friday where I joined a long excited queue (no, I’ve not seen one of those at a hack event before either) forming outside of InSpace. We were signed in on iPads (running the EventBrite app which seemed very effective for checking in) then got to participate in our first creative exercise: the labelling of our badges. Various coloured dots denoted role and type of organisation. I went for orangey red (developer) and blue (other).
As we started to meet and greet each other Rohan Gunatillake and Ben Werdmuller introduced us to the background, aims and sponsors of the event. They had, ahead of the event, sent various cultural organisations postcards asking them to speak about “the potential of open data in the arts”. Each recipient had recorded a short homemade video for the event which had been appearing on the Culture Hack Scotland website in the run up for the event – a great way to build excitement and help potential attendees understand what and why they were participating in the day. During the event itself we got to see two great wee videos from Hugh Wallace of the National Museums of Scotland and Margaret Robertson of Hide and Seek (developers of the fantastic Tate Trumps):
We were also treated to a talk from Rachel Coldicutt of the Royal Opera House – a veteran of the Culture Hack Day that took place in London earlier this year. To close Rachel asked us to think of names, ages and cultural events you might attend, this triggered some improvisational work on user needs and profiles. Though we were all a little too crowded and warm to be the best engaged audience this was a much needed reminder that any projects we would be building should start with the audience and what their needs are (and not just what could be cool to code).
As the night was properly kicked off we started sharing ideas – some had already started thinking about ideas and playing with the data ahead of the event (such as FestaFriend and Fake Fringe) but lots of us were also thinking of new ideas, many inspired by both the data and the talk from John Maxwell Hobbs, Head of Technology for BBC Scotland. Two full windows of InSpace quickly filled up with Post-It notes as we thought of quirky ways to exploit the data that had been made available…
As ideas were gathered dinner was served for developers (this was a feature of the event – hackers got free grub, non-hackers did not) and computers were grabbed from bags. Having not had a chance to look at the data before the event I decided my best bet was to leave my ideas on the window and return for a full day of creative hacking on Saturday. By the time I left at 9.30(pm) three long tables were hacking away on either their own or others’ ideas, teams were forming, existing projects were finding new collaborators, and people were generally sharing ideas and problems and doing wonderful stuff…
I returned at 9.30(am) on Saturday to find that even more hackers had assembled. There were perhaps 40 laptops (almost all Macs!) and longer rows of tables and power sockets had been provided in our Saturday venue, the Informatics Forum. Although most had gone home to sleep for at least an hour or two several teams had coded throughout the night to get to their idea realised.
As many others were mid way through serious coding work I decided to have a look at new ways to explore performance data – one of the largest and most interesting data sets being made available for the first time. This is data on every performer and show on the Edinburgh Festival Fringe and is rich with context since it includes URLs, awards won, nationality of production, and venue and timings of shows. Initially I wanted to think about ways to present this as a sort of IMDB for performance as there are various theatrical databases but nothing comparable or as useful as this for the performing arts outside of film.
I started to explore the data and although it was rich – this is, after all, the data used to publish the fringe guide and to track productions internally by the Fringe Office – there were also some immediate issues.
- Company information provided for the programme does not give you a useful idea of who is performing in that company at that point in time and, with descriptive text under a strict word limit, you do not usually have a definitive list of identities for event key talent for most Fringe shows
- The inclusion of international performers and both amateur and professional performers can mean that names of individuals are not unique (in the UK Equity tends to ensure that professional theatre performers register a unique professional name) and are not easily connected to identities elsewhere. Further the copyright around photography of performance can be tricky so pictures of productions are frequently limited to one or two cast members rather than the full company making visual disambiguation difficult as well.
- Information on production, design and tech staff is often non existent or restricted to short run paper pamphlets.
I was still keen to see how information from the Fringe perfomance data set could be combined with other information – Wikipedia or IMDB listings both being an obvious first step to finding out more about a company or individual. However this did seem far too ambitious an exercise for the time available.
As my colleague Gavin arrived we started to chat about more creative ways to explore the performance and venue data and he suggested we work with Inform7, a clever and intuitive natural language system for creating interactive text adventures. Our nearest tame coder, Joe Halliwell of SoDash, added an enthusiastic thumbs up for the idea and so we began to see what we could do with the data.
The first step for me was to look at how to use Inform7. The idea of this software is that you type instructions in natural language and the system uses this to create a text world that can be explored and experienced by a reader/player. Although it classifies itself as a tool for interactive fiction many of the works created have a game sensibility as works can be the text equivalent of a rich virtual world. Although the themes and genres are quite different the software reminded me of the GTA games since characters, objects, rooms can all be made interactive and richly described though this is something charming about this being in text form creating a very modern and very retro exploration experience.
Having looked at some works created with the software it was time to download and try to build something myself. I began with something simple building a room, then an object and seeing how it worked. I quickly realised both the best and worst thing about the software was the fact that one writes in natural language but that is, to a former data munger like me, a slightly unnatural process compared with designing in pseudocode or building scripts based upon clear rules and filtering. A quick search of the manual made a huge difference though and soon I had a tiny text that allowed you to be in a location and look at the area in more detail.
Next we needed to establish how the text/game would work. What was the story? How would points be scored? How would could the data from the Fringe be used in a way that kept the character of Fringe but was also a compelling stand alone experience rather than just another listings tool? Having decided to base the game around a challenge to have the most adventurous Fringe day possible (full decription here) I decided that points would be scored based upon the novelty and/or riskiness of seeing a particular show (full scoring guide here). The player would look around the city helped or hindered by people and items encountered along the way with obstacles including overly aggressive people handing our flyers for shows and opportunities such as recommendations for shows, an overheard comment, etc.
Having the narrative in hand the work split was clear: I would build up some characters to encounter and some basic opportunities for exploring to demonstrate how the game might work, Gavin would look at how to use the data on the venues to create a suitably formatted map of the venues. Richer experiences based upon show listings, the names and locations of real performers, and chance encounters were all aspects that would need to be added at a later date.
Each venue in the csv provided by the Fringe was provided with it’s Fringe Venue number, a description, a list of sub venues, and a location (in decimal latitude and longitude) and, after a little Googling, and a helpful diversion into the Haversine formula, some perl libraries were found that enabled a quick conversion of lat long info into the natural language mapping expected by Inform7 – in which every location is notionally a “room” which is connected to others via movement along the 8 compass directions (N, NE, E, SE, S, SW, W, NW). The venue number (assigned by the Fringe for use on maps etc.) was used as an identifier and each venue and it’s description were coded with these Inform7 friendly geo references. Less straightforwardly the rooms now had to be connected to each other – in this software one usually needs a statement such as:
“X” is a room. “X” is north of “Y”.
Thus some further processing was required to connect up the venues so that they could be loaded into Inform7 as an immersive series of connected venues. This proved to be challenging but eventually a mixture of Inform7 descriptive statements and a little perl wizardry accomplished this just in time for the showcase of ideas…
Show & Tell
At 4.30(ish), and after some controversial sound tests that involved hearing Europe’s The Final Countdown several times over at the most stressful point in the afternoon, both the room of hackers (many still hard at work on final fixes and tweaks) and the room of cultural sector folk who had been taking part in the Culture Hack Express! session gathered together for the Showcase of projects.
In readiness for the projects created Tom Scott, a seasoned hacker described variously as a “celebrity geek” and a “geek stand up comedian” (neither of which he thinks is true), gave us a run down of what can be created at hack days – favourite hacks, ideas and projects from previous days he’d attended. Some were playful, some silly but most were useful and some had become important and widely used (such as a government energy use website that had become a source of competition for staff trying to reduce their consumption faster than their peers!). It was a fun talk and the perfect warm up for those who had not attended a hack day before to get a sense of what hack days actually involve.
Next up were the presentations of the 30 projects created over the previous 24 hours. Yes, thirty. These did not include the projects that had not been completed or were not yet presentable – Rohan estimated the coding time and effort that had taken place as having a value of £20k. Given the calibre of hackers (many were notable and experienced members of Edinburgh’s tech community) that certainly seemed fair.
Projects ranged from games and visualisations to fully functional mobile phone applications. The full list is currently being assembled as a web showcase with many already available to view and explore on the Culture Hack Scotland Showcase area and I would recommend browsing through them. Some of my favourites included:
- Bookfest: a hugely slick book festival phone app to enable you to recognise favourite authors and find further reading as well as providing an easy way to explore listings.
- An app for finding shows to fill the gap between two other Fringe shows – including a guestimate of travel times that allows you to ensure you will actually be able to fit everything in.
- Steal It: an app to explore cultural works by pretending you are a master criminal and finding items around town to “steal” as part of a location based game that uses FourSquare-like checking in ideas.
- Edvent: a text based game, this one modelled firmly on retro gaming, that allows you to explore the Fringe as an adventurer with 100 gold coins to spend!
- Edinburgh Footfall: a visualisation and audio experience based upon footfall data for the city of Edinburgh for a whole calendar year. This led to some amusing tweets but it was a fascinating attempt to turn the data into art:
Our own presentation was based around a hastily created website explaining our game, now christened Edinburgh Culture Vulture, and a live demo of the game itself which looks a bit like this:
In addition to the stylish Culture Hack Scotland mugs that all hackers received the organisers also had prizes to award. Although I had to leave before the announcements I was delighted to hear that some of the most creative apps had been rewarded but these were not all the most technically complex.
Reflections on Culture Hack Scotland
There were some really lovely ideas and, coming from an organisation with a strong geo emphasis, it was super to see the way in which location was used playfully and skillfully in many of the projects. Many of the hackers choose to answer very specific questions but there was also a lot of creative, abstract and artistic work going on.
Seeing what others were thinking about and/or developing really helped inspire new ideas and encourage cross-pollination and it was rewarding to see that ideas on Friday’s post-its had fed into the ideas or functionality of some of the final hacks in one way or another. It was interesting to see how the format for the day worked as I have attended various hack days and bar camps before with some prooving hugely fruitful and others frustratingly lacklustre. In this case the mixture of pure enthusiasm from the organisers, the quality of the facilities, and the fact that unique data had been sought and requested specially for the event, was really successful.
Like others I left the day inspired and enthused to continue thinking about not only the data provided for this event but also to try to find new creative perspectives to exploring data we have and use here around EDINA projects and services. There is something refreshing about seeing entirely frivolous visualisations of data as an appealing alternative view to the more functional academic styles of visualising data.
There are hopes to follow up Culture Hack Scotland with another event next year and I hope that does go ahead as it was a really creative and successful hack day format and atmosphere.
The Twitter hashtag for the event was #chs11 and I would highly recommend looking through some of those tweets, as well as the Culture Hack Scotland blog, as alternative / more detailed reports of the event.