EuropeanaTech Hackathon: Interview with Mark Hall, University of Sheffield
We interviewed PATHS team member Mark Hall before and after his visit to the October 2011 Hackathon at the EuropeanaTech conference in Vienna about his thoughts and involvement in the event.
A week before the event:
Hi Mark, thanks for doing this interview. First of all; what do you do, and what is your involvement with the PATHS Project?
I am a post-doctoral researcher at the University of Sheffield working in the area of Information Retrieval and Information Visualisation. My involvement in the PATHS project is primarily on the user-interface side of things, researching and developing user interfaces that support the exploration and visualisation of large digital cultural heritage collections such as those provided by Europeana.
You’ll be attending the upcoming Europeana Hackathon which runs 3-5 October 2011 - can you explain to us a little about what a Hackathon is, and more specifically, what the Europeana Hackathon involves?
A Hackathon is a short event focused on quickly developing prototypes around a certain theme or data-set. In this case Europeana are providing access to their search API (Application Programming Interface) and their semantically enriched data-sets and the participants have two full days to develop prototypes around the themes "Multilinguality" and "Linked Open Data". Examples of the results from previous Europeana Hackathons can be found on the Europeana Libraries site
What will be your involvement in the Hackathon? Will you be working on any prototypes using Europeana data, and if so what are your plans?
In total there are three of us from the PATHS project that are going to the Hackathon. Aitor and Oier from the University in the Basque Country and me from Sheffield. We have not discussed in any detail what we intend to do, but might do something around the idea of linking Wikipedia and Europeana. At the start of the Hackathon there will be an introduction to the Europeana search API and data. Only after that will we be able to fully determine what we will implement.
What are your expectations for the event and what are you hoping to get out of attending the Hackathon?
I am hoping to be able to try out some very experimental ideas in how to visualise and provide access to such a large collection of data as Europeana. Apart from that it will simply be fun to spend some time coding and discussing ideas with my collaborators and also the other Hackathon participants. You never quite know what those discussions might spark off.
A week after the event:
You’ve recently come back from a few days at the Europeana Hackathon. How did the event go, and how was it structured?
The event went very well and was good fun. It started at lunchtime on Monday and we then spent Monday afternoon and all of Tuesday working on our prototypes. Representatives from Europeana were present, so if any technical or content questions arose, there was somebody to ask. Wednesday morning the teams presented their prototypes to the panel of judges and the presentation of the winners in the afternoon wrapped up the event. Plenty of drink, fruit, and sweets were provided to keep us going.
PATHS team members Oier Lopez de Lacalle, Mark Hall and Aitor Soroa spent their time at the Europeana Hackathon building the Wikiana prototype, winning third place for their efforts. You can try Wikiana out at http://paths.sheffield.ac.uk/wikiana/ .
Could you briefly explain the idea behind Wikiana - the prototype built by yourself, Aitor and Oier and how it works?
The basic idea is to provide the user with an overview of what data is available in Europeana, which is quite hard to do with the standard Europeana search interface. The concept we followed was to let the user browse a collection of documents that they are likely to be more familiar with (Wikipedia) and augment the Wikipedia articles with images taken from Europeana. That way as they browse the Wikipedia articles, the users develop an overview over what kind of data is available in Europeana. They can then either click on one of the Europeana images to be taken to that specific item in Europeana, or now that they have an overview, use Europeana's search facility to find specific items they are interested in.
Technically it works by loading the Wikipedia page that the user wants to see and for each paragraph determining the salient keywords. These keywords are then used in Europeana's opensearch API to find images that fit to the paragraph and the Wikipedia page is then displayed with the images.
Were there any other prototypes that came out of the Hackathon that you found particularly interesting or creative?
Could you see the technologies in Wikiana or any of the other prototypes being developed further and used in future products and projects?
A difficult question. The methods applied in the Wikiana prototype could be developed further, however our goal was primarily to try out an experimental approach and develop an understanding of what is possible with the Europeana opensearch API. This goal was achieved, but at the cost of further development requiring a major re-write of the prototype. Of the other prototypes the recommendation prototype has definite potential for inclusion into Europeana, although I do not know how well it would scale.
Is there anything else you got out of the event? If there are future Hackathons planned, are you hoping to attend them?
It was nice to be able to spend some time trying things out that were not core to the PATHS project, but where the ideas might influence the future direction of research. At the moment I know of no further Hackathons. When they appear we shall see what to do.
Many thanks for your time, Mark.