The humanities student has the question. The computer science student has the method. These two worlds collided at the Helsinki Digital Humanities Hackathon 2016. During the hackathon, students of humanities and computer science worked in close cooperation on various digital materials to find answers to their research questions.
“The hackaton is the crowning event of the Digital Humanities module, but it is open to everyone. In addition to the University of Helsinki, we have people from Aalto University as well,” explains Mikko Tolonen, professor of digital material research.
Four groups are each working on their own set of material. Each group has students from both the humanities and computer science.
“During the week, the groups create a small-scale research project. They are unlikely to finish it, since the processing of the data takes up a great deal of time, but the projects can continue afterwards,” says Tolonen.
At the end of the week, the groups draft posters and present their results.
“We teachers and group leaders are available throughout the week, but the groups are fairly independent.”
The participating teachers from Aalto University were Jukka Suomela and Eetu Mäkelä.
A Who’s Who of YLE programmes
The Finnish Public TV Broadcasting group was tasked with studying the Finnish Broadcasting Company YLE’s digital archive of television output from the period 1957–1990. This was the first time the archive was provided to researchers in such breadth.
“First we had to determine what was in the data. Even YLE couldn’t tell us that exactly,” explains Salla-Maaria Laaksonen, communictions researcher and group leader.
The group had no TV programmes at their disposal, just the digital metadata on the programming collected in YLE’s archive, including the broadcast time, title and description of the programmes.
“We conducted many different kinds of tests to acquire an understanding of what was in the data. We began by categorising the programmes into nine types, such as news, factual programming, sports and culture," Laaksonen explains.
“We also had to cope with the limitations of the data, for example not all programmes were archived. There were sizable gaps as late as the 1970s, because tapes had been reused.”
The refined data can be used to answer many different kinds of questions. For example, the group examined the prominence of President Urho Kekkonen in the material. First they had to filter out all the other Kekkonens; but even after that, they got 1,202 hits for the former Finnish president. The data show which years Kekkonen was most prominent on YLE channels and in which programme types.
The group experimented with a number of big data processing methods developed through research in computer science. These can be used to study many of the issues featured in various programmes, such as the visibility of the Cold War, or Finlandization (the period when Finland was under the sway of the Soviet Union), to see how different genders or political parties were presented, or even how much airtime was given to different sports at different points in time.
“Our data is a kind of national memory bank, a Who's Who of YLE's TV programming,” Laaksonen quips.
A new language of cooperation
“This has been a great week. I was a little intimidated to join in, but cooperation with people from different backgrounds has gone really well. We’ve learned to understand each other,” explains Julia Poutanen, a student of French Philology who studies in the Digital Humanities module.
At the end of the week, the YLE group reported that in addition to new tools, methods and big data processing, they had learned a new language –the language of asking for help. With each other’s help, they can solve the problems posed by a massive dataset, crashing computers and endless error messages.