July 2017, every Monday to Thursday forty students from all over Belgium come to Brussels. I am one of them. We wake up in the morning looking forward to #oSoc17, have breakfast and commute to Brussels. During the day we hack, drink coffee, have stand-ups, pitch our project and banter. By the evening we are ready to commute back home with the warm, fuzzy feeling that we’re making a difference.
As a high school student in Ghent I commuted every day by tram from home to school and back again. It took me a few years to gain enough knowledge to know beforehand which trams would be crowded and where I would have to board to ensure that I would still have a seat. Now, years later, I find myself commuting to Brussels by train not knowing a thing. How could I find out which train is most suitable to my needs? I have somewhat flexible hours and like to read on a train so delays in the morning aren’t that bad (more reading, I’ll make up for the delay in the evening) whereas a crowded train would be horrible (I can’t concentrate on reading while standing up).
Team oasis to the rescue! What if you could know beforehand which trains are more prone to delays, crowdedness or other inconveniences? Based on your own personal preferences you could easily pick whichever train suits your needs the most. Guess what: that’s exactly what we’re doing right now.
How can we make this dream come true? Open Data. The more data is published by your public transport providers, the more aspects (delays, fare prices, frequency, …) about commuting we can take into account. Using this data we compare their services to your expectations and predict your quality of experience so you can pick the right train, tram or bus.
But there’s even more: Linked Open Data. Whenever we talk about trains and delays everyone knows what we’re talking about. Computers on the other hand are dumb and have no notion of these concepts. So we need to define a vocabulary or ontology about public transport for them to understand these concepts (somewhat similar to a dictionary for us to look up what a word means). These vocabularies can be made interoperable by linking a term in one vocabulary to a term in another vocabulary. For example all over Europe there are many public transport providers with their own vision on publishing open data. If you wanted to work with their data you’d have to take into account all the differences. Whereas if you’re using Linked Open Data the many different datasets can be used together because we can translate the terms from the vocabularies.