By Kristinn Edvardsson,
Being a first year PhD student can be quite confusing, especially when you are hired into a project such as SAF21. While I was taking my first unsure steps in this project I kind of knew what I wanted to do but I still felt something was missing. My aim was to use Geographical Information Systems (GIS) to analyse the fisheries of Iceland, which was all well and good, but I was getting interested in this agent-based modelling everyone was going on about. Luckily for me (or so I thought) one of our partner institutions here in Iceland, MATÍS, hired an ABM expert named Cezara Pastrav who apparently loved ABM. In my naivety I thought it would be a grand idea to do a secondment at MATÍS and try to see what all this ABM was about, and perhaps I might include a small ABM in my work. After all, ABM is on the rise within the discipline of GIS.
I was received well at MATÍS courtesy of Sigríður Sigurðardóttir (who has changed jobs since then) and Jónas Viðarsson. I cannot go on without giving thanks to the cooks at MATÍS, who are probably partially responsible for what is to follow by keeping our energy levels up with wonderful food.
As it turned out I had a wonderful dataset on Iceland and its fisheries that could be used as a basis for an agent based model. Cezara had a thing called Unity (which I thought was Ubuntu´s graphical user interface) and undying confidence and a vision to do something spectacular. I got more confident since I consider myself an Ubuntu expert and suggested “why not model the whole Icelandic fishing industry?” Cezara´s mind, already three steps ahead of me said “yes! and turn it into a computer game while we are at it!”. I agreed, at the time thinking of something basic looking like most agent-based models are, something like an old eighties computer game.
Then it dawned on me that Cezara wanted to implement the ABM in a 3D gaming engine called Unity (not to be confused with Ubuntu except that it also runs on Ubuntu) and set it up more as a computer game. Already confident, I loved the Idea of making an ABM that would be nice to look at and said “yes” enthusiastically. It quickly dawned on me that I was in for a ride, a crash course in model design, C# programing, UML, and database design to name a few. None of these things where unfamiliar to me as I have had a brush with them through my GIS work, but the level of sophistication needed for this project would challenge most GIS experts to the breaking point.
What had initially been intended as a few weeks of learning about ABM ended up being six months of design sessions, programing, presentations and general fun on the job (even the bitter arguments were fun) with no end in sight. During this extended few weeks sessions we managed to create an ABM model of the Icelandic fishing industry that should work in theory, and need a lot more work in practice, but we proved to ourselves, and others, that it is doable.