HomeNews and blogs hub

From zero to a responsible software developer in a week

Bookmark this page Bookmarked

From zero to a responsible software developer in a week

Author(s)
University of Oslo by Alexander Ottesen (under CC-BY).

From zero to a responsible software developer in a week

Posted on 14 April 2016

Estimated read time: 5 min
Sections in this article
Share on blog/article:
Twitter LinkedIn

From zero to a responsible software developer in a week

Posted by r.silva on 14 April 2016 - 10:18am

University of Oslo by Alexander Ottesen (under CC-BY).By Iza Romanowska, Institute Fellow and PhD Student at University of Southampton.

Learning a new computational technique, be it simulation, specific type of data analysis or even lab-based methods, can be a daunting task. You could start by reading up on all the previous applications and methodological papers but it can leave you frustrated with the technical nitty-gritty which is virtually impenetrable without a good knowledge of the tools that were used. So perhaps, it is better to start from the other end and learn how to use the software first? Sounds like a reasonable plan until we are reminded of the legions of early career researchers trawling through literature looking for a nail they could hit with their shiny new hammer.

This year, thanks to the support of the Software Sustainability Institute we embarked on a project to break this circle and to train a group of students rarely associated with cutting edge, high tech academic software in a novel but rather complex simulation technique – Agent-based Modelling (ABM). On the rainy Monday morning of March 28th, 35 archaeologists arrived still sipping their coffee at the historical museum in Oslo to learn how to build simulations. Most of them had never seen a line of code before. By the end of the week 87% of them declared that they are confident enough to use simulation in their research.

Agent-based modelling has taken by storm disciplines from all corners of the scientific spectrum, from ecology and health planning to transport and various branches of social sciences. Surprisingly, despite its broad appeal to researchers who deal with social systems and humans in general, humanities seem to be late to the party of using computational modeling to investigate the complexities of social interactions. Using the medium of the largest conference on digital methods in archaeology, the Computing Applications and Quantitative Methods in Archaeology (CAA), we took on the challenge of breaking this impasse by providing a comprehensive set of events to create a cohort of well-informed and well-trained leaders in simulation software whose influence, we hope, will ripple throughout the archaeological community in the years to come.

It all started on that rainy March Monday morning when we began a two days workshop dedicated to one of the most commonly used ABM simulation platforms – NetLogo. NetLogo is an open-source platform for building agent-based models, which combines user-friendly interface, simple coding syntax and a vast library of model examples, making it an ideal starting point for entry-level agent-based modellers, as well as a useful prototyping tool for more experienced programmers. Two days and several hours of swearing at the screens later we had three dozen specialists in roman coins, bronze age pottery and Maya burials who could develop, test, document and disseminate a working simulation. The enthusiasm was sizzling in the air. It was the right time to squash the all too rosy picture of simulation as an infallible tool for doing ‘real science’ and prepare them for the difficult road ahead.

We held a roundtable on the first day of the conference consisting of eleven 5 minutes long lightning talks on the limitations and pitfalls of the method, each followed by a 10 minutes discussion slot. Experts in agent-based modeling delivered the talks, answered questions and engaged in, what in many cases became a communal moan about the quality of the data we have to work on, the difficulty of testing the code and problems with getting credit for research software. Finally, to finish on a more positive note, on the last day of the conference we held a full day paper session during which researchers from Europe, USA, South America and Oceania presented simulations they developed to assist with their research.

For data poor disciplines, among which archaeology is definitely one, simulation is often the only technique that can bridge the gap between theoretical divagations and the sparse empirical record. Unfortunately, the lack of formal training is holding the discipline back. The series of events (the workshop, the roundtable and the paper session) we ran in Oslo was addressing this skill gap, but we wanted to reach further. Thanks to the support of the Institute we are now able to increase the impact of our work by turning the handouts, recordings of talks and short interviews with the experts that gathered in Oslo into a self-study online course (follow us at simulatingcomplexity.wordpress.com for updates regarding the launch of the course). This permanent open access resource will serve the community at large providing a one-stop website for learning about agent-based modeling and good practice in research software without the need to brave through Norwegian weather.

Share on blog/article:
Twitter LinkedIn