At Global Campus, each of us gets one development day a week to spend on an external project with a partner organisation. Martin spends his at the ICAI OpenGov Lab at the University of Amsterdam.
Over the past four months Martin has been working with the team behind WooZM in order to extend the project and help make government data open. At the ICAI OpenGov Lab there has been an ongoing collaboration with the current team of WooZM - David Graus and Maik Larooij.

WooZM (previously called Woogle) is a search engine for Dutch government documents. Under the Open Government Act (Woo) many government documents have to be made public. WooZM then takes information from all categories and administrative bodies and indexes it, making it easy to find and reuse. The project was initially carried out by scientists at the University of Amsterdam with the search system designed by Maarten Marx and Jaap Kamps, and developed by Maik Larooij and Maarten Marx. The two deeper goals are helping (especially smaller) governing bodies to make reuse of released documents possible and easy and improving the digital sustainability of government information. With the current team we have added a third goal - make the project as a whole more easily accessible and explainable. More on that in the next paragraph.

The changes we have made can be split into two categories. The first is technical: we have been trying to modernize Woogle's stack to use emerging software tools and practices. Thanks to David Graus, who is the lab manager at the ICAI OpenGov Lab, the system is fully dockerized and equipped with an automated deployment pipeline. On our side, we built a new streaming pipeline for processing. It is our belief that this modernization will help with the second and more important kind of change - making the project easier for students and people interested in Open Government data to pick up and use. That would enable people to add data more easily to existing datasets, or to contribute by adding logic to the project and testing its performance. Because each processing step is independent and the storage layer is the source of truth, a contributor can work on a single component without needing to understand or run the rest of the system. Contributing to an already rich project would let people focus on the areas they are most interested in. Another perk is that each user would have their own single source of truth, so contributors would be fully independent of each other, or if collaboration is needed, all you would have to do, is share your datasets.