Ireland’s Open Data Strategy

by dixie

Writing about Atlanta is proving tricky, and in the meantime life is galloping away with me. Just so you don’t forget about me, I’ll offer another non-travel blog to tide everyone over.

Two weeks ago I attended the public briefing at the Department for Justice and Equality on the state of Open Data in the State. The minister for Public Expenditure and Reform has expressed an interest in encouraging public bodies to make their data sets available to the public. Some offices like the Central Statistics Office are already pretty amazing about this. Others less so.

A good number of people turned up on an otherwise quiet Monday night. There were a few academics, a handful of activists, a good bunch of civil servants, and a few representatives from startups and industry. After the formal presentation discussion was opened up, and I enjoyed the variety of perspectives. How reliable are the data sources going to be? How will sustainability be ensured? Will there be any enforcement, and what form would that take? What about groups that don’t want to release their data at all?

Since it was an information-gathering evening as much as a briefing, D/PER didn’t have answers for a lot of those questions. They invited further comment, and the deadline for submissions is this Friday. The format for submissions is…wait for it…an e-mail address. Not a forum, not even a form on the website. So in the spirit of open discussion, this is what I’ve sent as my input towards Ireland’s open data strategy.

Thank you for engaging the public the issue of Open Data and creating a strategy for Ireland. The public briefing on 8 September brought together a fantastic array of perspectives and concerns, generating a good set of questions that a complete open data strategy should address.

I’m a researcher in Trinity College, so my interest in data availability focuses primarily on having that data available as the raw material for research. When I’m developing predictive models or exploring the effects of change, I look for existing bodies of data to test those new methods with, ensuring the model is accurate. The key concern I have, then, is that the data is easy to find and obtain, and that it’s in a format that’s easy to read (machine readable for preference, but having multiple formats available would be even better) and work with. Simple is better. And when data is updated, older sets should remain available with version numbers — this is key for ensuring research is reproducible for other researchers.

One of the greatest risks in this undertaking is that the work will end up being lost due to lack of maintenance. As technology and the means for accessing data change with time, the data itself will need to be taken care of to ensure continued access to it. The best way to achieve this is by hiring people to do just that. It’s hard to justify in a tough climate for public service hiring, but without this investment there’s a great risk of all the current effort (and funding!) being wasted. Even if it’s just a small team of people based in one department who are responsible for curating Ireland’s public data, that might be enough. This could also be an incentive for various public bodies to make their data available, if a team of data experts could be made available to anyone who wanted or needed that expertise.

It’s an investment, certainly, but I think the return would be dramatic.

Thank you again for making this a priority, and I’ll be very interested to see how the strategy develops.