City Camp MN: Open data discussion

Otto Doll, CIO of Minneapolis, led a great discussion on open data. Here are my notes on that discussion.

The city of Minneapolis has a lot of data. That gets put together in information presented to the community by the city. However, what raw data should be shared publicly? (Not all data can or should be shared.) How is that data shared? If we put this data out there, do we put it out there forever? Do we keep publishing it forever? Archiving it forever?

What happens if public data gets put out there, an industry starts around it, what happens if data gets shut off at some point? Or if government starts charging for data access? Provide different levels of data: ie Service Levels of Agreement? Google Labs model? Are there models in the private sector to see what happens when data gets shut off or when costs go up?

Usually governments have records retention policies.

Do we start with putting everything out there and decide what not to put out? Put out XML data that can get manipulated?

Do we put out data and not tell you what it is? Do you share the metadata?

Do you believe Google Maps? Sort of. Some data years old. Wouldn’t it be nice to be able to see age of that data?

Pushback: We can’t just put it out there, people will abuse it! Will people just manipulate data we already package anyway?

How many ways do we need to make data available? Is there one perfect data format? Or should we just put out there whatever we have right now, Excel files, for example, and then let private sector convert it into better formats? XML if possible, is nice.

Look at data from perspective of users, not from government. How do users of this data want it?

How do we choose what data is relevant and valuable? Allow citizens to determine is valuable? Allow people to vote either through a voting system or through usage statistics.

If some data is pushed into the cloud, does that free up staff to work on other issues? Minneapolis already outsources a lot of IT already, so maybe not. They may already have data in the cloud, but that doesn’t mean it is accessible to the public.

Should city develop apps? Should business always be the ones who develop apps, build a business around it, and allow other cities to use that service? Are there particular services that a city delivers that could be done well with an app? What if there isn’t a good business model around a particular service? Is open source a good way to handle this? Civic Commons is working on open source sharing for government. If there is a business that does the app and charges for it, if the cost is too high, is it right for government to not make that available to all? What if government needs to charge for it? Should we subsidize some people?

Having all data available is very important to the public.

Putting data out there can help tie data together over multiple levels of government.

What about creating APIs to feed data to the government. Open 311 is a great example of that. Citizens as sensors. Open 311 allows people to use different apps and have all of them feed into government internal data systems.