Earlier this week I was in Washington, D.C. where I addressed the Association of Public Data Users at their Annual Conference (#IgniteAPDU). This conference brings together data users and data producer, mostly at the federal and agency level. Topics included those covering statistical policy, data production, data collection, data delivery, emerging trends, emerging challenges, and new techniques for dissemination. At a time when budget cuts are forcing producers and users of data to think creatively about how to do more with data with far fewer dollars, inspiring new thinking is integral. This is why the theme of this conference is innovating with public data.
New thinking about traditional agencies was also the theme of a presentation by Daniel Castro, CP of the Information Technology and Innovation Foundation. Daniel is a policy analyst and writer who has a particular focus on data.
When it comes to public agencies, he sees them in falling into two categories: Libraries and Museums. When you are a library, you store data. You want people to use that data. And you are physically built for that purpose. He used this image of the Trinity College Library in Dublin to make this point:
In this case, the library was architected to store books. They want you to sit quietly, solo, and digest their collections. But libraries are changing. The architecture and design are reforming to accommodate a new intent around collaboration and collective learning. And not necessarily using books.
As an example, Castro showed a picture of several newly designed libraries with a diverse set of spaces, including collaborative commons, such as this one from the 
University of Texas.
When you take a closer look at this space, it features diverse ways of both working together or working individually, letting the users decide which is the most appropriate for the task at hand..
Next, Castro turned his attention towards museums. An agency is a museum when its
function is to curate the data, tell the stories, and create the narrative. And museums are adapting to new models as well. Museums of the past were goaled on “cultural preservation and also to wow people with a sense of awe”. As an example, he showed a picture of the Hermitage in St. Petersburg, Russia. Notice in the picture, you have a sense of grandeur, yes, but more importantly, permanence. Exhibits are placed to stay. You would never think of nailing something to that intricately decorated floor or hanging something from that exquisitely designed ceiling. After all, the building itself is a very formal backdrop to what artists are contributing, and a work of art in themselves.
Now think of the art museums of today. In this picture of the Whitney, the floors are wood (recycled). An exhibit can be placed, replaced, moved, and changed. The ceilings are open. Nothing is precious in the infrastructure. Artists are invited to come in and experiment with the space. It is designed to be dynamic. Artists needs will change, as will visitors. And it is not designed to stand in silence. Many parts of today’s museums are designed to inspire discussion and collaboration.
And so it is today with respect to public agencies. Public agencies have always produced data. They may have technology today that provides ways of producing more of it, but they have always done it. The task of providing data was characterized by the following:
- It was a solitary endeavor designed to provide data for the individual.
- It was refined. Permanent. Designed for static production and use.
- It was communal. One size fits all, as if it data were served from a vending machine.
- It was fixed, designed for what the system was capable of providing at the time.
- And the goal was giving answers to those who asked for them.
Like libraries and museums, agencies of tomorrow will change. And the change will be driven not only the technology available, but by the expectations and usage patterns of the user. Data production and use by agencies going forward will be:
- Collaborative, making heavy use of APIs and open distribution methods like GitHub
- Experimental in nature – leveraging technology practices such as user testing, alpha testing, beta testing
- Personalized and customized to the individual requesting the data for a variety of use cases
- Dynamic and flexible so that it can adapt to both the technology evolution and usage patterns.
- And finally, it will be driven by a goal of raising questions in addition to providing answers.
My experience in Chicago is that city government is already ahead in moving in this direction. The department of innovation at the city has been thinking this way for some time. Now they are working on inspiring other departments to adopt this kind of thinking. It was very enlightening to see that agencies at the federal level are looking to adopt the same ethos.
