Blog post Goes Live: What’s Next?

By Andrea Di Maio | May 22, 2009 | 0 Comments

open government data

With a lot of anticipation,, the “one-stop repository of government information and tools to make that information useful” (as defined in the relevant White House announcement) has been launched on May 21st.

As one may expect, there are still few data sets, but the approach is quite clear. Data are provided in raw form (in XML, KML, ESRI, Excel, text, etc) and are described using a common metadata, covering attributes category, data, agency, description and so forth. The site also provides tools such as “widgets and data-mining and extraction tools, applications, and other services” to help use data.

I am sure critics will be abundant. Some will claim that data that are most important to them are not there yet. Some will say that tools could be better. And I guess some will complain that the site provides data in proprietary formats (such as XLS): I can hardly imagine what reaction this single fact would create in many government circles on the other side of the Atlantic.

Who cares though? This is a pragmatic initiative, and the real issue is not how many data gets published and when, by whether and how they will be utilized to create value.

Everybody agrees that public government information has a great value, and the ability to mash up data across the whole public sector and beyond has a great potential. But the problem is exactly that: we have been talking about potential for too long.

Let me take one of the most recent success stories about open data and mashups: This is something I followed with interest last year, and one of the last initiatives that Vivek Kundra, Obama’s CIO, took when he was still the DC CTO. The idea was simple: get people to submit ideas to show how value can be created from existing public data. DC had over 200 data sets and feeds, and in 30 days they got 47 submissions, all using a very small subset of those data.

Was that really a success? Certainly yes, if one takes into account that it was one of the first times anything like that had been tried in government. Less so, if one considers that DC has been providing those data for years.

This teaches us (and the Obama administration) an important lesson: providing a wealth of data is a necessary condition for open government and value creation, but not a sufficient one. More emphasis is needed on stimulating people’s and vendors’ imagination on how to leverage those data.

Therefore I do really welcome a joint initiative by Sunlight Labs, Google, O’Reilly Media and Techweb initiative called Apps for America 2: The challenge. This works pretty much like AppsForDemocracy, in so that it solicits applications that use data on and makes provisions for prizes up to 10,000 dollars.

Comments are closed