Shaky Foundations of Federal IT


The White House blog today featured a new post about the “Building Blocks of a 21st Century Digital Government.” If these are the building blocks of reinvented government, however, we’re on shaky ground.

Most agency CIOs don’t know what their agency’s major IT holdings are. Really. Decisions determining what data will be released, and how it gets released, are routinely made by individual departments, outside public view, and without review from the federal CIO or CTO, Congress, or the public.

This is a shame, because the $80 Billion+ federal IT budget contains a wealth of vital information, and should be considered a national asset, deserving thoughtful consideration. Instead, the identity of the major information holdings of the US government are still essentially opaque, even to the government officials supposedly in charge of managing them (with only a few exceptions).

Unfortunately, the major information policies of the Obama administration have all punted on this essential point. Transformative open data policy isn’t just about APIs and data portals.

This is frustrating because the White House is clearly spending an enormous amount of time and attention on the Federal Strategy, and doing some fantastic, innovative work. The GSA’s data stream of agency progress on the new requirements is an exciting new way for the White House to track compliance on explicit requirements (even if some of them are redundant with existing requirements). And the new Innovation Fellows program shows an administration clearly receptive to new ideas and personalities, who will inevitably make some progress facing important challenges.

But there’s still a glaring, glaring omission. This strategy still doesn’t address existing information that isn’t open, and it doesn’t empower federal managers or anyone else to get new access to existing information. When the White House says “open up” they mean “build apis,” not release information for the first time.

Well formed federal transparency policy has to be built on a circumspect, comprehensive foundation, with knowledge of all major information holdings.

Government information policy can’t be about opt-in, voluntary policies that encourage agencies to try some new things. Ultimately, that’s what the Federal IT policy preserves: the status quo, where agencies pick comfortable data to release, often without even knowing what they have to choose from.

You can’t manage what you can’t see.