Standards in Government

It's been a really exciting first day out here at Harvard. Jerry Mechling and the rest of the folks at the Kennedy School have put together an outstanding workshop. There are a whole host of impressive attendees (from CIOs of major departments to ambassadors). At lunch Nicolas Carr (author of "Does IT Matter") presented about the importance of standards in IT. I just got out of a discussion led by Malcolm Sparrow but also including Karen Evans, William Bratton (formerly of the NYPD and now chief of the LAPD) as well as Harry Spence. It was a great discussion around the need for better top down communication and a better way to empower those who are closer to the problem so that they feel enabled to come up with the solutions.

There was a really interesting talk with Charles Havekost who helped to get Grants.Gov running with standards. They focused on finding a standard that would cover the core set of data that the various grant applications gathered. This allowed for an online application system that was much more streamlined. It really hit home on the importance of custom defined schema within Office documents. There is no way that folks working on document standards would be able to understand that specific business (grants) and build support directly into the document formats. Instead, with custom defined schemas, any organization can create their own schema or take an already existing standard schema and use it on top of the document. This means documents aren't just limited to specifying formatting and layout information, but they actually can directly identify the particular data that matters in the different scenarios (like applicant's name; address; social security number; etc.).

The other important thing with moving to standards is to look at what the current uses are and see if you can make the move that causes the least amount of disruption, while still achieving the goals you have in the first place. It can be tempting at times for folks to think they should just blow everything up and start from scratch, but this can lead to all sorts of pain, and in the end it's rarely worth it. That's not always the case of course, but that's why you should always look around and see if there is something that will work within your existing systems.

I really got a lot out of this first day. I wish I had more time to write about it, but I have to get back to the dinner and more presentations. I'll fill you guys in more tomorrow.

-Brian