Don't dream in vain
Interoperable federal information technology architecture--that phrase, or variations,
keeps popping up in pronouncements from the Chief Information Officers Council.
The CIO Council wants to have a governmentwide architecture.
As Agriculture Department CIO Ann Thomson Reed recently said at an Armed Forced
Communications and Electronics Association conference, "This is one of those issues
where we, as a CIO Council, are going to say, 'This is important to us' " [GCN,
March 9, Page 12].
The council is to be commended for the seriousness of purpose with which it is pursuing
its mandate. In this, the second year since passage of the Information Technology
Management Reform Act, which chartered the council, the group is raising all the right
But it strikes me that the concept of a single federal architecture sounds suspiciously
like the kind of unattainable grand design that simply tires people out and never
materializes. If not approached carefully, such a vision or project--or whatever it
ultimately turns out to be--could, in the end, hurt the credibility of CIOs and the
Reed elaborated at AFCEA, zeroing in on the lack of an e-mail system that spans the
The council hopes that agencies can coordinate their IT plans so that they add up to a
uniform e-mail strategy. Again, a laudable goal. But is it worthwhile to pursue in those
You could argue that the Internet puts everyone on e-mail. But not all federal
employees with PCs have an Internet connection. With the Internet, there's nothing
approaching a suitable directory system, delivery is less reliable than with
business-grade e-mail, and attachments are still pretty hopeless.
At the other extreme, getting every agency to agree on a few compatible e-mail packages
is hardly likely.
As the council refines its definition of this piece of its vision, it should also urge
each CIO to see that interoperable e-mail becomes a goal of his or her own agency. That
alone would be a huge accomplishment for many agencies. At least then, interagency e-mail
uniformity would be starting from a base of a couple dozen disparate systems, not from the
mishmash of thousands that exists now.
Still, it's good that the CIO Council dreams. Somebody has to.