Why not more software reuse?
Blogger Steve Kelman believes the federal government could make big gains by making it a priority to reuse software wherever possible.
This has been a great week for learning from my students. I just blogged earlier this week about a student discussing risk-taking and short-term government appointments with me. Since then, another student -- who had been a Marine Corps captain and is now a reservist, working in Marine Corps IT -- came by to talk with me about some of the things he was working on, involving faster and better ways to analyze intelligence, before he came to the Kennedy School.
He said a number of interesting things, including a bunch of sources of frustration with paperwork-heavy review processes that clog the arteries of the system and dramatically slow down decision-making. But he said one thing in particular that caught my attention.
He noted that, in private-sector IT development projects he had examined or studied, companies always first asked, when starting a new system, what they could reuse from applications they had already developed. Reuse, he noted, was both a cost saver and a time saver for new project development. By contrast, in his military IT experience, people seldom if ever looked at what could be re-used when starting a new project -- and IT vendors did nothing, he noted, to help the government ask that question or to suggest re-use opportunities.
We talked about this a bit. I made the observation that, unlike the case with a number of other dysfunctional government practices in the IT development arena, there wouldn't appear to be any government-unique regs or requirements driving this behavior -- only cultural, organizational or SOP kinds of problems. My student agreed. If this is the case, incorporating into the culture an automatic look at reuse opportunities might be a low-hanging fruit for IT development improvement.
Having hung around the government IT development management work for a while, I have come across mention of this issue on occasion. I believe it came up at least peripherally in the work of the recent Government Technology Opportunity for the 21st Century commission that I co-chaired with Linda Gooden from Lockheed. But I don't think anything about this got incorporated in our report.
I am raising now this issue in the blog to get reader reactions and comments. Is software reuse in developing new applications something that deserves seriously more attention in the federal IT world? Why or why not? If yes, why aren't we doing this more already, and what can be done to increase attention to this?