- By Gary Swindon
- Mar 15, 1998
The only activity of consequence in a public-sector chief information officer's or senior information technology manager's professional life, I believe, is making the right infrastructure decision.
Senior managers in very large jurisdictions-such as major federal agencies and the 10 or so largest states and counties, such as Los Angeles and Wayne County, Mich.-are under enormous pressure to produce results that save large sums of money, speed up service delivery and vastly improve government business processes. If that weren't enough, they face the daunting challenge of actually carrying out the vision behind their infrastructure decisions and taking responsibility for the impact of those decisions in the long run.
Unfortunately, one consequence of having to deal with monolithic problems is that CIOs sometimes start looking for an easy way out.
The result is often a solution that is itself monolithic in nature and that turns for answers to the major IT industry trends of the last five years-client/server computing, the Internet and intranets, network computers (or was that networked computing?) and the privatization and outsourcing movement in its various forms. Considering the mixed successes of each of these technologies, one wonders whether a simple monolithic solution to a complex problem is indeed too good to be true.
Of all the trends noted above, the most insidious has been outsourcing-that is, the simple act of giving your troubles to someone else so you won't be bothered by them any more. Current examples are not hard to find, including California's decision to outsource its networks and Pennsylvania's intention to outsource its data centers. But top honors must go to Connecticut, which is attempting to give all its IT functions to some lucky vendor. While many states are larger, none had previously stepped up to outsource everything.
On balance, it seems likely that if "sticker shock" doesn't get them, then the challenge of putting everything into the vendor community's hands probably will.
Perhaps the most basic question raised by these cases is whether governments should own some or all of their IT infrastructure. To arrive at a conclusion that is right for your jurisdiction requires an understanding of what you expect your government to be when it grows up. It is a simple truism that good decisions come from a basic understanding of where you are going as an organization, who you serve and what trends will play a part in any solution.
Sadly, any decision requires tenacity in the face of pressure from outside groups that something be done right away. (And, by the way, it had better be the right decision!) But the bottom line is that CIOs and other IT leaders in government must be held to a higher standard than their industry counterparts. If you serve the taxpayers-and in the end, they are the only constituents who matter-then you must keep them in mind as you decide whether to give all your computing power to someone else. After all, government is not about saving money, giving infrastructure away or taking the easy way out of situations. It is about serving the public. Pity the CIO who believes that his job is about technology. Even more, pity the taxpayers who will reap the benefits of having a CIO who holds that benefit.
Should governments own their IT infrastructure? Absolutely. Should governments investigate outsourcing? Yes, but only where the outsourcing decision doesn't become a surrogate for intelligent and informed decision-making. Public ownership doesn't necessarily mean owning everything, any more than outsourcing should be all-inclusive. But remember, even if outsourcing some IT infrastructure seems to be the wise course, it does not absolve government of its responsibility to manage what is outsourced.
-- Gary Swindon is the director of the Office of Computing and Telecommunications for the state of Michigan. He can be reached at firstname.lastname@example.org.