Stop them before they code again
November 15, 2006
Grandiose software projects rarely succeed, and when government's involved the long odds get even longer. In the new Baseline, Laton McCartney chronicles the disintegration of the vast, multibillion-dollar project aimed at overhauling England's health system. Four years into the 10-year effort, called the National Program for Information Technology, the original $12 billion budget has already ballooned to at least $24 billion, which is more than it cost to dig the tunnel under the English Channel. Two members of parliament recently wrote that the project "is currently sleepwalking toward disaster. It is far behind schedule. Projected costs have spiraled. Key software systems have little chance of ever working properly. Clinical staff is losing confidence in it."
On this side of the Atlantic, we've had plenty of governmental IT disasters as well, including the IRS's systems modernization project and the FBI's ill-fated "Virtual Case File" system. And now we have another to add to the list, and in some ways it's the most mind-boggling of all. A Senate Committee is scheduled to have a hearing Thursday on the star-crossed "Defense Travel System" (DTS) being built by the Defense Department. Launched in 1998, the project is currently four years behind schedule and has cost $474 million, $200 million more than the original budget. And few people in the Department of Defense show any particular desire to use the system.
So what is DTS designed to do, exactly? It's a system for booking airline tickets and making other travel arrangements. Yep, the government has spent nearly a half billion dollars to essentially replicate the functionality of the many Internet travel sites that are already up and running and used daily by millions of people. And after eight years, it's still not working. A spokesman for the main contractor, Northrup Grumman, tells ABC News, "This is a program that the American taxpayer, the Department of Defense can be proud of." Oklahoma Senator Tom Coburn begs to differ. He calls DTS "a hoax on the American taxpayer." His colleague Norm Coleman, of Minnesota, says, "DTS does not list all available flights and does not always identify the lowest available fares, does not take rail reservations, does not reserve hotel rooms or rental cars." He says that only 17% of 755,000 trips recently booked by Department of Defense workers were processed by the system; the rest went through travel agents.
What's incredible is that this is a perfect application to be supplied as a service over the web - all the information, after all, is already out there. Indeed, Rearden Commerce, which yesterday announced a big partnership with American Express, offers a state-of-the-art, software-as-a-service system for managing travel and other personal business expenses. It interconnects with the myriad web-based systems run by airlines, car rental agencies, hotels, airport limo companies, package-delivery firms and so on to provide a one-stop portal for employees. Companies can build purchasing rules into the system, routing employees to preferred suppliers or specially negotiated deals, for instance, while still giving users the simplicity and flexibility they've come to expect in booking travel and arranging other services online. I asked Rearden CEO Patrick Grady what he thought about the travails of the DTS project. He offered two words in reply: "outrageous" and "unreal."
It's one thing to fail in crafting a complex software system that hasn't existed before. It's another thing to waste huge amounts of money and time trying to build a proprietary software system when all you needed to do was launch a browser.
Are these cases of IT mattering just too much?
Posted by: Jon Collins at November 16, 2006 03:02 AM
Add the FBI attempt at a LIMS system (http://www.usdoj.gov/oig/reports/FBI/a0633/index.htm)
But what does it mean? No IT? Move everything to Google?
I would say that these are prime examples where IT (and more so IT management) does matter. These are projects best broken down to smaller modules united by policy, standards and infrastructure services.
Each module should both adhere to the above and be able to function on independently.
Consider health care. There is no reason the GP, Labs, accusation and other systems should depend on each other for development, usage and day to day operations. Only data interaction matters.
Maybe DTS lists the black bag flights from Kabal and Baghdad to Gitmo. You wouldn't want to miss one of those.
And wouldn't you just love to be working on one of those projects? The death spiral begins early on, not long after the first individuals realize they're going down.
There's this idea that, "Well, you're getting paid, aren't you? Do what you're told, whether or not it's going to produce a functional system." is a valid management technique. It's especially true of consultant-driven projects - cash cows for the consultancy, boondoogles for the client.
Posted by: marc moore at November 16, 2006 08:37 AM
Profit, efficiency, and budgetary discipline are NOT goals of Government. In fact, they are counter to the goals of most lobbyists, politicians and lifelong government employees.
It's not really about the software at all. In fact, I would contend that $24 billion dollars is really not that much at all in comparison to so many other failed ventures.
Posted by: coders4hire at November 16, 2006 11:47 PM
coders4hire, this statement: "In fact, I would contend that $24 billion dollars is really not that much at all in comparison to so many other failed ventures" makes me think you are absolutely stark raving bonkers.
On the other hand, if you count $200 billion or so for Operation Desert Boondoggle, maybe you're right.
(In terms of software, however, it's got to be the biggest toilet flush ever drained. Even more than Vista.)
Posted by: John Koetsier at November 22, 2006 02:02 AM
We've had a similar disaster story here in Ireland.
The Personnel, Payment and Related Systems (PPARS) project, which was supposed to automate payroll and other employee-related services in state-run health organisations, turning into a fiasco.
Before the Government pulled the plug on it in 2005, PPARS was "25 times the original estimate, and the system is four years behind schedule."
In one pilot scheme, an nurse was accidentally overpaid by 1 million euro! (She reported the overpayment immediately. Perhaps she was too honest?)
Posted by: Michael Heraghty at November 22, 2006 07:36 AM
Post a comment
Thanks for signing in, . Now you can comment. (sign out)(If you haven't left a comment here before, you may need to be approved by the site owner before your comment will appear. Until then, it won't appear on the entry. Thanks for waiting.)
"Riveting" -San Francisco Chronicle
"Rewarding" -Financial Times
"Ominously prescient" -Kirkus Reviews
"Riveting stuff" -New York Post