COBOL and governmental efficiencies
- 12 January, 2009 08:10
Welcome to a New Year. For the US, 2008 saw our 401ks become 201ks, our worries about the price of gas come and go, our house values plummet, our economy implode and our IT budgets shrink. It was not a good year.
Will 2009 be better? Well, at least we have a new administration about to take office and the transition already looks orders of magnitude more organized and, dare I say it, more professional than the last four administrations.
One of the recent appointments President-elect Obama made was to a new post, chief performance officer, and the lucky victim, er, office holder is Nancy Killefer, who has impressive credentials.
The CPO's job will be to increase governmental efficiencies and eliminate wasteful spending. This is not a job for the faint of heart.
When he announced Killefer's appointment Obama said: "We can no longer afford to sustain the old ways when we know there are new and more efficient ways of getting the job done." Never was a truer word spoken. With the latest projections calling for a fiscal 2009 U.S. budget deficit of US$1.186 trillion, a bigger vision and a more strategic approach is needed, and government IT operations have got to be a crucial concern.
The reason for this is that government IT has been extremely conservative. Consider COBOL -- most of us probably think it's already dead, but no so fast. For all the advances we've had in programming languages and techniques over many decades, it turns out COBOL won't be pushing up daisies any time soon, particularly in the government sector. According to COBOL purveyor Microfocus, COBOL programs today process 75% of the world's business data and around 90% of all financial transactions.
Take a look at California. As part of his efforts to cure the state's appalling budget deficit, Gov. Arnold Schwarzenegger demanded state workers be paid on a minimum wage basis, but was told by the state controller, John Chiang, that no such thing was about to happen.
This was not some kind of disobedience on Chiang's part, but rather a result of the state having a payroll system written entirely in COBOL. To save money, California had laid off all temporary and part-time workers, including a lot of consultants. You guessed it: The only people who know COBOL are the consultants. And even if they were still employed by the state you don't change COBOL installations the size of California's payroll system in weeks or even months.
California has been trying to revamp its payroll system for a decade and it is estimated that the cost would now be almost $200 million!
This is but a single story of the government's use of COBOL. You can bet that there are thousands of COBOL-based systems in use at both federal and state levels that desperately need overhauling. Unfortunately these projects have been put off for years just like California's payroll overhaul, and in just the same way, the costs of making any meaningful change are going through the roof.
Here's my concern: Getting government systems away from outdated platforms is a huge undertaking and the true costs are completely unknown. But unless we do upgrade we're doomed to stagger on with the inefficiencies and inflexibilities of systems that are decades past their prime.
Government computing is where this country desperately needs to have a strategic focus and spend a lot of money because the longer we put off this work the more it will cost and the less able the government will be to address the data processing demands of the 21st century.
Good luck Ms. Killefer.
Australia lags Mongolia in Internet speeds
Salesforce.com to buy Clipboard, shutting down service
Investor tips on how to propel a startup
Review: Nokia Lumia 520
Review: Nokia Lumia 520