Letters to the editor
I am reacting to Milt Zall's column, "Outsourcing conundrum" [FCW, June 25].
Fortunately, we live in a capitalistic country. This means that government should be limited to performing what is "inherently governmental" in nature police, fire department, border patrol, the military, etc. Everything else belongs in the private sector. The reason we are the richest and most powerful nation on Earth is precisely because the capitalistic system outperforms socialist systems every time.
Zall argues that the only reason to "outsource" is for the government to determine the lowest cost of performing a particular function. As a contractor who has bid on those types of competitions, I know that they are essentially "fixed." By Zall's own admission, government officials cannot determine what it truly costs to run a particular function. They have no reason to because they do not "compete" with anyone other than occasionally other government agencies for work. They cannot do activity-based costing because they cannot determine "overhead, support and nondirect costs associated with their programs."
In the case of the Defense Finance and Accounting Service, its first nine A-76 competitions all came out in favor of the government. Facing criticism that these were fixed, DFAS officials recently announced that the most recent competition was awarded to industry. They "gave one" to industry to avoid further criticism.
I am a retiree from federal service. I'm glad I served, but I'm also glad I am no longer surrounded by those who could not hold a job in the private sector if they had to.
Name withheld upon request
The articles in your magazine on gridlock and highway information support ["Grappling with gridlock," FCW, July 2] miss the primary question: "Why are we still driving to work?"
There is no reason why most federal workers have to travel, other than that is how it's always been done. Options include working at home with videocams or at local work centers. Face-to-face interaction may be needed for some applications, but does it have to be all the time on the same 9-to-5 schedule? Many (most) federal workers are in a cubicle with limited interaction.
Tuning up Performance
Your article "OMB turns up heat on performance" [FCW, June 25] was timely and interesting but missed the causes and cures of the difficulties with the information technology performance issue. The essence of the issue is confusion about what constitutes IT performance.
There are two dimensions of IT performance: (1) new applications, including major enhancements, and (2) existing applications and resources (e.g., hardware, software, networks and support organizations). Both are important and each requires a different approach to performance meas.urement and management.
Existing applications and resources are today's production resources. They typically consume 70 percent to 80 percent of the IT budget. The money for them has already been spent. Measuring their performance is about cost- efficiency (i.e., how do our costs to support an end user compare to others with similar distributed environments) and/or productivity (such as the number of end users supported per support person vs. comparable environments).
Comparing similar workload environments is critical. These performance measures and management techniques are oriented toward cost reduction. For new development (including major enhancements to existing systems), performance measures reflect investment decisions and must be tied to organizational performance. Since the function of these metrics is to allocate limited IT resources to deliver maximum "value," it is critical that they be tied to outcomes of major importance to the organization.
The best performance measures link IT performance to organizational and/or program outcomes in a cause/effect relationshipthey predict outcomes rather than account for past results. They also describe mission performance from more than one perspective, not just financial. As an example, the balanced scorecard model recommends four perspectives: financial, customer, business process and learning/growth. The issue is not just measurement; it is delivery of maximum "value" as measured from multiple predictive perspectives.
Finally, in my experience, the major obstacle to effective IT performance measures is fuzzy business goals. The real work lies in clarifying the goals (of which IT measures are a subset). As a result, one of the most important benefits of an IT performance initiative is getting IT managers and business managers working together to identify the goals and performance measures that define program success, including the IT contribution to achieving them.
Mark Hess, President
Weston Technology Management Group LLC