Letter to the editor
As a taxpayer, I took note of the attention-getting facts and figures quoted in your March 5, 2001, Federal Computer Week editorial and feature story, "12 years, $1.6 billion and counting; How DOD plans to make its messaging system work."
As someone very familiar with the Defense Message System program, however, I am compelled to submit corrections to the published articles in keeping with your high standards of quality and professionalism in journalism.
The $1.6 billion figure you identified as the DMS contract value is misleading. The contract is of the indefinite-delivery, indefinite quantity (IDIQ) variety and allows a maximum of $1.6 billion dollars worth of goods and services to be purchased cumulatively by all services and agencies. This is entirely different from the estimated value of the contract at the time of award ($499,609,671), and represents the potential cost if perhaps all DOD members had been intended as DMS end users.
The comparison Autodin costs, which you cite as "$30 million yearly" for dedicated telecommunications circuits, actually average upwards of $65 million annually, even after recent declines brought about by steady migration to DMS.
An FCW.com article March 7, "Pentagon finds holes in DMS," states that "DMS is not fully secure" but fails to note the specific DMS version in question. In fact, DMS Releases 2.1 and 2.2 have incorporated automated scripts to properly secure the Unix and Windows NT operating systems. The results of most recent testing found the most serious vulnerabilities were, in general, related to security practices rather than DMS design. Most importantly, test results indicate that when sites properly implement the DMS security guidance, the DMS components are secure.
DMS is a writer-to-reader system and is not reliant on message centers. Message centers exist only as long as Autodin remains operational. DMS is installed today on most command centers servers with organizational messages sent and read from clients in the appropriate hierarchy of the organization, as determined by the organization. I currently send and receive DMS organizational messages from my desktop client for my organization.
The bottom line is: DMS does work today, but achieving the full extent of all service and agency requirements requires an iterative development process responsive to evolutions in requirements definition and the technological state of the art.
Your editorial comment, "Procurement Ten Years After," fails to recognize DMS as a pioneer of commercial off-the-shelf-based acquisition.
Early in the reform movement, when building DOD-unique solutions from the ground up was the status quo, DMS used COTS technologies to the maximum extent possible. Only those confirmed, critical warfighter requirements deemed unachievable (within required time lines) using commercial products were pursued as DMS-unique product add-ons.
The DMS objective has been to evolve until DOD messaging requirements are incorporated or converge with those of standard commercial products. That's why the program has focused diligently on constant revalidation of core requirements. Rethinking of the desktop-to-desktop requirement is just one example of DOD efforts to maximize acceptable trade-offs between cost and functionality.
The DOD decision to use COTS e-mail with DOD public-key infrastructure (PKI) is another. It represents an evolution of DMS products based on customer needs and commercial product capabilities, balanced against program cost, which has saved the DOD millions of dollars.
Procurement and IT management reforms were critical and needed events within DOD, and DMS is proud to have been among the first wave of COTS-based acquisitions. More recent IT efforts benefited from lessons of the DMS acquisition, but DMS itself represents a successful outcome "snatched from the jaws" of potential procurement tragedy.
The program successfully overcame a 22-month delay in initial operational capability due to protests of the initial contract award, and a failure of commercial product providers to introduce the proposed underlying commercial solutions at the forecasted time. COTS-based acquisition does have its risks: dealing with commercial practices is one of them.
In closing, I'd like to point out that the letter to the editor "Getting to the source of DMS problems" published March 12, 2001, is an example of gross misinformation and obfuscation of facts by emotional rhetoric. The replacement of Autodin was solicited neither as a firm-fixed price effort, nor a two-part effort and always consisted of a two-year base period contract with six one-year option periods. To state that Autodin closure may now occur (in 2003) "four years past the end of the original DMS contract" misrepresents the facts. The DMS contract is not a vehicle for commercial software products, and because unit prices include the costs of DMS specific add-ons, it could never be competitive as such.
The contracting agency and the DMS Program Management Office are both held accountable through the standard mechanisms. The Air Force contracting unit exemplifies the highest standards of acquisition skill and diligence. The PMO answers not only to the DISA director through procurement and budget reviews, but directly to the Military Communications/Electronics Board, the Joint Chiefs of Staff and the Office of Secretary of Defense.
Chief of public affairs
Editor's note: Of all the agencies interviewed for the DMS article, only the Air Force and Navy provided cost figures.