[Infowarrior] - Analysis: A litmus test for info-sharing

[Richard Forno]

Analysis: A litmus test for info-sharing
By SHAUN WATERMAN
UPI Homeland and National Security Editor
http://www.upi.com/inc/view.php?StoryID=20070109-041222-3710r

WASHINGTON, Jan. 9 (UPI) -- The litmus test for the success of post-Sept. 11
efforts to overhaul and integrate the information systems of U.S.
intelligence agencies is an ambitious new project code-named Railhead, which
aims to create a seamless network of networks within which data about
terrorist threats can freely and quickly flow to those who need, and are
cleared, to see it.

Dale Meyerrose, a retired Air Force general who is in charge of information
technology for the 16 agencies, told C-SPAN Television at the weekend that
the project would create a single "information sharing environment within
the National Counter-Terrorism Center," the multi-agency hub that vacuums up
and analyzes terrorist threat intelligence from every corner of the U.S.
government.

Meyerrose revealed the project's codename for the first time in a later
interview with United Press International, saying Railhead was one of
handful of programs his office was pushing forward as models of how agencies
could integrate their information technology in a "synergized and
cooperative" fashion.

The National Counter-Terrorism Center is the poster child for the efforts to
reform U.S. intelligence, by forcing the sprawling and secretive
bureaucracies of the spy agencies to collaborate more closely. But it has
also become a symbol of the hoops analysts have to jump through, switching
from network to network, sometimes on different workstations, to access
material from different agencies' databases because policy development and
government acquisition have lagged behind technological advance.

"Wrestling the (information technology, if you will, becomes more of what
the person does in their workspace rather than adding intellectual value to
the (intelligence) product that they're supposed to," Meyerrose said. "My
job is in essence to make the (information technology) invisible."

Railhead aims to do that by putting into practice at the center Congress'
vision of an Information Sharing Environment -- a sophisticated platform
that will allow counter-terrorism information, including personal data about
Americans, to be securely shared in a variety of ways that reflect and
respect the different rules in place in different agencies to protect
individual privacy and information security.

Meyerrose said Railhead and the other pilot projects were also testing a
controversial new acquisition strategy called "spiral development."

He said a revolution in the way the government spends billions of dollars a
year on computers and software is essential to keep up with fast-changing
technologies.

"What is possible, what is doable, and what is probable, changes every 18-24
months in the information technology world," he said.

"Traditional, 'big bang' acquisition strategy, where you outline your
requirements one year and take delivery three or four years later ... can't
keep up" with that rate of technological change, he added.

Meyerrose said traditional acquisition was a "failed strategy," which had
contributed to debacles like the National Security Agency's disastrous and
highly classified Trailblazer program, and the FBI's failed Sentinel
project, which congressional critics have lambasted as hundred-million -- or
in Trailblazer's case multi-billion -- dollar boondoggles.

To fix the way the nation's spy agencies develop and deploy their
information technology, Meyerrose said, spiral acquisition was a way of "not
designing to requirement, but designing to opportunity," because
intelligence agencies and the firms they are working with on new technology
could change the specifications of a project as new capabilities became
available.

Conversely, he said, if a project's requirements could be satisfied by a
cutting-edge technology that wasn't stable enough yet for deployment,
project managers could buy "bridging technologies" to use for a year or two
in developing the project until the more advanced product was ready to be
incorporated.

Prof. James Hendler, former chief scientist for information systems at the
Pentagon's hatchery for cutting edge technology, the Defense Advanced
Research Projects Agency, told UPI that the classic model of spiral
development was the way Microsoft builds software code, in a series of
versions.

"You're developing (version 1.2), testing (version 1.1) and deploying
(version 1.0) all at the same time," he explained.

In acquisition terms, he said, it was helpful to think of a decision to buy
a television set. "You say you want a set that does this, this and this, and
then you buy one." In spiral acquisition, he said, "I'm buying, not a TV
set, but a continually improving way to watch television."

He cautioned that the term was a catch-all used to cover a number of
different approaches to acquisition, all designed to reduce the enormous
time lags that traditional procurement created between concept and
deployment, and free up government buyers to take advantage of emerging new
technologies.

Spiral acquisition is about "designing something that won't be obsolete by
the time you deploy it," Hendler said.

But critics say the strategy can be a ticket to expensive programs that
develop technologies which are late and don't meet requirements.

"We definitely have some concerns" about spiral acquisition, Ryan Alexander,
president of non-partisan government spending watchdog Taxpayers for Common
Sense told UPI.

"The bottom line is, with anything that fails to identify upfront what the
technology it's buying is supposed to do, it's hard to know if you're
getting your money's worth," she said.

Meyerrose acknowledged the criticism, but said his office was working hard
to develop ways to measure the efficacy of the new way of doing business.

"It will be as important and just as tough for us to come up with the right
performance measures ... as it will to come up with the right technologies
to fit with the right processes," he said.

Hendler said that spiral procurements sometimes looked more expensive, but
used the analogy of buying a personal computer. "I can buy the cheapest
computer that does what I want right now. Or I can spend a bit more money
and buy something that I think will last three years. It's spending more now
to save money in the future."

He said that spiral development was a model for acquisition in fields other
than just computers and software. "It's good for integrating any fast
changing technology," he said.

Meyerrose warned that he expected some resistance to the new strategies.
"There are folks who don't like these kinds of changes," he said.

But he added that the Director of National Intelligence John Negroponte, who
announced last week that he was leaving for a job at the State Department,
nonetheless "intends to exert leadership" to make sure needed changes
happened.

"Fixing the acquisition capacities of the Intelligence Community ... is
exactly what the (director of national intelligence) is supposed to do,"
said Meyerrose, vowing "take on" any would-be foot-draggers.

"There can be no pocket vetos in this business," he vowed.

Del.icio.us | Digg it | RSS

[Get Copyright Permissions]E-MAIL | PRINT | SAVE | LICENSE
© Copyright 2007 United Press International, I