re: FBI CIO Exits, With Sentinel Project Nearly Done
As the program has yet to deliver, the belief that FBI is out of the woods is a bit premature. Studies by Gartner and others tell us that s/w development if very high risk (63% failures in industry and 84% failure rates in govt), regardless of what new buzz work method is employed (Agile, Scrum, etc). As with all government agencies, the FBI will lack access to the high quality developers who demand premium rates ($500/hr plus). The reason that FBI has failed twice is because their suppliers are forced to deliver low cost developers who are inexperienced.
FBI would have had a working system by now had they not followed the design to spec weapon systems approach perpetuated by our FFRDCs (Mitre), which both violates FAR restrictions onFFRDCs providing a commercial service and Clinger Cohen Act directive requiring that COTS be favored over custom development. Aerospace Corp, another FFRDC, found three COTS Case File Solutions that could have been modified for a fraction of the $700M already spent. Mark my words, FBI will fail again, but Chad will have already washed his hands of this mess. Very timely departure if you ask me.