The FBI’s Upgrade That Wasn’t — More taxpayer money down the drain. This is an interesting story to read since it involves both incompetency on the one hand and taking the money anyway on the other hand.
Within a few days, Azmi said, he warned FBI Director Robert S. Mueller III that the $170 million system was in serious trouble. A year later, it was dead. The nation’s premier law enforcement and counterterrorism agency, burdened with one of the government’s most archaic computer systems, would have to start from scratch.
The collapse of the attempt to remake the FBI’s filing system stemmed from failures of almost every kind, including poor conception and muddled execution of the steps needed to make the system work, according to outside reviews and interviews with people involved in the project.
Here is what one of my readers reports on this matter:
“As an independent software developer, I have encountered this situation on multiple occasions where I’m told to be quiet and just “give the customer what they ask.” I would then ask something silly like, “What if what they ask is entirely unreasonable and doomed to fail?” To which the response would be something along the lines of “the customer is always right.” The fact that the FBI refused to go with modifying an existing file system and wanted to rewrite one from scratch is a huge indicator that they had no intention of listening to the contractors.”
Someone should be fired for this mishap.
The fact is that people should not be specifying computer-related systems if they do no understand what is possible and what is impossible. It’s particulalry worrisome when they demand the impossible and don’t even know that is what they are doing despite being told.
I’m a “government consultant” too. Working with government entities can be very different from working with commercial enterprises. Even the non-uniformed branches tend to have more of a military-style management than in the outside world. Although program managers DO have bosses, they tend to have a very very wide degree of latitude in how they accomplish any particular “mission”… it’s often a win-big or lose-big scenario. And those working under such a manager may bitch and moan at the watercooler, but it’s very unlikely that anybody will try to go outside the organizational chain-of-command to do anything about it, even in the face of obvious disasters-in-the-making.
I’ve never worked with the Fibbies, but I suspect that they’re pretty much just like the rest. It isn’t really that nobody CARES, it’s just that once they invest authority in a subordinate, they tend to let it go, avoid micromanagement, and trust that person to get the job done. If they fail, it’s unlikely that anyone will be fired. It goes in the “lessons learned” file, and whoever screwed up probably doesn’t get promoted next time around.
I think it’s probably a legacy of the military, since government and military were much more intertwined in the early days of the US. (I could also be wrong.) But the essence of the military is to take young men and women, give them adequate but not overwhelming training, and then throw them into the fire and see who survives. (If you’ve never really been around the military you would be surprised at the huge responsibilities that are thrust onto VERY young men and women.)
I’m not making any conclusions. Just making an observation.
People don’t realize the value of a feasibility study. A little bit of planning ahead by someone other than the consultants being paid to build it would have fixed this. Unfortunately, most managers I know don’t realize how important this step is.
I’m sure someone will lose their job over this, but I’m equally sure it will not be the person responsible.
3 (If you’ve never really been around the military you would be surprised at the huge responsibilities that are thrust onto VERY young men and women.)
Ha! yeah…
Reminds me of serving on Trident submarines when the missile/bomb command was consolidated, and missile designations were changed from “Strategic” to “Tactical” weapons [since “Strategic” made it sound like we didn’t intend to use them…] and an inspection team with Air Force & Army types came to visit us.
The Air Force types in particular were shocked at how few officers were involved in the process of launching a missile – just a few key positions – with the majority of the work being done by NCOs… [since just about all enlisted guys on subs are NCOs].
many years ago,
thae IRS wanted to update their systems..
They got the quotes, and settled on one.
They THEM had to take it to congress to be ‘Ok’d’
3 years later they got the OK, and gave the contractor a call.
He loved it,
He installed the system and everything worked ACORRDING TO THE CONTRACT…
In those 3 years 386’s had become Pentiums…
He installed the 386’s
Put the profit in his pocket, and LEFT the country.
What is it to SAY, about our country, IF…
They cant handle a decent Contract?
Do know a Serial port from a Parallel port and dont have the BASIC knowledge to get someone WHO knows…
Why should anyone be surprised? This kind of thing is common in industry, too.
I think this specific FBI snafu has been common knowledge for some time. I’ll bet that when the project is analyzed, the cause will be conflicting design requirements. The reader’s report points to that very problem.
Chances are that, by contract, SAIC gave them what they wanted instead of what they needed, conflicting requirements and all.
My experience has been the same as Smith/7, it is a universal problem. Getting a design spec hammered out with input from folks that don’t know what to ask for and supplied to developers that have little experience in the application area is close to if not being in the center of tech hell. It isn’t unique to any specific business or government agency.
You can be up to your ears in money, need, metrics, consultants, case tools, good hardware infrastructure and project management methodology and still end up with a water-proof sponge if the design spec is FUBAR.
When are we going to learn that if you want something done right, you have to offshore the work to someone else.
I am currently working on a project that failed, trying to discover why the company spent so much money, diverted so many people and resources, and allowed profitable ventures to be side tracked.
Investigating Cause and Effect always looks easy but usually turns up a lot more problems then anticipated.
Follow the money on this one.
Software seems to have become the biggest areas of pork, these days. It’s simple! :
1) Start a fly-by-night software company
2) Wine-and-dine a representative or Senator, slip him/her a few thousand dollars to get re-elected.
3) Get a 100 million contract to provide the government some ambiguous shoddy thrown-together piece-o-crap.
Don’t worry about it since nost congressmen know nothing about computers.