Posted on 08/18/2006 7:13:39 AM PDT by Small-L
It was late 2003, and a contractor, Science Applications International Corp. (SAIC), had spent months writing 730,000 lines of computer code for the Virtual Case File (VCF), a networked system for tracking criminal cases that was designed to replace the bureau's antiquated paper files and, finally, shove J. Edgar Hoover's FBI into the 21st century.
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.
In a 318-page report, completed in January 2005 and obtained by The Post under the Freedom of Information Act, Aerospace said the SAIC software was incomplete, inadequate and so poorly designed that it would be essentially unusable under real-world conditions. Even in rudimentary tests, the system did not comply with basic requirements, the report said. It did not include network-management or archiving systems -- a failing that would put crucial law enforcement and national security data at risk, according to the report.
(Excerpt) Read more at washingtonpost.com ...
From personal experience, that's the SAIC management philosophy--Get a Time and Materials contract, burn through the money as fast as possible, book the profits as early as possible, and worry about delivering a product later--usually with dozens of Engineering Changes to pad to total cost and profit.
Corruption in govt contracting? Whodathunk? (sarc off)
Look on the bright side. Some FBI middlemanager is going to get a nice retirement job.
And people thought that contract corruption is limited to the DOD.
...the SAIC software was incomplete, inadequate and so poorly designed that it would be essentially unusable under real-world conditions.
What the heck are they talking about? It works just fine.
(sheesh, what did they expect for $170 million, Ms. Pac-Man?)
This story took up an entire page and was so disorganized I had to read it twice. Too bad they didn't bother to edit it. Same thing happened to the Border Patrol a decade ago. Why can Visa, Mastercard, ect manage to obtain effective programs for dealing with huge masses of info, but the government can't? Ditto communicating airline passanger lists from abroad before planes take off.
So will SAIC be prohibited from competing for other government contracts.
Sadly, some college students could set up the system using available software. Too many government agencies assume they need new software but don't look at using what is commercially availbe and paying someone to set it up. I bet Oracle or Microsoft could have done the project using an existing product.
;-) One could only wish. The difficulty is that every other Beltway Bandit is just as bad.
ROTFL. SAIC holds more government contracts than anyone else if I'm not mistaken. Plenty of them involve military use. Let's hope they turn out better than this one.
Maybe. There are objected oriented APIs for building database interfaces.
What the government needs to do is develop an extension of one of those API sets for government use. That shouldn't take more than $7-10 million dollars and 10-18 months to develop, and simultaneously develop several government applications, including the FBI application. And any new development after that would cost significantly less, and time to deployment would be minimized.
Open source CMS solutions are great examples of how the modular approach and common, object oriented API can allow the creation of unique websites from a common code base.
From my own experience, it's not just SAIC. This philosophy is rampant in all areas of government contracting.
I did a project for the Navy where the project was managed by a committee that met once a month. None of the people on the committee had any software experience and every time they met they changed the specifications for the project. Every change they made required months of rework and introduced new problems to the software. For example, they asked for a database that would run on Windows 98. We were in the final testing phases when they decided it needed to be compatible with DOS. It started as an MS Access program and they decided it needed to run over the internet.
The initial concept was straight forward and could have been done for a couple of hundred thousand. In the end the project ran over 5 million and was scrapped before it was ever operational.
I'm sure the who thing could have been done with off the shelf software using any number of available data-base programs. The bigger the budget the bigger the boondoggel.
This is what happens when
1) non-tech administrators have to make decisions about tech.
2) mid-level knowledge people have been thinned out in budget reductions or outsourced.
3. See number 1.
Lol, used to work for SAIC myself. It was all about billing.
Its also obvious that the FBI did not look and see what other government/military agencies are using. The FBI is not the only agency that needs to be secure.
I'm sure they are of the mindset that they need a "different" software when one might already be made and just needs to be modified.
But, the GOOD news is that SAIC got $170 million!!
Moral of the story: DON'T buy a cow until you've tested the milk FIRST!!!
Agreed about probable cause.
SAIC, CSC, and a bunch of others do really good work for the Navy, Intel community, and many more technically competent government agencies with a good cadre of technical program managers and a clear idea of what they really want.
DOJ, and certainly the FBI, aren't among those agencies. They have a cop attitude, and if you don't wear a gun, you don't really count.
Looks like SAIC was getting paid in proportion to the number of staff who were "working", with no relation to the amount of work that was getting done. I've seen this happen with military contracting.
If the programmers were put into direct communication with the actual users of the product, there would be a joint effort to develop something useful. But there are always layers of stupid useless bureaucracy on both sides.
The end result is that the proper goal of doing something good for the country while making an honest buck becomes merely a game to keep the billable hours flowing.
Actually, a lot of software developed for the U.S. government appears to be "single use" and is often locked down pretty tight in terms of where and how it is going to be used.
What the U.S. government needs to do is develop a high level, object oriented API toolbox for developing a secure, modular systems for storing and recalling content in government databases. The front end application would merely be an interpreter for the high level code, much like a web browser... In fact, it could be a web browser designed and compiled to work with enhanced security protocols for the new toolbox.
That would be an interesting aspect of a special, secure browser for accessing secure government database applications. By default, if a browser can't convince the database application that it is a secure browser, its access to the database would be limited automatically, regardless of what access level the user may have usually.
Being here in the market area, almost anybody in this field knows something about this failed project. I heard it was about 60% SAIC and about 40% FBI's fault. Sometimes these government clients can be difficult, speaking through experience.. I heard that new and revised requirments for the project were like 10+ a day. Thats lack of focus on the FBI part and lack of skills on SAIC's part to set them straight. Lack of professionalism on both parts. This does not excuse SAIC, because they have to be held accountable for garbage they produced, and from my experience, I have not been all that impressed by SAIC.
Here is how the game is played in DC though, if SAIC tells the FBI "You are screwed up", the FBI then says "You're fired" and then offer the contract to a competitor.
You get what you pay for!
If the contract is set up so the company is paid for a worthless product, or worse yet, paid before even delivering a product, then that's what you'll get: crap.
OTOH, if you set up a contract to pay only after you receive a functioning, quality piece of work that meets all requirements, then you'll get that, and more quickly too.
Don't pay a dime until you have what you ordered. It solves many problems all at once.
You haven't seen what the army uses now for operations. It uses a MS office platform. The maps are digital, and you can do digital overlays(no more acetate and perm markers) and you can then send what you have done to other units.
Compared to the days of having to trace and using copiers, the army has come a long way to being paperless.
I spent five years working with SAIC...and there were rare occassions that I could see terrific work being done and no fingerprints left by management. But note...I said rare occassions.
The company has an interesting approach....they have tons of people on the payroll throughout the organization who simply sit and wait. When a juicy contract comes up...they look for a connection to the top customer...and they find some guy who worked with him years ago. They bring in their budget analysts who figure how to get the most cash out of the customer. They size up their competition and often get the smaller and less costy guys out of the way by sub-contracting part of the deal down to them if they just withdraw. Finally, upon winning, they find division chief who rarely knows anything about project but is a favorite or darling amonsgt the SAIC management. Other favorite managers and team leaders are brought in...and they forge ahead. The real players...the planners, the programmers, the analysts....are usually kept in check by SAIC's no-risk concept...preached over and over. Don't do anything that might upset the customer....invoke as little time with the customer as possible, and simply look busy (you must turn in monthly activity reports, which are edited by your team chief, then re-edited by assistant division chief, then re-re-edited by the organizational grammer queen and the division chief himself...which is all billed to the customer in man-hours).
I personally watched SAIC bury itself into the ground with a customer and then over five years....the whole core of support with the customer just rotted away. The monthly accomplishment report....which consumed well over $12k monthly to produce for sixty personnel...was never read by the customer. The customer eventually lost respect for SAIC and their management concepts.
I won't miss dumping on the FBI in this case...because they should have had montly reviews and knew that things weren't working right. Probably none of the FBI management people involved in this episode will be demoted or disciplined...even though they share alot of the pain in this episode.
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.