Slashdot Mirror
US Virtual Border Fence Doesn't Work
lelitsch writes "The Washington Post reports that the initial pilot of the Virtual Border Fence planned by the DHS and subcontracted to Boeing has been a miserable failure. A lot of the points in the report have the hallmark of death-march software development projects. Some choice quotes include 'did not work as planned or meet the needs of the U.S. Border Patrol,' 'DHS officials do not yet know the type of terrain where the fencing is to be constructed,' and 'the design will not be used as the basis for future... development.' The article notes that Boeing was forced to deliver 'something' early as President Bush pushed for immigration reform in Congress in 2006. That reform effort died last year in the Senate."
But how will we stop all those virtual Mexicans now?
"You'll get nothing, and you'll like it!"
Outsourcing the software development to Mexico was a terrible idea.
If we just annexed Mexico we'd only have to build half as much fence to keep the Guatamalans and Hondurans out. Plus, they have margaritas.
Do what you can, with what you have, where you are.
The Six Million Peso Man
Someone needs to do all those shitty jobs that your average Second Life citizen thinks they're too good for.
Relevant quote: If you're so woefully underinformed, just keep from commenting, ok?
-- Language is a virus from outer space.
It amazes me that the Mexican president encourages illegal immigration into this country and calls those of us who want immigration laws to be followed racist or anti-Latino. All the while, they have stringent immigration laws for those coming to Mexico and are trying to build a fence with Guatemala.
The chutzpah is unbelievable.
Except for ending slavery, the Nazis, communism, & securing American independence, war has never solved anything.
One reason all those crap jobs have such lousy wages is that employers know they can always hire illegals, who are in no position to complain about wages or working conditions. I'd rather pay more and see an American citizen get the job.
Mea navis aericumbens anguillis abundat
...a wide area surveillance group, I would like to suggest a few reasons why this occurred, especially given what we know of Boeing's attempt to provide a solution.
;). There's a vast amount of 'learning through pain' which (of course) teaches you how to avoid stumbling blocks in the future. WAS is a fusion of architectural planning, mechanical engineering, network engineering, environmental engineering, and software engineering. It is also one of the more difficult management projects due to the fact that very few companies (almost none) have the in-house departments/divisions to handle all aspects of it; ergo, most companies do the more natural 'I am the lead contractor, you all can sub-contract to me for utilities, HVAC, network topology, integration software, camera systems, electromagnetic fences', et cetera.
;)
Wide Area Surveillance is, like any real world 'enterprise' solution, complex. That is not to say it is not achievable, it is just not something you decide to do on a whim
This means that during the bidding process for these jobs, as with any $$$LARGE$$$ government contract, much of the sub-contracting can be political and very rarely results in a proffered solution that is 'best in breed' in all (or even most) areas.
This is all very normal. The real difficulty is in identifying which aspects of a WAS solution will kill your project. For example, the article claims that using off the shelf commercial software for dispatchers was a serious issue. I can tell you from experience, there's no way that this derailed the project. There are several companies (the one I used to work at is one for example) that specialize in integrating their 'command & control' (for lack of a more encompassing term) suites with 3rd party streaming video, network systems, hardware devices, et cetera. The relative cost of these systems varies from very low (with a fair amount of services work being entailed) to moderate (where you get far more C&C stuff than you plan to use but it's there if you need it in the future - but they fully integrate the things you do need off the bat.)
Usually the biggest problems are from poor planning at the start or 'mid course correction' by people who didn't make careful consideration of their options up front regarding the physical infrastructure required. A good example of this is 'pole placement.' One of the easiest, conceptually, methods of watching swathe of territory where there isn't supposed to be much activity is to use a high quality camera mounted (usually mounted on a Pan/Tilt/Zoom gimbal) on a tall pole. How tall? THAT is the question my friends. From a cost point of view you want to put them up as high as is feasible given the terrain and what the local survey should be. This means less poles, less cameras, and less overall costs to cover a wider area; HOWEVER, the higher you put that camera the more difficult the installation of the pole because I assure you that putting a camera 60 feet off the ground results in shaking, deflection, twisting, and all kinds of other frame stabilization nightmares. Usually what happens is that the project denotes the max camera heights, assigns what types of poles/towers will support the cameras, how they will be built in order to overcome problems like these and then 6 months later they change the camera heights (usually because they want to cut out a few poles and the neighboring cameras must take up the slack), bingo you're well thought out and budgeted pole no longer serves your needs.
It is at this point that the reader will think 'ok, then we need to redesign the poles right? No big deal...' Sadly this does not usually happen. The change request costs associated outweight the money saved on the pole changes but that doesn't mean they won't still use the wrong poles and save a hundred thousand on camera costs, they'll just try to hack some solution like putting a frame stabilizer black box on the back of the camera, because that should work, right?
Loading...
I'd rather pay more and see an American citizen get the job.
No, you wouldn't.
Ideology is easy when it doesn't hurt you (or, in this case, your pocket).
Just build the Great Walmart of America. One side is the employee entrance, the other side for customer.
Life is not for the lazy.
I work for DHS and a friend of mine runs a small program that's been managing sensors on the border for 25 years.
Boeing was hired as the project's integrator and instead of subcontracting or working with the existing systems tried to do everything themselves. Why? To keep as much money for themselves, of course. They ignored, at first, all the existing systems and tried to replace them with proprietary technology that would anchor them into govermnent contracts in perpetuity.
They failed. Now they have to rely on refined data from a government-developed system to produce any results at all. This is a pattern I've seen in 26 years of working for the government: we hire an outside vendor who comes in and has to rely on our knowledge to make anything work. In a lot of cases they get us to do much of their work for them. The vendor's employees get huge bonuses and we get downsized. Granted there are times where if you don't bring in someone from the outside nothing will change, but the number of times internal staff saves the vendor's ass has been, in my experience, much higher than the other way around.
Sometimes it's better to spend your money on what your own staff can do instead of just assuming that an outside vendor will automatically develop something better. For some reason, too many executives undervalue the abilities of their own people and hire big names like Boeing for many times what it would have cost to develop better systems in house. The Secure Border Initiative is apparently one of them.
TLR
A man no more knows his destiny than a tea leaf knows the history of the East India Company
It was set up to fail from the very beginning. Its no secret that every power in the American government wants more illegals. The republicans want more cheap labor. The democrats want more poor voters. This fence was never more than distraction. Just a way for government to pretend they are doing something, while actually doing nothing.
It can be go tiem now plees?
Forcing someone to deliver a proof-of-concept or sample product for an arbitrary date is just common business practice. In this case, if Boeing wanted to continue the contract with the US Government they needed to prove they were up to the challenge. The arbitrary date was set by Bush because of his agenda. Just last week our company had to build a demo sandbox for a potential customer to play around in. We had a restrictive timeline in which to build it because the customer sponsor had a deadline to receive funding. We had to deliver something mostly untested and with deprecated hardware. The only difference is we had done it before so was able to deliver quickly. In my mind it's just finger-pointing if a vendor agrees to a certain date and then can't deliver. Just like our company should be willing to accept the consequences if our potential customer comes back with complaints that the system we gave them is too slow or buggy and "doesn't work."
And the Bush administration once again puts politics above effective governance and management.
And a slashdot user once again trots out their favorite villain without actually using their damn head.
So, you're saying that Boeing told DHS that this would not work in its first prototype/deployment? They were under orders to deploy something they knew would not work? Or is it possible that the procurement people said, "We need something that can do X, can you provide that on this timetable?" And the vendor said they could, and that it would work. Is your position that the president looked over their proposal, saw the technical flaws and systems integrations problems with the laptops and software, and said, "no one will notice, do it anyway," or that perhaps it's not the executive branch's leadership job to know when a vendor is lying about the compatibility of the components they're stitching together? Why aren't you complaining about Boeing, for lying about their ability to actually do this, and agreeing to take the contract?
Don't disappoint your bird dog. Go to the range.