We Paid What for That Virtual Border Fence?

Well, they've done it again. They (those who govern us) have figured out how to design a mouse the size of an elephant. Here, read this first:

U.S. scraps $20 million prototype of virtual fence

TUCSON, Ariz. (AP) -- The government is scrapping a $20 million prototype of its highly touted "virtual fence" on the Arizona-Mexico border because the system is failing to adequately alert border patrol agents to illegal crossings, officials said.

The move comes just two months after Homeland Security Secretary Michael Chertoff announced his approval of the fence built by the Boeing Co. The fence consists of nine electronic surveillance towers along a 28-mile section of border southwest of Tucson.

Boeing is to replace the so-called Project 28 prototype with a series of towers equipped with communications systems, new cameras and new radar capability, officials said.

Now I am not an engineer or a systems administrator nor do I play on TV. I'm just a guy who purchased the first PC sold to the public back in the '70s. (Tandy TRS80, 9 Mghz, 16K, two 5" external drives and a dot matrix printer that sounded like a machine gun.)

But as I read that story, I said to myself, "Don't these people ever choose cheap generic solutions to these kinds of problems? Why is their first choice always to, not only reinvent the wheel, but do so in the most expensive and risky ways possible.? Which almost always results in cost over-runs and failure."

If someone asked me to design a virtual border fence I'd begin by finding a top-of-the line Web cam, one that can be operated (moved and focused) remotely over the web able to see in both daylight and total darkness. Then I'd find the best ruggedized laptop on the market, design a weather-proof, secure box for it.

So, let's see how much under the $20 million (failed system) they choose, I can come:

Length of virtual fence: the same 28 miles.

• Number of poles: 60 (Since Web cams tend to have lower resolution than high-end cameras, I will place them at half-mile intervals rather than one-mile intervals. (Example of a panning Web cam -- Legal Sea Foods Boston Harbor.)
• Number of Web Cams: 60
• Number of Laptops: 60

How am I doing? I've now got a fence the same length as they tried, and twice the number of cameras.

Let's see how the costs rack up:

Cameras: I have chosen a top-of-the line RISYS IRI-1011 camera/thermal imager, or equivalent. Cost: web-capable cameras like this, with good lenses and nightvision, retail for around $2900. But I figure if we buy 60 at one crack, they'll give us a break. So I will log $2,500 each x 60 = $150,000.

Laptops: I'm going expensive here. I could try individual Wi-Fi networked Web cams, but I am afraid half a mile is just to far between cameras for that to be reliable. So I have chosen to hook each camera to its own dedicated laptop. The computer I chose for this is Getac A790 Rugged Laptop, retail price, $4985. Let's call it $5,000 each x 60 = $300,000.

Lockboxes: 60 weather proof, secure lock boxes plus installation: $250 each x 60 = $15,000

Poles: 60, 40 foot metal poles: fabrication, shipping and installation: $4,000 each x 60 = $240,000

Desktop PCs: Even though I assume Homeland Security has warehouses full of PCs just sitting around collecting dust, nevertheless I will throw in some desktop computers to monitor the cameras. One desktop for each 2 Web cams: 30 x $700 = $21,000.

Connectivity: High-speed T1 line connecting the laptops/cams to the Net: $1,000 a month. (Maybe the government can get those friendly phone companies that were so anxious to help them spy on us to run this line for them too.)

Total Cost of My Virtual Fence: $735,000