Dalke Scientific Software: More science. Less time. Products
[ previous | next ]     /home/writings/diary/archive/2004/04/14/SDI_spiral

`Star Wars' and the Spiral Development model

I heard on the news this afternoon that the US Strategic Defense Initiative project (better known as `Star Wars') is using a spiral development model taken from software development projects instead of a more traditional one. This allows them to start development before the whole system is complete.

I use the spiral development model a lot. It's one of McConnell's "Best Practices" in his book Rapid Development. One of the key points is to have many deployables, to test and get rapid feedback for the next loop through the spiral.

The SDI spiral doesn't have that essential feature. There have been only a fraction of the tests done as were planned for this time, and those tests were not against a realistic target. Yet the project is already building a launch site in Alaska as well as a radar system for it. Why there? Because the only close-to-realistic threat is North Korea and China.

A much more likely scenario is to deliver any such nuke via container ship since 1) it's a whole lot cheaper than developing an ICBM, and 2) less than 5% of the shipping into this country is inspected. That ratio could go way up with some of the $10 billion requested for next year, much less the $130 billion wasted so far on a weapons system which has been 21 year in development.

This new installation they're building up north? They had to get a waiver before doing it! See, most projects have to show that they'll be sucessful before being deployed. But not this one. Go ahead and build it 'cause the schedule says we need to have this highly experimental project done by this time. Even it there's a remote chance of it actually working "at least it's better than nothing."

Except that isn't the only way to spend the money. $120 billion + $10 billion/year can make for a very good espionage network and a lot of bribes to find out just who is doing what. And fund a lot of cultural exchanges, study programs, and other forms of soft power to make people less likely to want to harm us. But this administration especially seems to think only hard power counts and amoung other things has made it harder to get a scientific or cultural visa to enter the US, and of course started treating all them furriners as criminals and now photographs and fingerprints (!) most every non-citizen crossing the border.

What a message we send to the rest of the world. I will recommend that we align next year's BOSC with a European conference rather than ISMB, which will likely be in the US next year. Two years ago it was in Canada. We almost didn't get the South African contingent because they didn't realize they needed a US visa to transit a US airport from one international flight to another. Luckily, the US had the TWOV program (Travelers WithOut Visa) which meant they could make it. But that was canceled last year.

Why is all this happening? Because US citizens are frighted of the next attack? No, it's because the people in charge don't want to be blamed for the next failure. The risk analysis isn't done based on how it affects America, it's on done with a personal calculus. "If I implement nasty, horrible, un-American policy X I can justify it on being the 'post-9/11 era.' If I don't do it and something happens people will blame me and I'll loose my job. If I do it and something happens I just say we didn't get enough funding and blame someone else." The only career risk then is to stand up for trust, respect, and dignity. And when did that pay?

Some people might argue that the money spent on SDI will have less tangible benefits, like the spin-offs of the space program. My response is that the National Science Foundation requested a budget of about $5.7 billion for next year, or just over half spent on SDI. Surely tripling the NSF budget will lead to a few more spin-offs than SDI.

As you might tell, I'm rather ticked off about this. It didn't help that I listened to Bush's speech which repeated yet again the canard that chemical weapons are "Weapons of Mass Destruction." If that's the case then the chemical weapons of World War I should each be comparable to a nuke in effectiveness. It's estimated that about 1.1 million people were casualties from gas attacks in that war with about 130,000 tons of gas used. That's over 100 kilos of gas per casualty.

The referenced page says that by the end of the war gas was pretty useless because of advances in protection against gas. If it takes so much effort to use gas in warfare than how was it so effective against Iraqi citizens? (In the late 1980s, btw, when Iraq was our ally against Iran - Operation Staunch - and well before the First Gulf War.) Because this was a Kurdish town already under attack by the Iraqi army, defenseless against any sort of bombing (and they were being attacked by conventional bombs as well) and ill-prepared. They went into bomb shelters not realizing gas is heavier than air, making a death trap. (Hence why people figured gas would be useful in the trenches of WWI.)

Don't get me wrong. Chemical gas attacks are nasty and their effects and linger long beyond the war even into the children of those attacked. That's the main reason they've been banned for almost a century. But that doesn't make it a weapon of mass destruction! You could talk about the potential for mass destruction, in which case plastic bottles filled with gasoline and lit by one person can kill 198.



Copyright © 2001-2003 Dalke Scientific Software, LLC.