That's all good and fine if you're actually a contractor and your daily travel involves hauling around lots of tools and large hardware. Nobody's going to say people shouldn't drive trucks if they don't have a real use for them on a daily basis. But so many people (men, mostly) drive their oversized monster trucks to work every day to sit at a desk and code all day. And those trucks? Not a scratch on them. No, I don't think they've ever actually been used for construction purposes.

As for me, I don't need to get windows replaced every single day. If I do, I'll rent a truck for a day. Or even easier: have them delivered. If it's local it'll cost like $20, or about 0.05% of the cost of the gas I save annually by not driving a goddamn monster truck to work.

Right--there's no problem with taking a mathematical model and tweaking the parameters until it fits experimental results.

But when you do that, it means your model is dependent on some basic assumptions. In the case of string theories, a well-known parameter is the number of spacial dimensions. If a model happens to fit experimental evidence very well, but only if there are 11 spacial dimensions, that's all well and good. But that model is useless if the assumptions are false.

We might still be able to design experiments to discover whether extra dimensions exist. Right now they're not feasible, due to the amount of energy it would take to explore those depths. But it may be possible. Of there turn out to be enough extra dimensions to fit a string theory, then we can put a check next to that parameter saying that it's consistent with reality. And that's a mark in favor of the theory.

It seems packets are being dropped by AT&T routers. A few other sites seem to be unreachable too. ZDnet has the scoop: http://blogs.zdnet.com/BTL/?p=18064