Russell's Blog

New. Improved. Stays crunchy in milk.

The Wrong Data

Posted by Russell on October 26, 2006 at 11:38 p.m.
I tracked down the problem that had been mucking up the unity-beta results from Ophidian. Either Pierre or I accidentally checked in a version of CUBE into the repository that wrote the output file data tables in the output in a different order. So, all this time I had thought it was a logic bug, or worse, a math error, turns out to be the result of importing the wrong table.

The weird thing is that it somehow passed my integrity tests. It was the right length, it had the right precision, it had the right boundary conditions, and it had an inflection point.

Er, it had lots of inflection points. I guess I should add a test for the entropy before I pass imported data to the rest of my code. These are all supposed to be nice, smooth, differentiable functions. I don't even use the data in this table -- I think it has something to do with benchmarking the convergence cycle. Maybe it's a poorly calculated magnetic shear. Maybe it's some sort of error distribution. All I know about it is that the values came out of a vector burried somewhere in the solver, and that it's labled "s".

That's the trouble with computers. You leave one thing out of place, and before you know it, they go stomping all over the place. Sooner or later, you will find yourself the victim of your own Techno Trousers. Anyway, here my plot from last week, but without the garbage.

No, I'm not going to explain what it means. You'll just have to wait for the paper.

Asshats

Posted by Russell on October 26, 2006 at 4:08 p.m.
I remind Rush Limbaugh apologists that Parkinson's disease is degenerative and ultimately fatal. Michal J. Fox is in the process of dying. He has decided to let America watch him die in the hopes that we might learn something.

This isn't some cheap political stunt. This is a man's life. Whatever your position on stem cell research, at least have the decency to respect Mr. Fox's effort to die a meaningful death. He is doing these commercials because he hopes, as most people do, that stem cell research may one day lead to treatments that help people suffering from Parkinson's disease (as well as other injuries and conditions).

It's too late for Mr. Fox. He's doing this to help other people.

I suppose Rush would have said Christopher Reeves was exaggerating the way in which he just lay there in his chair, or deliberately let his sentences get cut off when his respirator whistled and clicked. You know, for dramatic effect. He was an actor, after all, and worse yet, a liberal Hollywood type. He'd do anything for attention. Sadly, it seems empathy is beyond the ken of some people.

The horrible truth of the matter is that someday soon, Michal J. Fox will die. Parkinson's disease does not usually allow a photogenic death, but given his bravery thus far, it would not surprise me if Mr. Fox keeps working until the bitter end. Then Mr. Limbaugh can talk about how "shameless" it was for Mr. Fox to exaggerate the pain and misery of his own death to push his "agenda."

Like most liberal types, I support stem cell research, and I'm proud to have Marty McFly, Superman and the great state of California on my side.