Russell's Blog

I will admit, YouTube is cool. You can upload video you made yourself. Fantastic. Great. Wonderful. But "vlogging"? That's just stupid.

Blogs are great because anyone who has...

• an idea
• a computer
• internet access
• modest command of a written language

...can start sharing their ideas with the rest of the world (at least, the rest of the world that has a computer). That's actually rather revolutionary. It's Many-to-Many communication on a scale never seen before.

But do you really want to see, you know, the masses? If we start with the quality of the average written blog entry, and we use that as a benchmark to predict the average quality of a vlog entry (were the practice to contaminate the general population), you get something rather like your uncle Ed's wedding video, except it's just a video of him trying to tell you something.

Please, for god's sake, spare us from this.

After more than a year of work, I've finally finished the calculations and coding for my paper on high-performance tokamak equilibria (working title: Comparison of Analytic and Numerical High-Beta Equilibrium Solutions). This is my first piece of truly original research. This is a little snapshot of the error analysis from my first parallel run:

The code is written in Python, with the bits that need to go fast (numerical integration and root finding) compiled and optimized for the local CPU flavor. The optimized functions were supplied by the wonderful SciPy project. Given how much work I put into this project, I probably could have done it without SciPy, but I wouldn't have wanted to. The end result would have been much less reliable, slower, and more annoying to use. I am also extremely grateful for the Debian project and its package maintainers, who deliver the only OS I can find that has all the necessary packages compiled and ready to use.

Now, I just have to prepare the cases, run them, and write the "Conclusions" and "Error Analysis" section. That, and cross my fingers and hope that I don't get any grumpy reviewers.

I have my first distributed parallel program up and running, and even (mostly) debugged! Python's XML-RPC made very short work of it.

This is the last step to finishing my research for the infamous High-Beta Paper.

I got to meet Al Gore at Borders today, and picked up a copy of the book for An Inconvenient Truth. I know it's been said many times before, but I really am astonished at how different he is in person than he was during the 2000 presidential race. He was funny, casual, and completely at ease, even as the requisite Westwood Insane Homeless Person screamed nonsense at him from the first floor. He signed a huge number of books in a very short period of time, and managed to do it without seeming to hurry anyone. He didn't rush around and scowl like I've seen authors at book signings do. He went through the two hour ordeal giving off the impression of being busy, but happy to listen to people if they had something to say. Several people in the line went on at length about what they were going to ask him "when their turn came," but when they got to the table, almost no one pestered him.

The other thing I didn't realize, even from his performance in his movie, is how bear-like he is. He's about my height, but if I were drawn in pencil, he'd have to be drawn with a Sharpie.

Anyway, I also picked up a bunch of Terry Pratchett books. Mimi's family and Pierre gave me a Borders gift cards for my birthday, and good use is being made of them.

While I'm happy to have a signed copy of his book, I'd much rather see his name on a ballot.

So, pretty much every race I cared about went the wrong way in yesterday's election. Wesley lost the nomination for governor, Francine Busby lost her gambit to take CA-50, and both ballot initiatives failed. I guess that's what happens when only 28% of registered voters show up. Lazy fucks. Maybe they should give out sometime more exciting that those little "I Voted" stickers. Maybe T-shirts, or mugs or something.