Atomspheric CO2 (PPM)



Uptime verified by

24 February 2006

White supremacists are stupid

I met Jo's friend Ajak today. He was super cool and way clever. We will start a bank together.

Which brings me to my main point, being that white supremacists are stupid. And here is my proof. Either white supremacists have never met Ajak, or they have met him. If they haven't met him, they are stupid because they're making huge claims about black people without really knowing enough black people to know much about anything at all. Which is obviously stupid. And if they have met him, then they would know that black people can't possibly be less clever than white people. Unless they're stupid.

So my theory is irrefutably proven whichever way you look at it.

SQL Views

I've used some SQL views for my research database. They're pretty damn good. I've even got nested views. So I can dump the file parsing output straight into various tables, and get out a nice data set with no duplicates and the attributes attached to the records. And stuff. It's good.

Them costs going up alright

I've read some stuff lately about the rising cost of health care. As I understand it, health care has been getting more and more expensive for quite a long time now. It's share of GDP increases by a good chunk each year. Governments have been troubled by this, and one of their solutions is to introduce private health care to reduce costs. The beauty of this solution is that even if costs per person aren't reduced, it will kind of look like it's worked because a whole lot of poor people will no longer be able to afford to get some sorts of health care. So the share of GDP dedicated to health care drops, government expenditure drops, taxes drop, the rich get richer, the poor get a little richer and a lot sicker.

It's very hard to see if costs are dropping or not. So it's very hard to see if health care is good value. It's changing so fast that you can't compare costs across time. Just like with inflation - when the things people are buying improve dramatically in a short space of time it's basically impossible to even measure inflation. Maybe the government is actually doing a bang up job of health care investment and expenditure. The rising share of GDP is a complete red herring. Who cares if it goes up? Who cares if it goes to 50%? What else are we going to spend it on? If 10% of our GDP (or something like it) is keeping us fed and housed who really cares what we spend the 90% on? If we spent a whole bunch of money keeping people alive until they're 150, then they'll probably stop wanting to run off to war and die at 20. So defence spending can drop.

Life expectancy in the US is expected to drop in 2006 for the first time. If life is valuable then that suggests to me that it wouldn't be so terrible if the US spent a little more on health. A lot of economists have done research suggesting that private health insurance is unfair to the poor and reduces overall health outcomes. And it may not even be that efficient. The default response to these economists usually involves pointing at the rising complexity of health and rising costs of health. I personally think that government is done a marvellous job of something that was always incredibly complex.

I think my basic point is that there's absolutely no reason why health care should be a constant share of GDP. There's no magic number somewhere between 10% and 20% that represents the optimal share for all eternity. And there's no reason why now is the time for the government to step aside and let the market rush in to fix everything up.

I'm obviously biased, because I'm opposed to private health care entirely on equity grounds. But I can see a good argument for efficiency at the expense of equity when I see it. And right now, if you ask me, no one is making one.

A far better solution to rising inefficiency in a market with loopy elasticities (like health and medicine)

23 February 2006


I'm getting a new phone. My old phone is only a year old and is already borked. And it only fell of the roof of a moving vehicle once! Don't make 'em like they used to.

It is a Motorola. I've heard they have bad user interfaces. I figure they can't be much worse than the current Nokias.

Motorola A1000

Update: It seems I was wrong. The user interface can be much worse. Just appalling.

Killer Dog

There is a dog that lives not far from us who we see whenever we go to St Peter's station. He's a friendly, sad-looking dog. He always gets happy whenever people walk past, but goes back to his bed when they don't stop to pat him.

Libby and I have wanted to pat him for a long time, but we were always too scared. He's a big dog, and you get the sense that if you caught him on a bad day he'd be able to snap your arm without even breaking a pant.

But today Libby told me I should pat him when I walked back home after dropping her off at the station. So I did. And he was a great dog. So lovely and generous. He got a bit riled when a skate-boarder skooted past, but who can blame him.

I read his collar too, looking for his name. It said Today we kill, Tomorrow we die. I can only assume it was an ironic statement by the dog, like the guy in Full Metal Jacket. Perhaps he was wanting to say a little something about the duality of dog. Sir.

22 February 2006


I started learning Ruby the other day. I'm scripting one of the things for my uni project in it. It's so easy to learn. And exceptions are way tasty.

Everyone raves about Rails, but I'm a much bigger fan of Ruby than of Rails. Rails strikes me as very bloated and restrictive. Maybe I'll still become a convert though.


I saw one of the guys in the office using ratpoison the other day, and I thought I'd try it because my laptop doesn't have much oompf. Now I really like it just because it's cool. You can set up all these key bindings to do whatever you want. You'd save so much time if you were doing the same sorts of things regularly.

It even has a key-driven little menu dooby, if you can't remember the keys for things. It takes a bit of getting used to, but it's really nifty.

15 February 2006

Assister of Research

I just got another job. It's kind of my third job now, although one of my jobs will finish in a few weeks. I'm going to be a research assistant for one of the professors at my university. He's a pretty damn clever professor, and it makes me pretty frightened to be working for him. I didn't a project for him over the summer, but that was OK because I didn't have to talk to him, or convince him I understood what I was doing. I kind of did understand, but I couldn't have convinced anyone that I did, least of all him. So this new job is scarier still. I think I mostly have to learn to use this statistical software and explain to him how it works when he doesn't know. Mum just pointed out, that if I try and focus all my energy learning one simple thing that he doesn't understand, then I can trick him into thinking I'm clever. I reckon it's a tops idea, so that's what I'm going to do.


Occam's Razor annoys me a lot. Or not the razor itself, but the way people use it so willy nilly. I was sure there must be some clever logical reasoning behind it that explained why people used it to justify so much of their thinking. But there's no logic behind it - just the idea that simple explanations are more likely than others. Which is entirely sensible, but that shouldn't give you any extra confidence in one explanation just because it's simple.

0.276 seconds