Environmental Monitoring and Citizen Science

The Economist has an article on environmental monitoring. It's the citizen science idea though which makes it particularly compelling. It involves tools to help interested people in gathering information that scientists can later use - and get long term data across a wider range on the planet than any one or team of scientists could possibly get. It's also great in helping kids learn about science: proper collecting of data, investigation, watching things over time, even making hypotheses about the data that they get.

There are also interesting technical issues with how to enable easy collection, processing, storage and analysis of huge amounts of data, but that can wait until next time.

Food Production in America

Thanks to the great Xeni Jardin over at Boing Boing I ran across an interesting article by Michael Pollan about the issues our next President will face about food production in the US. Food prices have crept up steadily over the last few years, driven by everything from skyrocketing oil prices (after transportation the agricultural system uses the most oil in the country) to ethanol production from corn. The latter has been contested in a few places, but it's hard to argue with the first. Pollan says that the era of cheap and abundant food is over and that the new president will need to reform the entire system. I've noticed that my food budget has definitely doubled over the last few years and I'm no larger than I was then so something has had to increase - if it's not my girth it must be the cost of the food :)

One of the most terrifying facts in the article was this:

a system that in 1940 produced 2.3 calories of food energy for every calorie of fossil-fuel energy it used into one that now takes 10 calories of fossil-fuel energy to produce a single calorie of modern supermarket food

At any rate as Pollan said:

If I could say it in 2 pages I wouldn't have used 8000 words.

I encourage people to read the article and think a bit more about the food they buy.

Just saw the post about google chrome...

Here:

http://blogoscoped.com/google-chrome/

Other than how absolutely cool it is for Scott McCloud to be doing a comic book about a web browser, there's some very cool and interesting technology in there. The separate process scheme sounds interesting and v8 sounds quite cool as well. This on the heels of TraceMonkey this week from Mozilla. A lot of good research going into dynamic languages and compilation. The fallout from all of this competition should be great.

Dumb ideas...

Ever had one of those dumb ideas that you start thinking about and go "hey, this just might work - and not suck too much"?

Yeah. Me too. This could be entertaining. (Stack slot indeed...)

Lots of talks lately...

One on more effective use of gcc, a pair of tutorials on how to use OpenMP and tonight's effort of a brief tutorial and explanation of autotools. If anyone is interested in my slides or pdfs send me email and I'll forward on a copy. I suppose I should make them available somewhere.

Tom and Ian both made posts on them recently - amazingly good timing for me. I agree that a new solution needs to happen, most of autoconf is a pain, takes too long, and is too hard for most people to maintain. I like automake, but am also sympathetic to wanting something new there too. I did look at cmake as someone mentioned in a comment to Ian, but didn't really have time to get into it.

I've been working on some moderately interesting stuff at work, but can't really post about it yet. Hopefully soon though, not very gcc related unfortunately though.

More pointers to other people's blogs...

I seem to be doing that these days, but I just ran across this thanks to ncm and found it to be great:

Developing for Developers.

Ran across it because of the "cache-oblivious data structures" post, stayed for dessert.

Update: Fixed cnp.

guess i'm really an employee now

Got cut off by SJ on my way up to the rink today...