SVG

By anders pearson 08 Dec 2000

i’ve been playing around with SVG lately.

Scalable Vector Graphics are an image format similar to Flash but implemented entirely as XML. Adobe has a browser plugin available and new versions of Illustrator (as well as the new Flash generator supposedly) can save as SVG. now, the Apache project is working on Batik which aims to be a nice core library for dealing with SVGs and mozilla has (currently very limited) native support for SVGs.

i’m very excited.

Flash is nice to look at and all but has serious limitations for web use (it’s pretty much impossible to link to anything in an all flash site from outside, text-browsers and screen readers don’t stand a chance of being able to render anything remotely useful, etc) and it’s fully proprietary (eg, i can view flash on linux but i can’t make flash files on linux because macromedia hasn’t bothered porting generator).

SVG gets around all of these problems. first of all, it’s a W3C standard so there are no licensing issues preventing anyone and everyone from making viewers/generators and incorporating them into other programs (Free or otherwise). since SVGs are just XML (ie, plain text), anyone with a text editor can make them or tweak them by hand (trust me, i’ve been doing this all afternoon. there is something very cool about the idea of making graphics in emacs :) or write programs to generate them dynamically. it also means that you can simply apply different style-sheets to them to do things like just pulling out any textual data for the benefit of text-browsers and screen readers. and since any XML file can be parsed into a DOM, you can script it to your heart’s content. so currently SVG has javascript support that lets you do all the nice interactivity and animation stuff of Flash but i expect to see support for other scripting languages as soon as it starts to catch on.

now we just sit back and wait to see how microsoft is going to come screw it all up (there’s no way they’ll let any decent, open, standards based technology go unmolested for long).

you have no free will.

By jp 04 Dec 2000

I was reading wired’s article on modern day supercomputers, and I got me brain to thinking about modeling. so given atomic cordinates, and the exact solute conditions and temps, you can effectivel model the behavior of organic and inorganic compounds to about 100% accuracy given the right data. and this with no more than a first-gen pentuim or eqivilent. so in the course of studying for my final, I get to wondering where it will end when the chips get big enough.

example. a virus has no free will. there are no genes to respond to envrionmental cues until after host infect. so all that cool syringe action DNA injection you see is purely thermodynamic, just as reliable as dropping a rock above your foot leading to broken toes. it just happens as soon as it binds a bacteria without requiring any energy. so we can model, given the exact coordinates, concentrations of viriods and substrates (bacteria) the behavior of the viruses. you’d need a slightly bigger computer.

now. move on to prokaryotes. a bacteria is technically an input-response machine. they “sense” the environment around them and respond through chemical cascades to response to the environment in a seeminly intellingent manner. penicillin got you down? no sweat, just make some drug-pumps to zap it out. it seems intelligent to us cause it’s what we’d do, given the choice. but in evolutionary terms, it was just the smart thing to do, so it stuck. but this raises the question, can we model something like an E. coli or a yeast? i think so. we know everything that would happen. it’s just a spankload more atoms to keep track of. need a slightly bigger computer.

so then what about us? given the world’s biggest computer from dimension X, shouldn’t we be able to effectively model humans? if we track every interaction they have with every single environmental stimulus (heat, cold, how much, how long) every toxin and ligand (EtOH, cholera toxin, influenza, etc) and every social interaction, and we know the exact celluar, developmental and nuerochemical cascades they bring about, can’t you predict this model’s behavior with some degree of accuracy?

so the question arises, if we can model this sort of system, does that mean we’ve created an AI, or that we’ve crushed the notion that people have free will? I kinda like the idea that we’re just overly complex stim-response machines. given the media you’re grown in, pardon the pun, it seems reasonable that we should know how any given person “thinks” on the atomic/molecular level, and therefore acts.

or at least it’d make me feel better when I do things and don’t know why I did them.

buckbuckbuck

By anders pearson 02 Dec 2000

i haven’t been to a McDonald’s in many many years so you could probably guess (correctly) that i don’t have anything particularly nice to say about them. however, i can only laugh when someone finds a chicken head in their wings and gets all grossed out and offended.

in other parts of the world, it is considered good form to serve you your chicken and fish along with the head to prove that it was freshly killed. however, someone should have warned the chinese that if you give a bunch of drunk american college students some chicken and fish heads, modern art is a likely outcome. tuck also has some fascinating footage of the entertainment possibilities inherent in chickenhead soup.

(my apologies for the poor quality of the chicken/fish sculpture. i scanned it in a few years ago when i didn’t really have access to a decent scanner.)