Reviewing TED

I really enjoyed Tom Wujec's earlier talk on the astrolabe, so I came into this one with high hopes. Wujec's summary of cognitive science leaves a good bit to be desired, however. The visuals are pretty good (though the visualization of 2008's TED horrifies me just a teensy bit), but his interpretation of the science doesn't really match up with what I've learned over the past decade – or at least, it only matches up in a very elementary (and relatively uninteresting) way.

Wow, almost forgot today. So much for that 30-days-to-ingrain-a-habit business, I guess.

Arthur Benjamin strikes me as the best sort of math professor – he's engaged and passionate about his topic, and he's just a little geeky. He's also got the advantage of supporting a trendy (though no less important) position: that statistics is more important to most people than calculus is.

The one other thing that stands out for me in this video is that Benjamin talks almost entirely with his hands.

An old one, this time - Jill Sobule's song for Al Gore is a cute ode, but it's another one of those that I just don't see as being TEDish. Actually, I preferred Sobule's joint performance with Julia Sweeney the following year.

The Long Now Foundation does some fascinating work, but I kind of think the Long News project (as presented by Kirck Citron) is fundamentally mistaken. The general idea is that of all the news stories reported in a given year, only some will be important 10, 50, or 100 years in the future. I see two problems with this:

1) This seems like a slippery slope; Citron appears to be advocating an approach that eliminates the personal aspect of news in favor of the long view. Frankly, the birth of my daughter in 2008 was one of the biggest stories for me, but the chances are pretty slim that it'll be important any time in the future to the majority of humanity (and even if she does, say, cure cancer, her birth itself still won't be the news story)

2) This is an attempt to predict the future impact of current events, which is notoriously, and inevitably, I think, unreliable. Citron can discount the death of an entertainer all he likes, but there are entirely possible futures where that has repercussions felt much more strongly in the future than this or that medical technique.

I've wanted to review this talk for a while – I saw it a while ago, but I've been repeatedly distracted. I love Chinese(-American) food, so this one's of special interest to me.

(sidenote, continuing my interest in presenter names: where does the "8" in Jennifer 8. Lee come from?)

The fortune cookie story is interesting; it's a great start to Lee's talk in that it points explicitly to the lack of interest in authentic food – we don't care that fortune cookies are actually Japanese in origin and are unknown in China, because they're part of Chinese food culture in America. So this might be controversial, but I think that's perfectly OK.

Perfectly OK, that is, as long as we don't project those attitudes back onto the supposed source culture. As Lee points out, food gets modified wherever it ends up – Italian Chinese food is different from American, Japanese, Brazilian, and every other sort of Chinese food. We need to understand that, and the power of every culture to acquire and modify the products of other cultures to fit into the local ecosystem.

This is a great talk, filled with humor and insight, and it's one more that makes me sad Taste3 appears to be defunct.

First off, "Hod Lipson" is a very cool name.

That aside, I love the opening question Lipson poses: where are the robots that we've been promised for the past forty years? (I'd also ask that about hovercars and jetpacks, by the way). I also like the conjunction of simulated evolved organisms and physical recreations of them; once you get the physical constraints encoded into your simulation, this seems like it'd be a great way to design any sort of complex machine.

On the other hand, the physical device with multiple, trainable brains seems like it's putting the cart before the horse. Certainly, the device as a whole can learn how to move forward, but it seems much less efficient to settle on the physical design before the software is written when doing it the other way 'round is possible.

Malaria kills 1 million people a year? Reused syringes kill 1.3 million a year? Holy crap. Those are the sort of numbers that remind me in no uncertain terms that I live in a privileged world.

OK, a few minutes in and I'm convinced this is one of the most terrifying TED talks I've seen.

Sometimes we joke about planned obsolescence, or about items breaking just as they come out of warranty. Marc Koska's one-time-use (enforced) syringes are that, but for the best of reasons. This is the kind of thing that will save lives.

There's a horrifying slide in the background of this talk when it starts – and upon the close-up, it's still bad. (Side note: the length of the health-care bill has been on my mind lately, what with people complaining about it being 2000+ pages. I'm sorry, how long should the rules governing a reform of such massive scope be? Is there a cheat sheet somewhere? Tax code: 1,000 pages; capital punishment: 23 pages; health care reform: 162 pages)

Alan Siegel comes from an advertising and design background, which provides an interesting point of view; his one-pager on credit card policies is nice, but I think it's a bit of a cheat – it's a very dense single page, and the interactions he mentions for the web-enabled version make it clear that not everything we might want it to capture appears on the page itself.

Siegel's pretty dry, but I think between the content and the effectiveness of his work saves the talk.

This talk follows on David Gallo's earlier talk. Interestingly, some of the deep-dive video seems more primitive the earlier visuals, but it's still just astounding.

On the shallow videos, I'm always fascinated with the visual communication abilities of octopi and squids. Forget robots and zombies – cephalopods seem much closer to being our new overlords than anything else.

Still, I'm excited that there's a huge part of our world that still allows for "classical" exploration and discovery. Deepest, darkest Africa might be wholly mapped, but the oceans of the world remain "Here Be Dragons" territory, and that's inspiring.

I'm going to watch two David Gallo talks back to back - this first one is from 1998. First off, I'm inclined to like him because he mocks Titanic (yay!). He's a very practiced, relaxed speaker, which is nice. Less compelling voices might be overwhelmed by the remarkable images onscreen, but he weaves in and out of them (as in his descriptions of the 150-foot long and X-wing death jellies).

I think the biggest lesson to take from Gallo's talk here is that the ocean was, back in 1998, the least known habitat on Earth – which, considering that the oceans are more than two-thirds the Earth's surface, means that we knew much less about our planet than we typically think we did.

Now I want to go watch SeaQuest DSV.

(Spoiler: we still don't know nearly enough about the oceans.)