Less Wrong is a community blog devoted to refining the art of human rationality. Please visit our About page for more information.

btrettel comments on Open thread, Nov. 7 - Nov. 13, 2016 - Less Wrong Discussion

5 Post author: MrMind 07 November 2016 08:01AM

You are viewing a comment permalink. View the original post to see all comments and the full post content.

Comments (123)

You are viewing a single comment's thread. Show more comments above.

Comment author: JohnReese 08 November 2016 09:41:53PM 2 points [-]

Any LWers attending Envision 2016?

Comment author: btrettel 09 November 2016 06:17:18PM *  -2 points [-]

I would have liked to, but they rejected my application. (Edit: I imagine I'm getting downvoted because I mentioned this. Note that this is not complaining. I tend to view application processes as close to lotteries, so I don't take this personally.)

The topics of the conference interest me greatly. Right now I'm planning on hosting some futurist related discussions through the Austin LessWrong group while coordinating with a local futurist group.

If you will attend and have the opportunity, I'd be interested in seeing a summary of your experiences at the conference.

Comment author: btrettel 10 October 2017 01:47:57AM 0 points [-]

Amusingly enough, I got into the conference this year early. This seems to be a small piece of evidence for my hypothesis that these sorts of applications often work as lotteries.

Comment author: JohnReese 10 November 2016 12:28:04AM 0 points [-]

Hiya!, Oh ok. Sorry to hear that. I will be attending - yes. That's a great idea, let me post a summary of what it was like and what I learnt from it in Dec. Thanks for the suggestion. All the best with your futurist group. Any themes you find interesting in particular?

Comment author: btrettel 11 November 2016 04:03:16AM *  -2 points [-]

I'm interested in predicting future events to prioritize technology research. I've been thinking about getting speakers with expertise in the future of computing, trends of resource availability and utilization, climate change, and clean energy to start.

Previously I thought futurism was all about making optimistic predictions, but since then I've found more futurists who make predictions I think are credible. I track my own on PredictionBook and am going to start using Metaculus and GJOpen soon.

And despite working on a PhD in engineering, I'm actually quite skeptical of technology. Technological solutions to problems don't have as good a track record as most believe, and I think this influences where I should focus my research. For example, I used to think clean energy research was very important, but I've since come to the conclusion that energy and climate change are social/economic, political, or even aesthetic problems, not so much technical problems. That's not to say technology won't play a role, but due to things like the Jevons paradox, technology's role isn't as obvious as people think it is. Improving efficiency can increase usage, not that even maximally efficient systems will necessarily solve the problem. You should get some idea of what the actual effect of the research will be rather than assuming the effect will be what you want. Similar things are frequently said about starting a business: Check if the market exists before starting the business.

Comment author: morganism 12 November 2016 09:20:57PM 0 points [-]

I like the way biological systems use waste streams from upstream to produce their own fuel.

There are some good projects to condense CO2 directly from the atmo , and convert to methanol and hydrogen. If placed next to manufacturing fuel cells and pipelines, these become economic leverages, and others will quickly implement them to pick up cost savings.

A lot is also regulatory, like electric co's restrictions against re-using the waste heat from processing ,

Comment author: btrettel 12 November 2016 11:28:32PM *  -2 points [-]

Good point about regulatory issues. I've been thinking a lot about working on standards committees and whatnot as they actually have influence and many standards/regulations/codes are bad.

Using waste streams is one of the more basic efficiency engineering approaches, and at this point I think if large gains were to be had from those, we'd have them already.

As for condensing CO2, there are tons of ideas along those lines, but I'm not sure carbon capture is worthwhile. I'd need to see more economic analysis of those ideas, or better yet, test implementations. That's more or less my point. There are a ton of ideas, many of which could work technologically, but which would work economically/socially/etc. as well?

Let's go back to biological systems. Even assuming that most people driving is a good idea (I don't assume this), cars are somewhat irrational for that purpose. You can cut down drag (and consequently increase efficiency) a lot with relatively basic (and well known) modifications, e.g., boat tailing. It seems to me that cars aren't built that way in the first place because even though people say they want fuel economy, etc., cars built that way won't sell.

There used to be a really interesting interview along those lines with Bob Lutz (a well known car company executive) here, but it seems to have since gone offline. Here's what I have quoted in my notes:

AlixPartners: I would love to hear your point of view on design. Is it becoming even more important or not?

Bob Lutz: If you look at the [auto] companies that are really successful today, they are heavily design-focused.

In an era of levelness in almost everything else—fuel consumption, safety (which is all mandated anyway), cost, et cetera, et cetera, et cetera—the one thing that sticks out that can give you a huge competitive advantage is design.

AlixPartners: There's always a trade-off between the design side and the engineer. What are some of the key lessons you might want to pass on about that give and take?

Bob Lutz: Well, my key lesson learned, and I pass on to anybody who is in any position of responsibility in the automobile business, is if you look at the automobile as a collection of rational traits, like fuel economy, shoulder room, elbow room, hip room, rear H-point to dash, et cetera, et cetera, et cetera--and you get all that down on paper and the car is totally defined dimensionally, and then you hand it to the designers and say, “Put a wrapper on this, please,” you're going to get a car that meets every rational, stated desire on the part of a potential customer—but nobody's going to buy it! Because it is a fundamental mistake to look at cars and their attributes in a rational way.

We're all rational people. but looking at cars as fulfilling rational needs and then designing to that is about as smart as designing men's wristwatches for function only. They're just not going to sell.

And it's the same with cars. Tumblehome, side sloping in, fast windshield, roof height—[it was usually a] struggle to get what the designers wanted, in terms of not having a very stiff-looking car. And [the engineers would often] say, “Well, but what you're doing is it's going to deteriorate head swing lateral [if we] go down another half inch in roof height.”

And I would say, “Have you ever seen people in a showroom with tape measures, where the husband and wife are in there measuring and finally one of them says, ‘You know honey, this Chevrolet has a half inch less [room] than the Toyota Camry we saw before. Let's go buy it.’”

That's just not the way people behave. People won't even be in the showroom unless the car fascinates them visually from ads or as seen on television or seen in the street, or whatever.