Overview for gwern - Less Wrong

gwern on Rationality Quotes June 2013

2013-06-18T02:00:35.315702+00:00

And the reason this hypothesis is so unlikely as to be not worth considering is:

During the Cold War, the US and British governments were shot through with hundreds of double agents for the Soviets, to an almost ludicrous extent (eg. Kim Philby apparently almost became head of MI6 before being unmasked); and of course, due to the end of the Cold War & access to Russian archives, we now have a much better idea of everything that was going on and can claim a reasonable degree of certainty as to who was a double agent and what their activities were.

With those observations in mind: can you name a single one of those double-agents who went public as a leaker as Snowden has done?

If you can name only one or two such people, and if there were, say, hundreds of regular whistleblowers over the Cold War (which seems like a reasonable figure given all the crap like MKULTRA), then the extreme unlikelihood of the Fox hypothesis seems clear...

gwern on LessWrong help desk - free paper downloads and more

2013-06-18T00:10:38.125354+00:00

Thanks.

gwern on LessWrong help desk - free paper downloads and more

2013-06-18T00:10:30.376247+00:00

Thanks.

gwern on LessWrong help desk - free paper downloads and more

2013-06-18T00:09:02.862078+00:00

Thanks.

gwern on The Classic Literature Workshop

2013-06-18T00:03:34.855785+00:00

It felt to me like George Lucas was being given too much of a free hand, and so there wasn't anyone doing that checking

FWIW, this is pretty much the story that I got from reading (ironically, I've forgotten which) Kaminski's The Secret History of Star Wars or Marcus Hearn's The Cinema of George Lucas: Lucas, for the original trilogy, was able to circulate drafts of the movie scripts among a circle of friends and acquaintances, who would edit and make suggestions and criticize it heavily. This circle had, however, disintegrated by the time the prequels got going, between normal life events, Lucas's divorce, etc.

(As far as Empire Strikes Back goes, you can't give any of the credit to Leigh Brackett; she was dying at the time, and supposedly Lucas wound up ignoring what she managed to finish.)

gwern on Rationality Quotes May 2013

2013-06-17T13:17:37.423398+10:00

So you were using Hmisc::spower... I'm surprised that there was even such a function (however obtusely named) - why on earth isn't it in the survival library?

I was going to try to replicate that estimate, but looking at the spower documentation, it's pretty complex and I don't think I could do it without the original paper (which is more work than I want to do).

gwern on Rationality witticisms suitable for t-shirts or bumper stickers

2013-06-17T13:09:01.905022+10:00

Special Pleading Objection?

But Johnny Depp is special: he's Johnny Depp. He's an elite. And breaking (fashion) rules may be part of why he continues to be perceived as an elite (I'm thinking particularly of Kleef et al 2011 in http://lesswrong.com/lw/dtg/notes_on_the_psychology_of_power/ ).

gwern on Open Thread, June 2-15, 2013

2013-06-16T22:32:07.663289+00:00

in sans-serif than in serif font.

FWIW, most of my pages on gwern.net seem like they'd count as 'long texts', but my just concluded font A/B test using 2 sans-serif and 2 serif fonts doesn't see any difference in reading time when you split by serif: http://www.gwern.net/a-b-testing#fonts

gwern on Open Thread, April 1-15, 2013

2013-06-17T08:30:24.023524+10:00

My A/B test has finished: http://www.gwern.net/a-b-testing#fonts

Baskerville wasn't the top font in the end, but the differences between the fonts were all trivial even with an ungodly large sample size of n=142,983 (split over 4 fonts). I dunno if the NYT result is valid, but if there's any effect, I'm not seeing it in terms of how long people spend reading my website's pages.

gwern on Near-Term Risk: Killer Robots a Threat to Freedom and Democracy

2013-06-17T03:05:27.096618+10:00

I don't think it saves Rolf's point:

The actual US armed forces are a few million. 5% would be a much better estimate. This aside, you are ignoring that "lethal autonomy" is nowhere near the same thing as "operational autonomy". A Predator drone requires more people to run it - fuelling, arming, polishing the paint - than a fighter aircraft does.

If you are getting >6x more flight-hours out of a drone for <2x as many people used as compared to its alternative, then by switching a fleet of alternatives entirely to drones, the effectiveness or lethality increases by >6x for an increased man power of <2x - even if you keep the manpower constant and shrink the size of the fleet to compensate for that <2x manpower penalty, you've still got a new fleet which is somewhere around 6x more lethal. Or you could take the tradeoff even further and have an equally lethal fleet with a small fraction of the total manpower, because each drone goes so much further than its equivalent. So a drone fleet off similar lethality does have more operational autonomy!

That's why per flight hour costs matter - because ultimately, the entire point of having these airplanes is to fly them.