Written a full response to your comments on Felicifia (I'm not going to discuss this in three different venues), but...

your opponent's true rejection seems to be "cryonics does not work"

This sort of groundless speculation about my beliefs (and its subsequent upvoting success), a) in a thread where I’ve said nothing about them, b) where I’ve made no arguments to whose soundness the eventual success/failure of cryo would be at all relevant, and c) where the speculator has made remarks that demonstrate he hasn’t even read the arguments he’s dismissing (among other things a reductio ad absurdum to an ‘absurd’ conclusion which I’ve already shown I hold), does not make me more confident that the atmosphere on this site supports proper scepticism.

Ie you're projecting.

Assuming you accept the reasoning, 90% seems quite generous to me. What percentage of complex computer programmes when run for the first time exhibit behaviour the programmers hadn't anticipated? I don't have much of an idea, but my guess would be close to 100. If so, the question is how likely unexpected behaviour is to be fatal. For any programme that will eventually gain access to the world at large and quickly become AI++, that seems (again, no data to back this up - just an intuitive guess) pretty likely, perhaps almost certain.

For any parameter of human comfort (eg 253 degrees Kelvin, 60% water, 40 hour working weeks), a misplaced decimal point misplaced by seems like it would destroy the economy at best and life on earth at worst.

If Holden’s criticism is appropriate, the best response might be to look for other options rather than making a doomed effort to make FAI – for example trying to prevent the development of AI anywhere on earth, at least until we can self-improve enough to keep up with it. That might have a low probability of success, but if FAI has sufficiently low probability, it would still seem like a better bet.

Seems like a decent reply overall, but I found the fourth point very unconvincing. Holden has said 'what he knows know' - to wit that whereas the world's best experts would normally test a complicated programme by running it, isolating out what (inevitably) went wrong by examining the results it produced, rewriting it, then doing it again.

Almost no programmes are glitch free, so this is at best an optimization process and one which - as Holden pointed out - you can't do with this type of AI. If (/when) it goes wrong the first time, you don't get a second chance. Eliezer's reply doesn't seem to address this stark difference between what experts have been achieving and what SIAI is asking them to achieve.

Hm. Interesting piece. I'm partially sold, but not on this: 'Further, I see little difference between how a Muslim "chooses" to get upset at disrespect to Mohammed, and how a Westerner might "choose" to get upset if you called eir mother a whore.'

I'm pretty content to call that a sort of choice, especially if you make it a fair comparison, ie a general remark not victimising one person that all mothers are whores. After all, there’s still a pretty big difference between that (or even the rather more inflammatory ‘all Western mothers are whores’), and (a sincerely offensive) ‘your mother is a whore’. One is basically bullying someone, assuming they’re not in a position to hurt you back equally; the other is the sort of casual prejudice that (cough) some of us discourage but don’t actually seek to ban.

On top of that, there’s a significant difference between drawing a picture of someone and drawing a picture of someone in a way calculated to piss people who like them off. In the Muhammad cartoons furore, it initially seemed to be Muslims who were trying to elide the difference – specifically by positioning the latter as very bad and the first as (almost) equally bad. If drawing the former is a political action against such a sentiment (or just an aesthetic statement, standing against those who’d repress a portrayal of something they thought was beautiful), then I hardly think it’s a reprehensible one. Here I think actual 'whores' - or rather porn stars - give a better analogy. Their portrayals offend a lot of people, but few sensible people think there’s a good argument for banning them a) because overturning our anti-censorship sentiments should require a pretty strong burden of evidence and b) because a lot of people very much like them, and why should they be deprived? After all, the naysayers choose not to look at something that exists, but the fans can’t do the reverse.

Lastly, (and leastly), there’s the question of accuracy of the original criticism. If your mother does sell herself for money, then, while victimising you for it is still pretty unpleasant, we would be more inclined to tolerate borderline cases of people pointing it out in a potentially offensive way than if it weren't true. But most of the times when someone’s mum is aggressively called a whore, she probably isn’t. On the other hand, by most accounts Muhammad was a brutal sex pest, who most likely would have ordered suicide bombings had the technology existed for him to do so.

I don’t know how relevant improv is to Less Wrongers, but I find it helpful for everyday social interactions, so:

Primary recommendation: Salinsky & Frances-White’s The Improv Handbook.

Reason It’s one of the only improv books which actually suggests physical strategies for you to try out that might improve your ability rather than referring to concepts that the author has a pet phrase for that they use as a substitute for explaining what it means. Not all of the suggetions worked for me, and they’re based on primarily on anecdotal evidence (plus the selection effect of the authors having run a reasonably successful improv group in the hostile London climate and only then written a book), but I know of no other book that has as constructive an approach. It also has a number of interview sections and similar, which are eminently skippable – only half the book is really worth reading for performance advice, but fortunately the table of contents make it pretty clear which half that is.

I’m recommending it over Keith Johnstone’s ‘Impro’ and ‘Impro for Storytellers’, whose ideas it incorporates, breaks down and structures far better, over Chris Johnston’s ‘The Improvisation Game’, which is an awful mishmash of interviews and turgid academic writing, over Charna Halpern’s ‘Truth in Comedy’, which has quite a different set of ideas but spends more time boasting about how good they are than explaining them, over Jimmy Carrane and Liz Allen’s Improvising Better, which has a few nice tips and is mercifully short, but doesn’t have anything close to a coherent set of principles, ‘The Improvisation Book’, which I haven’t read in depth but seems to be little more than a list of games, and Dan Patterson and Mark Leveson’s ‘Whose Line is It Anyway’, which unsurprisingly is very heavily focused on emulating the restrictive format of the show of the same name.

Secondary recommendation: Mick Napier’s Improvise, which comes from a different school of thought to TIH’s – the same one as ‘Truth in Comedy’.

Reason It's the only one of any of those I’ve mentioned (TIH included) to explicitly suggest scientific reasoning in developing and assessing improv methods. After the author’s initial proclamation to that effect, he doesn’t really communicate how he’s tried to do so, and his advice seems to assume you’re already quite comfortable with being in an unspecified scene with no preset rules (one of the hardest things for an improviser to find himself in, IME), so I wouldn’t recommend it as a beginner’s guide.

How does one create an open thread? The only options I had available were this and comments. Is it something you need minimum karma for?

If I was reliable enough to be sure, then I'd probably know where the comment in question was :P But I am fairly confident. They weren't professionally made videos (nor was the website professional-looking), just a set by this one guy explaining one bit of maths at a time.

We've changed it to a more standard font now. How's that?

We've been experimenting a bit with the font. A bit of Googling failed to yield any scientific data on what font types people prefer, though there's a fair bit on font size. Sounds like a new test is in order, though... (incidentally, if you or anyone else reading this happen to know of any decent studies on fonts for reading online, I'd be keen to see it).