TrE comments on Harry Potter and the Methods of Rationality discussion thread, part 17, chapter 86 - Less Wrong

9 Post author: Alsadius 17 December 2012 07:19AM

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

Comments (606)

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

Comment author: TrE 17 December 2012 08:27:27PM *  2 points [-]

Moody would have to take into account the coriolis non-force, at least for very-long-range shots. How fast does a killing curse move? Also, to what amount is a curse affected by curved space? Do they react to gravity at all?

Comment author: DanielLC 18 December 2012 02:16:24AM 15 points [-]

It's not clear what it would mean to not be affected by curved space or gravity, since there's no "straight" besides geodesics and no "non-accelerating" besides freefall, but that doesn't seem to stop much in the HP verse.

Speaking of curved space, I've noticed that there are spells to make things bigger on the inside. If you do this just right, you can create a singularity known as a cone point that ought to send any spell fired at it back in the direction of the caster. Also, you could make a faster-than-light drive, which could be used as a time machine.

Comment author: TrE 18 December 2012 07:38:32AM 2 points [-]

If the inertial mass of a spell is greater than its gravitational mass, it would appear that the spell doesn't react to gravity as much as it should. It is also possible that spells work a bit like brooms.

Comment author: DanielLC 18 December 2012 08:32:32AM 2 points [-]

Gravity doesn't work that way. Something not reacting to gravity under general relativity is like something stopping under special relativity (or even Galilean invariance). However, considering that there's a spell that does just that, this doesn't mean much.

Comment author: Eliezer_Yudkowsky 19 December 2012 05:22:51AM 8 points [-]

I hereby declare Arresto Momentum to match the velocity of a small mass to the velocity of some much larger mass that the wizard thinks is a reference frame.

Comment author: MugaSofer 19 December 2012 12:32:43PM 6 points [-]

Does that mean Harry can't use it (becuse there is no universal reference frame) or he can use it in all sorts of munchkiny ways (I stop the car ... relative to the moon!)

Comment author: Eliezer_Yudkowsky 20 December 2012 12:29:54AM 4 points [-]

Well, for one thing, he's not powerful enough to cast it period, but if he were, I expect it would only work on near / nearest masses.

Comment author: shminux 20 December 2012 12:20:21AM 1 point [-]

munchkiny ways (I stop the car ... relative to the moon!)

That requires original research, like partial transfiguration.

Comment author: MugaSofer 20 December 2012 05:59:10PM -1 points [-]

That's what I had in mind, yes.

Comment author: faul_sname 21 December 2012 10:14:20PM 1 point [-]

Original research is dangerous. This original research in particular seems dangerous.

Comment author: MugaSofer 22 December 2012 02:10:57PM 1 point [-]

Now I come to think, he would probably just conclude it uses Earth as a reference frame, just as broomsticks use Aristotelian physics and transfiguration uses form/substance duality.

Comment author: Alsadius 19 December 2012 09:37:41AM 2 points [-]

So what would it do with a wizard who had truly internalized the principle of relativity, and who understood that there was no privileged reference frame? Could he use it to de facto impart an arbitrary velocity to an arbitrary object?

Comment author: shminux 20 December 2012 12:28:15AM 0 points [-]

impart an arbitrary velocity to an arbitrary object?

You can equalize velocity of a given small object with that of the caster's reference frame. The excess momentum is transferred to the Earth (or potentially another massive body, though there is no proof that magic works outside the Earth's atmosphere, which incidentally means that the Pioneercrux may have decayed). This way there is no preferred reference frame.

Comment author: pedanterrific 22 December 2012 12:23:55AM 1 point [-]

Of course there's no proof, but I don't think it's really possible, if only for literary reasons, that the Sphere of Stars spell is just a fancy hologram. I think we can take that as fairly strong evidence it's working as intended.

Comment author: Alsadius 20 December 2012 12:45:36AM 0 points [-]

So "arbitrary" might be a bit strong, but what's to prevent him from using the Moon or the Sun as a reference frame? Or even the Earth as a whole, instead of the local part? Even imparting a relative 1000 mph to an object near the equator would be pretty powerful.

Comment author: shminux 20 December 2012 01:13:47AM 0 points [-]

"the caster's rest frame" is relativistically invariant.

Comment author: Alsadius 20 December 2012 02:42:11AM 0 points [-]

But he didn't say "the caster's reference frame", he said "some much larger mass that the wizard thinks is a reference frame".

Comment author: Kawoomba 19 December 2012 12:49:59PM 1 point [-]
Comment author: MixedNuts 19 December 2012 11:49:23PM 1 point [-]

You can't use Arresto Momentum for that; you need a much more massive object as a reference, but your reference frame only contains massless objects. I guess you could say that massless objects have much larger mass than other massless objects, and use Arresto Momentum to make photons go in arbitrary directions with negligible losses.

If you can make a surface cast a long-lasting Arresto Momentum, you have a near-perfect mirror for all wavelengths from radio to gamma and all angles. That would be useful. (Of course you can also make the light slower if you wish, but the fun you can have with arbitrary refractive indices with no wavelength dependence and no absorption is redundant here.)

Also, if we figure out the necessary ratio of masses at given magical energy, we can use the spell to measure mass. Assuming your aim is precise enough to Arresto Momentum a neutrino, anyway.

Comment author: TrE 18 December 2012 08:46:31AM 1 point [-]

Magic works like the author of the story wants it to work.

Comment author: Alsadius 18 December 2012 10:42:32AM 0 points [-]

I think it's safe to say that the HPMOR universe still runs on some sort of rules, even if they're not the usual ones Einstein taught us.

Comment author: MugaSofer 18 December 2012 08:11:31PM 3 points [-]

Actually, Eliezer has stated that he deliberately inserts impossible details in all his stories because he's disturbed by the notion that they might be simulated somewhere. I wouldn't expect it to affect the plot, though.

Comment author: DanielLC 18 December 2012 10:56:24PM 1 point [-]

Didn't he say that he mentioned the time-turner because it disproves the simulation hypothesis? It would also guarantee that nobody would simulate it.

He seems to have figured out how to simulate it later, but that requires simulating everything, and there's no guarantee that there will even be a reality with a stable time loop, unless you have the infinite number of realities necessary to allow connectedness, in which case you can't possibly simulate all the possibilities.

Comment author: MugaSofer 19 December 2012 01:25:47AM *  1 point [-]

Actually, there is a guarantee that there will be a stable time loop. Look up the Novicov Consistency Principal some time. And I think that was to stop people speculating that magic was being provided by the Matrix Lords, although sadly it doesn't. I was referring to the corridor tiled with pentagons (although it never actually says they were regular pentagons) and the spiral staircase that lifts you by rotating.

Comment author: DanielLC 19 December 2012 02:08:20AM *  3 points [-]

The Novicov Consistency Principal relies on the Kakutani fixed-point theorem, which relies on convexness, which implies connectedness, which implies an infinite number of realities.

Comment author: Armok_GoB 23 December 2012 02:02:59AM 2 points [-]

I can think of ways to simulate those quite easily. It does involve cheating with the environment, but not really cheating with any minds. Mostly the same kinds of tricks mentioned in Eliezers "that'd it take to make me belive 2+2=3" article.

Like, for example, deforming the tiling pattern constantly so that it was always the right type of angles and side lengths where the eyes were looking, and stopping the motion detection from going of on those changes. Or have the stairs tile exactly, just snap people on them up in exact increments of tiles, and again doing clever things with the way motion is detected in the eye.

Comment author: ygert 19 December 2012 08:17:13AM 1 point [-]

Probably, but these are not mutually exclusive. If Dumbledore is right, the set of rules that the HPMOR world runs on are the laws of Narrative Causality.

Comment author: MugaSofer 19 December 2012 12:30:09PM -1 points [-]

You know that spell was invented for the film, right?

Comment author: DanielLC 19 December 2012 09:57:06PM 0 points [-]

According to the wiki:

Appearances Harry Potter and the Prisoner of Azkaban (First mentioned) Harry Potter and the Prisoner of Azkaban (film) (First identified as Aresto Momentum)

It looks to me like they established that there is a spell that does that, but it wasn't named until the movie. I suppose without the name there's no real evidence that it doesn't just accelerate the subject in any way the caster pleases.

In any case, in the MoR verse, the brooms act in a manner that involves some sort of rest state. Also, as of Eliezer's reply to this post, that spell canonically exists in the MoR verse.

Comment author: MugaSofer 19 December 2012 10:18:20PM 1 point [-]

Well, it's the name that actually tells us it's "arresting momentum" and not, I don't know, stopping him getting too close to the ground or some sort of anti-gravity spell. The scene is in the book, but the spell itself is not.

(The wiki treats the films a canon when they don't actually contradict the books, instead of a separate canon like most sane people would. They do the same for everything, in fact, from videogames to trading cards, although they draw the line at fanfiction.)

Comment author: MugaSofer 18 December 2012 08:09:56PM -1 points [-]

"Making things bigger on the inside" does not equal "bending space into any damn shape we please."

Comment author: DanielLC 18 December 2012 10:52:06PM 1 point [-]

A cone point only requires being able to make things bigger on the inside. Just make a series of concentric spheres that are the right amount bigger on the inside and outside. If you don't make it smooth, it won't be perfect, but it just has to be good enough.

I don't know much about the faster-than-light stuff. From what I can find, an Alcubierre drive needs at least some negative mass, which, from what I understand, would translate to having to make things smaller on the inside. It seems likely that wizards would be able to make things smaller on the inside too, and they just don't do it much because it's pretty useless, at least for things wizards would think of.

Comment author: MugaSofer 19 December 2012 01:19:54AM 0 points [-]

Good point on the spheres, although I think Extension Charms are tied to enclosed spaces (hell, they might not even bend space. They could tap into another dimension or move you "out of phase" or something.)

Regarding the FTL, Alcubierre drives need a ring (or donut in the latest designs) of negative energy (or matter, I guess.) I can see that translating into "smaller on the inside" constructions, although I doubt there's actually a spell for that (although you never know.) Might be able to transfigure the stuff, though.

Comment author: DanielLC 19 December 2012 02:00:07AM *  3 points [-]

I think Extension Charms are tied to enclosed spaces

Perhaps that's just because no wizard is well-versed in non-euclidean geometry well-enough to understand how it would work in a non-enclosed space.

Besides, enclosed is relative. Maybe light doesn't travel through it, but neutrinos and your enemy's spells do.

(hell, they might not even bend space. They could tap into another dimension or move you "out of phase" or something.)

It doesn't matter. As long as it takes longer to travel through, it works. For example, you could use different kinds of glass to make retroflectors using the same principle, since light will travel at different speeds.

although I doubt there's actually a spell for that (although you never know.)

Perhaps the same spell works. If the same spell could make something two or three times bigger on the inside, then why not one half times bigger?

Comment author: MixedNuts 19 December 2012 02:27:10PM 3 points [-]

What happens if you make it bigger on the inside and then turn it inside out?

Comment author: MugaSofer 19 December 2012 02:47:14PM *  -1 points [-]

Perhaps that's just because no wizard is well-versed in non-euclidean geometry well-enough to understand how it would work in a non-enclosed space.

Could be. Hard to tell without the author telling us.

Besides, enclosed is relative. Maybe light doesn't travel through it, but neutrinos and your enemy's spells do.

I think most spells are disrupted by solid objects. And most offensive spells would risk destroying whatever you charmed. Would be interesting to try it on something made of glass, though.

It doesn't matter. As long as it takes longer to travel through, it works. For example, you could use different kinds of glass to make retroflectors using the same principle, since light will travel at different speeds.

Oh, I see. Cool. I assumed they were purely theoretical.

Perhaps the same spell works. If the same spell could make something two or three times bigger on the inside, then why not one half times bigger?

Interesting point. I doubt it works by specifying the increase as a number, though.

Hmm, there are probably other uses for containers with shrunken insides.

Comment author: Kindly 19 December 2012 08:00:27PM 3 points [-]

Hmm, there are probably other uses for containers with shrunken insides.

Inflatable mattresses (or waterbeds) that can be filled quickly/efficiently.

If rooms can be enchanted in this way (houses can certainly be bigger on the inside) then enchanting a long hallway in this way would allow people to walk longer distances in a short amount of time.

If the container is transparent, it could be used as a magnifying glass for things you placed inside.

Comment author: MugaSofer 19 December 2012 09:10:37PM 0 points [-]

Wizards already have shrinking charms for the first one, but well done.

Comment author: DanielLC 19 December 2012 09:55:11PM 1 point [-]

I assumed they were purely theoretical.

In a sense they are. I've never seen anything about anybody actually doing that. I just know it would work in theory.

Lenses and optic fibers and such work on the same principle, but they aren't cone points.