1063: "Kill Hitler"

This forum is for the individual discussion thread that goes with each new comic.

Moderators: Moderators General, Prelates, Magistrates

Edrees
Posts: 32
Joined: Wed Dec 23, 2009 8:16 pm UTC

Re: 1063: "Kill Hitler"

Postby Edrees » Tue Jun 12, 2012 6:39 pm UTC

Technically, if you had one use backward and one use forward, you could kill hitler and return to the moment just before you used the time machine, so you could use it again since you aren't returning to the present.

But if you didn't have a use back and a use forward but rather a round trip use where your return just puts you back to the moment you started, then yeah you're boned with one.

J Thomas
Everyone's a jerk. You. Me. This Jerk.^
Posts: 1190
Joined: Fri Sep 23, 2011 3:18 pm UTC

Re: 1063: "Kill Hitler"

Postby J Thomas » Tue Jun 12, 2012 9:45 pm UTC

gmalivuk wrote:
J Thomas wrote:You hypothesize that there is a single exception, and all you hypothesize about that exception is that it looks like a blue bottle and it is affected less by gravity but has the same mass when subject to other forces. The only way for me to test that is to weigh each blue bottle I find, and also subject it to electromagnetic force, and look for an anomaly. But since the chance is very small that the bottle I test is the one that is anomalous, this is an extremely expensive test.
What does probability have to do with it? Weighing an individual blue bottle is cheap, and on your account each of the individual hypotheses (this blue bottle is the unusual one) is worth testing.

By pointing out that testing *all* of them is expensive and time consuming, you're really just making our point for us: testing absurd hypotheses in their entirety is often expensive and time consuming and thus not worth doing. Because the expected informational payoff is so incredibly low compared to the required effort.


It turns cheap after you tell me which one you think is different.

Now note, of course, that it's entirely possible to stumble across the alleged anomaly without ever wasting time testing for it. If I happen to be using blue bottles in some entirely unrelated experiment, and notice that one isn't affected by gravity as much as I'd expect, I have managed to confirm the wacky hypothesis without ever wasting time studying it.

And your wacky neutrinos-aren't-real hypothesis is the same way. Discrepancies from how we expect "real" things to behave in situations where we (perhaps mistakenly) believe there are neutrinos can show up without anyone needing to actually seriously test the hypothesis that neutrinos aren't a real thing. Just like apparent velocity discrepancies showed up without anyone wasting time designing a test specifically to determine whether neutrinos might travel faster than light. After that was the correct time to investigate further and try to figure out what was really going on, not before.


Try playing the Eleusis game. It will give you experience using scientific method. In an artificial situation, but still....

This is something I know about and you don't. You are not open to learning from me about it. Let's agree to disagree until one of us changes his mind on his own.
Last edited by J Thomas on Tue Jun 12, 2012 10:32 pm UTC, edited 1 time in total.
The Law of Fives is true. I see it everywhere I look for it.

User avatar
gmalivuk
GNU Terry Pratchett
Posts: 26767
Joined: Wed Feb 28, 2007 6:02 pm UTC
Location: Here and There
Contact:

Re: 1063: "Kill Hitler"

Postby gmalivuk » Tue Jun 12, 2012 10:21 pm UTC

J Thomas wrote:It turns cheap after you tell me which one you think is different.
Why does that matter, when you've asserted that *every* alternative hypothesis is worth testing? Simply treat my hypothesis (there is a bottle that has this property) as a collection of alternatives (Hi says bottle i has this property). Now each one of those hypotheses is worth testing, and a decision not to bother every time is "bad science" in your ever so wise opinion.

This is something I know about and you don't.
You have yet to provide me with any reason to believe this statement. For one thing, if I didn't understand the scientific method but you did, I would expect all or at least some of the other scientists having this discussion to disagree with me and agree with you.

Or are all the physicists as hopelessly ignorant as I am?
Last edited by gmalivuk on Tue Jun 12, 2012 10:42 pm UTC, edited 2 times in total.
Unless stated otherwise, I do not care whether a statement, by itself, constitutes a persuasive political argument. I care whether it's true.
---
If this post has math that doesn't work for you, use TeX the World for Firefox or Chrome

(he/him/his)

J Thomas
Everyone's a jerk. You. Me. This Jerk.^
Posts: 1190
Joined: Fri Sep 23, 2011 3:18 pm UTC

Re: 1063: "Kill Hitler"

Postby J Thomas » Tue Jun 12, 2012 10:31 pm UTC

gmalivuk wrote:
J Thomas wrote:It turns cheap after you tell me which one you think is different.
Why does that matter, when you've asserted that *every* alternative hypothesis is worth testing? Simply treat my hypothesis (there is a bottle that has this property) as a collection of alternatives (Hi says bottle i has this property). Now each one of those hypotheses is worth testing, and a decision not to bother every time is "bad science" in your ever so wise opinion.

This is something I know about and you don't.
You have yet to provide me with any reason to believe this statement. For one thing, if I didn't understand the scientific method but you did, I would expect all or at least some of the other scientists having this discussion to disagree with me and agree with you.

Or are all the physicists as hopelessly ignorant as I am?


I invite you to agree to disagree.
Also I want to encourage you to play Eleusis, which is a fun game.
The Law of Fives is true. I see it everywhere I look for it.

User avatar
gmalivuk
GNU Terry Pratchett
Posts: 26767
Joined: Wed Feb 28, 2007 6:02 pm UTC
Location: Here and There
Contact:

Re: 1063: "Kill Hitler"

Postby gmalivuk » Tue Jun 12, 2012 10:48 pm UTC

Regarding that card game as it relates to this discussion, you still seem to be missing just how absurd some hypotheses are. Suppose the sequence so far is

3C, 6D, 5C, 2D, KS, 2H, AC, 8H

and suppose further that every attempt to play two odds or two evens in a row has been rejected, along with every attempt to place a red on a red or black on black. Nothing, however, has thus far disproved the possible rule that you have to play a red 6 on an 8. Is it worth using up my turn to test that particular hypothesis? Will I have really gained useful information if I play 6H next and it is rejected?

More generally, given any sequence of already played cards along with any cards that were rejected (and where in the sequence they were rejected), and given any other individual card, there exists a rule whereby that card is playable next. This does not, however, mean that I should give all possible rules equal weight, because different possible rules have different prior likelihood of being the one chosen by the dealer.
Last edited by gmalivuk on Tue Jun 12, 2012 10:54 pm UTC, edited 1 time in total.
Unless stated otherwise, I do not care whether a statement, by itself, constitutes a persuasive political argument. I care whether it's true.
---
If this post has math that doesn't work for you, use TeX the World for Firefox or Chrome

(he/him/his)

J Thomas
Everyone's a jerk. You. Me. This Jerk.^
Posts: 1190
Joined: Fri Sep 23, 2011 3:18 pm UTC

Re: 1063: "Kill Hitler"

Postby J Thomas » Tue Jun 12, 2012 10:50 pm UTC

gmalivuk wrote:Regarding that card game as it relates to this discussion, you still seem to be missing just how absurd some hypotheses are. Suppose the sequence so far is

3C, 6D, 5C, 2D, KS, 2H, AC, 8H

and suppose further that every attempt to play two odds or two evens in a row has been rejected, along with every attempt to place a red on a red or black on black. Nothing, however, has thus far disproved the possible rule that you have to play a red 6 on an 8. Is it worth using up my turn to test that particular hypothesis?

More generally, given any sequence of already played cards along with any cards that were rejected (and where in the sequence they were rejected), and given any other individual card, there exists a rule whereby that card is playable next. This does not, however, mean that my best strategy is to play cards completely at random.


I invite you to play the game and find out what works.

Note that to keep it fun, it's good to have rules that are reasonably simple and I think it helps if at any turn the fraction of playable cards is somewhere between 1/4 and 3/4. If you play with some real sophisticated people they might like rules like: play black on an even card and red on an odd one but try not to get tricky until they start getting all the easy ones real fast.
The Law of Fives is true. I see it everywhere I look for it.

User avatar
gmalivuk
GNU Terry Pratchett
Posts: 26767
Joined: Wed Feb 28, 2007 6:02 pm UTC
Location: Here and There
Contact:

Re: 1063: "Kill Hitler"

Postby gmalivuk » Tue Jun 12, 2012 11:04 pm UTC

J Thomas wrote:Note that to keep it fun, it's good to have rules that are reasonably simple and I think it helps if at any turn the fraction of playable cards is somewhere between 1/4 and 3/4. If you play with some real sophisticated people they might like rules like: play black on an even card and red on an odd one but try not to get tricky until they start getting all the easy ones real fast.
What I'm still confused about is how you think this strategy supports *your* claims about how science should work.

At any point, after collecting any finite amount of data (in the game or in real science), there are infinitely many hypotheses that have not been disproven, and literally any test you do can eliminate some of them. But since we don't have infinite time or memory, it is irrational to keep in mind the set of all such "tentatively true" hypothesis, and make note at each step of *every* one that remains possible after that step.
Unless stated otherwise, I do not care whether a statement, by itself, constitutes a persuasive political argument. I care whether it's true.
---
If this post has math that doesn't work for you, use TeX the World for Firefox or Chrome

(he/him/his)

J Thomas
Everyone's a jerk. You. Me. This Jerk.^
Posts: 1190
Joined: Fri Sep 23, 2011 3:18 pm UTC

Re: 1063: "Kill Hitler"

Postby J Thomas » Tue Jun 12, 2012 11:58 pm UTC

gmalivuk wrote:
J Thomas wrote:Note that to keep it fun, it's good to have rules that are reasonably simple and I think it helps if at any turn the fraction of playable cards is somewhere between 1/4 and 3/4. If you play with some real sophisticated people they might like rules like: play black on an even card and red on an odd one but try not to get tricky until they start getting all the easy ones real fast.
What I'm still confused about is how you think this strategy supports *your* claims about how science should work.

At any point, after collecting any finite amount of data (in the game or in real science), there are infinitely many hypotheses that have not been disproven, and literally any test you do can eliminate some of them. But since we don't have infinite time or memory, it is irrational to keep in mind the set of all such "tentatively true" hypothesis, and make note at each step of *every* one that remains possible after that step.


I think if you play the game you'll come up with something workable. It really doesn't matter how I do it, right?
The Law of Fives is true. I see it everywhere I look for it.

User avatar
gmalivuk
GNU Terry Pratchett
Posts: 26767
Joined: Wed Feb 28, 2007 6:02 pm UTC
Location: Here and There
Contact:

Re: 1063: "Kill Hitler"

Postby gmalivuk » Wed Jun 13, 2012 12:18 am UTC

J Thomas wrote:My position on science is that all hypotheses that have not been disproven should be considered tentatively true. Any test that can eliminate some of them is worth doing, no matter how unlikely or ridiculous some people believe those hypotheses are.
Do you still stand by this claim? Forget the card game, since it's even farther off topic than this whole discussion of how science should work. Do you still claim that "any test that can eliminate some [hypotheses not yet disproven] is worth doing, no matter how unlikely"?

Because that this is absurd.

Your claim implies that literally everything is worth doing, as a scientific experiment, because observing the results of any event will eliminate some (an infinite number, in fact!) of those "tentatively true" hypotheses that fit with all the evidence about the universe that we have so far.

You are arguing against *any* form of inductive bias, which means we cannot reason inductively, which means we cannot do science.

And yet I'm apparently the one who doesn't know how science works...
Unless stated otherwise, I do not care whether a statement, by itself, constitutes a persuasive political argument. I care whether it's true.
---
If this post has math that doesn't work for you, use TeX the World for Firefox or Chrome

(he/him/his)

J Thomas
Everyone's a jerk. You. Me. This Jerk.^
Posts: 1190
Joined: Fri Sep 23, 2011 3:18 pm UTC

Re: 1063: "Kill Hitler"

Postby J Thomas » Wed Jun 13, 2012 1:43 am UTC

gmalivuk wrote:
J Thomas wrote:My position on science is that all hypotheses that have not been disproven should be considered tentatively true. Any test that can eliminate some of them is worth doing, no matter how unlikely or ridiculous some people believe those hypotheses are.
Do you still stand by this claim? Forget the card game, since it's even farther off topic than this whole discussion of how science should work. Do you still claim that "any test that can eliminate some [hypotheses not yet disproven] is worth doing, no matter how unlikely"?

Because that this is absurd.

Your claim implies that literally everything is worth doing, as a scientific experiment, because observing the results of any event will eliminate some (an infinite number, in fact!) of those "tentatively true" hypotheses that fit with all the evidence about the universe that we have so far.

You are arguing against *any* form of inductive bias, which means we cannot reason inductively, which means we cannot do science.

And yet I'm apparently the one who doesn't know how science works...


Do you really care to understand my ideas about this? I don't think so. What does it matter whether you understand what I'm saying or not? Let's agree to disagree.
The Law of Fives is true. I see it everywhere I look for it.

User avatar
gmalivuk
GNU Terry Pratchett
Posts: 26767
Joined: Wed Feb 28, 2007 6:02 pm UTC
Location: Here and There
Contact:

Re: 1063: "Kill Hitler"

Postby gmalivuk » Wed Jun 13, 2012 1:45 am UTC

J Thomas wrote:Do you really care to understand my ideas about this?
I asked you what you think, didn't I? If you don't care to explain it, fine, but please don't ascribe motivations to me.

Also, don't confuse my disagreement with a lack of (interest in) understanding your ideas.
Unless stated otherwise, I do not care whether a statement, by itself, constitutes a persuasive political argument. I care whether it's true.
---
If this post has math that doesn't work for you, use TeX the World for Firefox or Chrome

(he/him/his)

J Thomas
Everyone's a jerk. You. Me. This Jerk.^
Posts: 1190
Joined: Fri Sep 23, 2011 3:18 pm UTC

Re: 1063: "Kill Hitler"

Postby J Thomas » Wed Jun 13, 2012 1:55 am UTC

gmalivuk wrote:
J Thomas wrote:Do you really care to understand my ideas about this?
I asked you what you think, didn't I? If you don't care to explain it, fine, but please don't ascribe motivations to me.

Also, don't confuse my disagreement with a lack of (interest in) understanding your ideas.


I think of all the times remaining to me to stop explaining it to you, this is the best.

I tried discussing your motives in private message but you haven't answered yet though you have replied many times since in public. I think under the circumstances it's OK for me to make hypotheses about your motives.
The Law of Fives is true. I see it everywhere I look for it.

User avatar
gmalivuk
GNU Terry Pratchett
Posts: 26767
Joined: Wed Feb 28, 2007 6:02 pm UTC
Location: Here and There
Contact:

Re: 1063: "Kill Hitler"

Postby gmalivuk » Wed Jun 13, 2012 2:13 am UTC

J Thomas wrote:I tried discussing your motives in private message
No, you said I was trolling you in a private message. That is not the same as trying to discuss something.

(In any case, my motives here are pretty much what they are every time I get in an argument with someone I don't know: figure out and/or develop what I think and my reasons for thinking so.)
Unless stated otherwise, I do not care whether a statement, by itself, constitutes a persuasive political argument. I care whether it's true.
---
If this post has math that doesn't work for you, use TeX the World for Firefox or Chrome

(he/him/his)

J Thomas
Everyone's a jerk. You. Me. This Jerk.^
Posts: 1190
Joined: Fri Sep 23, 2011 3:18 pm UTC

Re: 1063: "Kill Hitler"

Postby J Thomas » Wed Jun 13, 2012 3:40 am UTC

gmalivuk wrote:(In any case, my motives here are pretty much what they are every time I get in an argument with someone I don't know: figure out and/or develop what I think and my reasons for thinking so.)


You have done admirably. Many students make less progress in a 1-quarter class. If I can find the time I'd like to review the thread to watch how you changed your thinking, and get ideas what might have prepared the way for that.
The Law of Fives is true. I see it everywhere I look for it.

User avatar
gmalivuk
GNU Terry Pratchett
Posts: 26767
Joined: Wed Feb 28, 2007 6:02 pm UTC
Location: Here and There
Contact:

Re: 1063: "Kill Hitler"

Postby gmalivuk » Wed Jun 13, 2012 4:46 am UTC

How about instead you stop patronizingly acting like I'm a student of yours?
Unless stated otherwise, I do not care whether a statement, by itself, constitutes a persuasive political argument. I care whether it's true.
---
If this post has math that doesn't work for you, use TeX the World for Firefox or Chrome

(he/him/his)

curtis95112
Posts: 638
Joined: Thu Jan 27, 2011 5:23 pm UTC

Re: 1063: "Kill Hitler"

Postby curtis95112 » Wed Jun 13, 2012 5:11 am UTC

J Thomas.

But since the chance is very small that the bottle I test is the one that is anomalous, this is an extremely expensive test.


No, it's very cheap. The hypothesis to be tested is "This bottle is the one that is affected less". I admit that it's very unlikely to be true. And that the expected returns are extremely small even when compared to the tiny cost. Which is why I personally wouldn't test this.

But according to you.
Any test that can eliminate some of them is worth doing, no matter how unlikely or ridiculous some people believe those hypotheses are.


Do you still stand by this?
Mighty Jalapeno wrote:
Tyndmyr wrote:
Роберт wrote:Sure, but at least they hit the intended target that time.

Well, if you shoot enough people, you're bound to get the right one eventually.

Thats the best description of the USA ever.

J Thomas
Everyone's a jerk. You. Me. This Jerk.^
Posts: 1190
Joined: Fri Sep 23, 2011 3:18 pm UTC

Re: 1063: "Kill Hitler"

Postby J Thomas » Wed Jun 13, 2012 6:06 am UTC

gmalivuk wrote:How about instead you stop patronizingly acting like I'm a student of yours?


You aren't a student of mine. You are figuring out for yourself what stand you want to take. I have no expectation that you'll wind up agreeing with me, and yet when it gels I fully expect you'll come up with something that makes sense given your assumptions.
The Law of Fives is true. I see it everywhere I look for it.

User avatar
PM 2Ring
Posts: 3713
Joined: Mon Jan 26, 2009 3:19 pm UTC
Location: Sydney, Australia

Re: 1063: "Kill Hitler"

Postby PM 2Ring » Wed Jun 13, 2012 8:00 am UTC

(I get the feeling I'm going to regret posting to this thread. Oh, well. :) )

The modern formulation of physics is all about invariants. Perhaps a more effective formulation can be found, but I'm sceptical. Abandoning any of the core invariants without good reason and without a more effective approach sounds like a recipe for unnecessary complexity. As was mentioned earlier in the thread, the laws of physics give us a way to compress the observed physical data in such a way that we can not only reproduce that data but we can also predict as yet unobserved data. A more effective theory ought to give us a better compression ratio, but a theory that invokes ad hoc exceptions gives us a worse compression ratio.

If you claim that neutrinos don't exist and thus the weak nuclear interactions do not conserve energy, momentum and angular momentum, then there are consequences beyond weak nuclear interactions. If the invariance of the laws of physics to (local) changes in time, space and direction is not correct, then there exist one or more preferred inertial frames, which breaks relativity for everything, not just interactions involving the weak force.

FWIW, here's what I said in another post re: Noether's Theorem:
PM 2Ring wrote:Noether's theorem says that the law of conservation of energy is equivalent to the fact that the equations describing a closed system are invariant under a translation in time. In other words, if you're in a sealed laboratory (which is sealed so well that you can't detect anything about the outside universe, and the lab is in freefall, so you can't feel any gravity), there's no experiment that you can perform which is dependent on what time it is in the outside world. So if you perform some experiment, and then your lab is magically transported into the past or future and you perform exactly the same experiment again, you'll get the same result that you got the first time.

Similarly, the law of conservation of linear momentum is equivalent to the fact that the equations describing a closed system are invariant under a translation in space. So if your lab is transported in space, it won't affect the results of your experiment. Conservation of angular momentum is equivalent to invariance under a change of orientation, so rotating your lab to point in a different direction doesn't affect your experiments either.

Another way to look at these conservation principles is that conservation of energy tells us that there's no absolute notion of time in the universe; conservation of momentum says there's no preferred frame of spatial reference, and no place that is the centre of the universe; and conservation of angular momentum says there's no preferred orientation, no "this way is up". OTOH, these statements don't necessarily hold at the instant of the Big Bang, since that instant is a special point in time.

User avatar
gmalivuk
GNU Terry Pratchett
Posts: 26767
Joined: Wed Feb 28, 2007 6:02 pm UTC
Location: Here and There
Contact:

Re: 1063: "Kill Hitler"

Postby gmalivuk » Wed Jun 13, 2012 1:25 pm UTC

J Thomas wrote:You aren't a student of mine.
Pretty sure I already knew that, which is why it's patronizing when you act as though I am one.

when it gels I fully expect you'll come up with something that makes sense given your assumptions.
What I started out with already made sense given my assumptions, and really hasn't changed:

Some hypotheses are too unlikely, given what we already know, to be worth taking seriously in the sense of running tests to differentiate between the unlikely hypothesis and one that's more reasonable.
Unless stated otherwise, I do not care whether a statement, by itself, constitutes a persuasive political argument. I care whether it's true.
---
If this post has math that doesn't work for you, use TeX the World for Firefox or Chrome

(he/him/his)

J Thomas
Everyone's a jerk. You. Me. This Jerk.^
Posts: 1190
Joined: Fri Sep 23, 2011 3:18 pm UTC

Re: 1063: "Kill Hitler"

Postby J Thomas » Wed Jun 13, 2012 2:48 pm UTC

PM 2Ring wrote:(I get the feeling I'm going to regret posting to this thread. Oh, well. :) )

The modern formulation of physics is all about invariants. Perhaps a more effective formulation can be found, but I'm sceptical. Abandoning any of the core invariants without good reason and without a more effective approach sounds like a recipe for unnecessary complexity. As was mentioned earlier in the thread, the laws of physics give us a way to compress the observed physical data in such a way that we can not only reproduce that data but we can also predict as yet unobserved data. A more effective theory ought to give us a better compression ratio, but a theory that invokes ad hoc exceptions gives us a worse compression ratio.

If you claim that neutrinos don't exist and thus the weak nuclear interactions do not conserve energy, momentum and angular momentum, then there are consequences beyond weak nuclear interactions. If the invariance of the laws of physics to (local) changes in time, space and direction is not correct, then there exist one or more preferred inertial frames, which breaks relativity for everything, not just interactions involving the weak force.

FWIW, here's what I said in another post re: Noether's Theorem:
PM 2Ring wrote:Noether's theorem says that the law of conservation of energy is equivalent to the fact that the equations describing a closed system are invariant under a translation in time. In other words, if you're in a sealed laboratory (which is sealed so well that you can't detect anything about the outside universe, and the lab is in freefall, so you can't feel any gravity), there's no experiment that you can perform which is dependent on what time it is in the outside world. So if you perform some experiment, and then your lab is magically transported into the past or future and you perform exactly the same experiment again, you'll get the same result that you got the first time.

Similarly, the law of conservation of linear momentum is equivalent to the fact that the equations describing a closed system are invariant under a translation in space. So if your lab is transported in space, it won't affect the results of your experiment. Conservation of angular momentum is equivalent to invariance under a change of orientation, so rotating your lab to point in a different direction doesn't affect your experiments either.

Another way to look at these conservation principles is that conservation of energy tells us that there's no absolute notion of time in the universe; conservation of momentum says there's no preferred frame of spatial reference, and no place that is the centre of the universe; and conservation of angular momentum says there's no preferred orientation, no "this way is up". OTOH, these statements don't necessarily hold at the instant of the Big Bang, since that instant is a special point in time.


Thank you. I will present an alternative point of view, not with the expectation that it is practical but mostly in hope of entertaining you.

Say that you are in a sealed laboratory, and you have a neutrino detector. Somebody brings a nuclear reactor close to the outside of your lab and you can detect it. They move the reactor far away and you can detect that. In theory, if you had a way to seal your lab so neutrinos couldn't get in then your neutrino detector would get the same result regardless. But in practice you can't do that. So in reality, conservation of linear momentum fails because you can detect neutrinos, and conservation of angular momentum fails because you can detect which direction they come from.

If the most widely believed theory of neutrinos is correct, then this is just a quibble. All it takes is to put your lab behind a lead wall a million meters thick, and the neutrinos would be blocked. And it makes some sense to believe that, because the number of neutrinos detected varies with the time of day which might mean that the bulk of the earth (8000 miles thick at maximum) affects neutrinos from the sun. With enough data about neutrino detection day and night across the seasons, you could infer the rate of neutrinos from the sky in all directions apart from the sun. All this is reasonable, and it's a collection of reasonable assumptions piled on each other. It could still be wrong.

I can certainly understand the esthetic value of data compression. That's one of the big virtues of poetry. With poetry you can say powerful things in a few words, and the compression is so valuable that it's worth losing precision. In physics often you can have concision and precision both -- you can express fundamental rules in a little space without being wrong. Of course, in most real situations there are endless complications because there's so much going on that interacts in complex ways, but all that complexity could all come from the simple rules, provided you have all the simple rules worked out correctly. But what if the reality turns out to be a little more complicated? Then you really ought to accept reality rather than insist on an artificial simplicity.

It makes some sense to assume undetectable neutrinos instead of assume conservation laws are violated. But if you try it out the other way and that suggests an interesting experiment, then you come out ahead. And you wouldn't have to give up conservation laws completely. Like, in my office most of the time there's conservation of paperclips. in reality occasionally paperclips get destroyed when somebody bends one out of recognition, and occasionally a new box of paperclips is bought, but most of the time they are conserved. Clearly the physics conservation laws people like are mostly true, and they could still be used.

You figure that if conservation laws fail then there must be a preferred inertial frame. Doesn't that depend on the pattern in which they fail? I have the idea they could fail in a way that wouldn't single out one inertial frame as the preferred one, though I don't have an example in mind right now that I'm certain would work. I could be wrong. If I'm right then your idea that if we give up anything we lose it all, is an example of a pitfall with the concise physical theory approach. When it all seems to fit together too precisely, when it seems like a perfect jewel, then it's only natural to believe that losing any single part would result in chaos, the jewel must be smashed to powder. But remove one part and there might be ways to salvage a lot of the rest. Perhaps it could fit together as a different jewel.

My hope is that the more different theories you can entertain, the better you may think of interesting experiments.

.... we can not only reproduce that data but we can also predict as yet unobserved data


There is room for problems here. Consider the following hypothetical: A variety of types of neutrinos have been postulated by symmetry, from theory. Some of them are detectable when conservation laws are violated in the amounts consistent with theory. Others are not yet detectable. So likely someone will find reactions that should violate conservation laws in the amounts predicted for these currently-undetectable particles, and when that happens there will be a big announcement that the predicted particles have been detected.

But what if a different theory predicted particles that correspond to different amounts of conservation violation? What if people who looked for those particles found them? But we won't find them because we won't think to look for them.

The harder it is to find things unless you look for them, the more likely you find only what you look for. Damn, that was tautological. But seriously, when a neutrino detector requires lots of computing power to toss out large numbers of observations that are not what you're looking for and keep only the ones that fit your theory, is there a chance that you have in fact overdetermined the system? Do they do the control of telling their computers to look for something else, for particles with randomly-chosen parameters, and see whether those might fit their predictions as well as the theoretically correct ones do?

Disclaimer: I did not do well as an experimentalist. I found too many alternative interpretations and did too many controls. I would have done better to make the most interesting interpretation only, and attempt to publish, and revise if reviewers objected.
The Law of Fives is true. I see it everywhere I look for it.

curtis95112
Posts: 638
Joined: Thu Jan 27, 2011 5:23 pm UTC

Re: 1063: "Kill Hitler"

Postby curtis95112 » Wed Jun 13, 2012 4:44 pm UTC

Say that you are in a sealed laboratory, and you have a neutrino detector. Somebody brings a nuclear reactor close to the outside of your lab and you can detect it. They move the reactor far away and you can detect that. In theory, if you had a way to seal your lab so neutrinos couldn't get in then your neutrino detector would get the same result regardless. But in practice you can't do that. So in reality, conservation of linear momentum fails because you can detect neutrinos, and conservation of angular momentum fails because you can detect which direction they come from.

If the most widely believed theory of neutrinos is correct, then this is just a quibble. All it takes is to put your lab behind a lead wall a million meters thick, and the neutrinos would be blocked. And it makes some sense to believe that, because the number of neutrinos detected varies with the time of day which might mean that the bulk of the earth (8000 miles thick at maximum) affects neutrinos from the sun. With enough data about neutrino detection day and night across the seasons, you could infer the rate of neutrinos from the sky in all directions apart from the sun. All this is reasonable, and it's a collection of reasonable assumptions piled on each other. It could still be wrong.


I don't understand what you're trying to say. Why do the conservation laws fail in your scenario? Could you explain?
Mighty Jalapeno wrote:
Tyndmyr wrote:
Роберт wrote:Sure, but at least they hit the intended target that time.

Well, if you shoot enough people, you're bound to get the right one eventually.

Thats the best description of the USA ever.

User avatar
gmalivuk
GNU Terry Pratchett
Posts: 26767
Joined: Wed Feb 28, 2007 6:02 pm UTC
Location: Here and There
Contact:

Re: 1063: "Kill Hitler"

Postby gmalivuk » Wed Jun 13, 2012 5:46 pm UTC

J Thomas wrote:Say that you are in a sealed laboratory, and you have a neutrino detector. Somebody brings a nuclear reactor close to the outside of your lab and you can detect it. They move the reactor far away and you can detect that. In theory, if you had a way to seal your lab so neutrinos couldn't get in then your neutrino detector would get the same result regardless. But in practice you can't do that. So in reality, conservation of linear momentum fails because you can detect neutrinos, and conservation of angular momentum fails because you can detect which direction they come from.
Yeah, I don't get why those observations involve or imply violations of conservation. It just means energy and momentum are getting into your lab from the outside.

it's a collection of reasonable assumptions piled on each other. It could still be wrong.
Yes, and no one is denying that, anywhere in this thread.

Where we disagree with you is your conclusion that, because the current theory could be wrong, therefore every conceivable alternative is worth testing.

But what if the reality turns out to be a little more complicated?
Then it will turn out our theories need to be a little more complicated. But it is *not* the case that therefore we should waste time and effort *now* testing all these more complicated possible theories, before there's any data whatsoever that suggests our current understanding is so fundamentally flawed.

My hope is that the more different theories you can entertain, the better you may think of interesting experiments.
Real scientists doing real experiments, on the other hand, value quality over quantity. The interesting experiments are the ones with a high probability of giving us information, and are therefore *not* the ones designed to test exceedingly improbable theories.

I did not do well as an experimentalist. I found too many alternative interpretations and did too many controls.
As I said before, you seem extremely bothered by the need for inductive bias and making assumptions before drawing conclusions, despite the fact that such assumptions are necessary for learning anything about the real world.

You also don't seem to like the implications of there being practical constraints on what experiments we can do. Since it is physically impossible to test every possible theory, we have to have some way to decide which experiments are worth doing. And because presumably the goal of science is to learn, we should prefer those experiments with a high expected return of information compared to the time and effort involved in running the experiment. For most scientists, the experiments likely to provide the most information are generally the most interesting to do. The theories these experiments test between really are the most interesting theories for those scientists.
Unless stated otherwise, I do not care whether a statement, by itself, constitutes a persuasive political argument. I care whether it's true.
---
If this post has math that doesn't work for you, use TeX the World for Firefox or Chrome

(he/him/his)

J Thomas
Everyone's a jerk. You. Me. This Jerk.^
Posts: 1190
Joined: Fri Sep 23, 2011 3:18 pm UTC

Re: 1063: "Kill Hitler"

Postby J Thomas » Wed Jun 13, 2012 5:58 pm UTC

curtis95112 wrote:
Say that you are in a sealed laboratory, and you have a neutrino detector. Somebody brings a nuclear reactor close to the outside of your lab and you can detect it. They move the reactor far away and you can detect that. In theory, if you had a way to seal your lab so neutrinos couldn't get in then your neutrino detector would get the same result regardless. But in practice you can't do that. So in reality, conservation of linear momentum fails because you can detect neutrinos, and conservation of angular momentum fails because you can detect which direction they come from.

If the most widely believed theory of neutrinos is correct, then this is just a quibble. All it takes is to put your lab behind a lead wall a million meters thick, and the neutrinos would be blocked. And it makes some sense to believe that, because the number of neutrinos detected varies with the time of day which might mean that the bulk of the earth (8000 miles thick at maximum) affects neutrinos from the sun. With enough data about neutrino detection day and night across the seasons, you could infer the rate of neutrinos from the sky in all directions apart from the sun. All this is reasonable, and it's a collection of reasonable assumptions piled on each other. It could still be wrong.


I don't understand what you're trying to say. Why do the conservation laws fail in your scenario? Could you explain?


It's a trick. Conservation laws say you get the same result from your experiment in your sealed lab independent of location. But this experiment gets different results in different locations. If you are close to a reactor you get different results. And we don't know an effective way to shield for that. You can use a faraday cage to mostly shield for electromagnetic things. You can use a few meters of lead to mostly shield from gamma rays. You can put the lab in free fall to shield most of the effect of gravity, though you might detect some tidal effects inside your lab. But you can't do a whole lot to shield against neutrinos.

One more time, it's two interpretations that both fit all the available facts. People prefer one, but both are tentatively true at this point. There are reactions in a nuclear reactor for which various things that are supposed to be conserved, disappear. There are other reactions that happen more when you're closer to a nuclear reactor, where the things that are supposed to be conserved appear out of nowhere. You can say that the things which disappear or appear are carried around by invisible elves, or you can say they disappear and appear. You get the same result either way.

If it's conservation laws violated then the same experiment will not get the same result in different places. Namely, the same experiment will get a different result close to a nuclear reactor. The reactor somehow changes the probability of some events happening. But if it's invisible elves that carry stuff around then the conservation laws are not violated after all because you have an explanation why the same experiment gets a different result. Close to the reactor there are more invisible elves to carry stuff around.

Both explanations work, both get the same result, so you can choose whichever you want. Or don't choose.
The Law of Fives is true. I see it everywhere I look for it.

User avatar
gmalivuk
GNU Terry Pratchett
Posts: 26767
Joined: Wed Feb 28, 2007 6:02 pm UTC
Location: Here and There
Contact:

Re: 1063: "Kill Hitler"

Postby gmalivuk » Wed Jun 13, 2012 6:02 pm UTC

J Thomas wrote:Conservation laws say you get the same result from your experiment in your sealed lab independent of location. But this experiment gets different results in different locations.
No, this experiment just hasn't ever been done in a sufficiently sealed lab.

Both explanations work, both get the same result, so you can choose whichever you want. Or don't choose.
Unless you are a scientist interested in theories capable of making predictions. In that case, you should choose the explanation that involves the least additional assumptions and most efficiently makes the best predictions about future observations.

Thus, the theory that neutrinos exist as subatomic particles, instead of either the theory that energy is carried around by elves or the theory that energy disappears and then systematically and predictably reappears elsewhere later but was never actually carried between those two points by any particle.
Unless stated otherwise, I do not care whether a statement, by itself, constitutes a persuasive political argument. I care whether it's true.
---
If this post has math that doesn't work for you, use TeX the World for Firefox or Chrome

(he/him/his)

J Thomas
Everyone's a jerk. You. Me. This Jerk.^
Posts: 1190
Joined: Fri Sep 23, 2011 3:18 pm UTC

Re: 1063: "Kill Hitler"

Postby J Thomas » Wed Jun 13, 2012 7:11 pm UTC

gmalivuk wrote:
J Thomas wrote:
Both explanations work, both get the same result, so you can choose whichever you want. Or don't choose.
Unless you are a scientist interested in theories capable of making predictions. In that case, you should choose the explanation that involves the least additional assumptions and most efficiently makes the best predictions about future observations.


If you're an engineer who needs predicted values, then you probably want the math that gives you the simplest calculations.

If you're a scientist, then it's good to look at the explanations that encourage the most interesting experiments.

If you're an educator, then you probably want the single explanation that's easiest to explain to people who have the usual preconceptions.

Thus, the theory that neutrinos exist as subatomic particles, instead of either the theory that energy is carried around by elves or the theory that energy disappears and then systematically and predictably reappears elsewhere later but was never actually carried between those two points by any particle.
[/quote]

See, deciding that it's subatomic particles is more additional assumption than noticing that energy disappears one place and predictably reappears elsewhere, since the latter is what is observed. But assuming it's subatomic particles might suggest experiments that the simpler assumption does not. Considering both might help you find experiments that either alone would not.

If you're a teacher then the subatomic particle idea gives you something concrete to teach that fits people's preconceptions, so that's good.

And if you're an engineer then it doesn't matter what the theory is, and currently it doesn't matter about the experimental results either -- you won't need it. If you did need it, what counts is how to calculate it and not what stories people tell about it.

Damn, I responded to you again. But it wasn't a bad exchange this time.
The Law of Fives is true. I see it everywhere I look for it.

User avatar
gmalivuk
GNU Terry Pratchett
Posts: 26767
Joined: Wed Feb 28, 2007 6:02 pm UTC
Location: Here and There
Contact:

Re: 1063: "Kill Hitler"

Postby gmalivuk » Wed Jun 13, 2012 8:09 pm UTC

J Thomas wrote:If you're a scientist, then it's good to look at the explanations that encourage the most interesting experiments.

If you're an educator, then you probably want the single explanation that's easiest to explain to people who have the usual preconceptions.
No, if you're a scientist you *also* want to be able to effectively calculate predicted values, because otherwise you have nothing to compare future observations to. Or to put it another way, the most interesting experiments are the ones where you can say something sensible about the results, which means they're the ones whose results are comparable to your predictions, which means you have to have calculated some predictions ahead of time. Granted, your *full* theory might take a lot longer to give you predictions than the simplified-for-engineering version, and so you might use that version instead of the full one in practice. But even the possibility of having a simplified version in the first place isn't something that all potential theories have, and so you'll preferentially use the ones that do whenever you want to run an actual experiment.

And if you're a teacher, you want the explanation that is most like the one scientists and/or engineers actually use or, for a lower level class, the explanation scientists used to use because it was generally good enough, leaving the newest theories for more advanced classes.

See, deciding that it's subatomic particles is more additional assumption than noticing that energy disappears one place and predictably reappears elsewhere
No it isn't, because the additional assumptions have to include everything that logically follows from the explanation. Which in this case means assuming, against all prior results, that energy actually vanishes from the universe for a time, and furthermore assuming that the appearance of time-symmetrical laws was only ever a coincidence, and so on.

But assuming it's subatomic particles might suggest experiments that the simpler assumption does not.
This of course being the other important bit.

After all, *everything* we have ever observed throughout the history of the universe (along with every imaginable set of alternative observations we might have made instead) is perfectly consistent with the claim that the universe is completely and fundamentally random and we just happen to live in one of the exceptional bits where it looks like it isn't. The problem with that explanation is that it makes no predictions and is completely untestable in principle, and is therefore completely worthless to anyone hoping to do science or otherwise learn anything. That doesn't mean it's not true, but assuming at any point that it is true means giving up on the possibility of any further knowledge (or any knowledge in the first place) about the universe.

Conveniently, every set of possible observations is *also* consistent with (an infinite number of) explanations which *aren't* fundamentally arbitrary, and some of them are even testable and simple enough for us to calculate predictions and effectively run experiments in reasonable amounts of time. So those are the explanations we preferentially stick closest to.

Considering both might help you find experiments that either alone would not.
Which experiments I still contend are not worth doing when one of the considered hypotheses is too improbable.
Unless stated otherwise, I do not care whether a statement, by itself, constitutes a persuasive political argument. I care whether it's true.
---
If this post has math that doesn't work for you, use TeX the World for Firefox or Chrome

(he/him/his)

User avatar
PM 2Ring
Posts: 3713
Joined: Mon Jan 26, 2009 3:19 pm UTC
Location: Sydney, Australia

Re: 1063: "Kill Hitler"

Postby PM 2Ring » Thu Jun 14, 2012 12:56 pm UTC

J Thomas wrote:Say that you are in a sealed laboratory, and you have a neutrino detector. Somebody brings a nuclear reactor close to the outside of your lab and you can detect it. They move the reactor far away and you can detect that. In theory, if you had a way to seal your lab so neutrinos couldn't get in then your neutrino detector would get the same result regardless. But in practice you can't do that. So in reality, conservation of linear momentum fails because you can detect neutrinos, and conservation of angular momentum fails because you can detect which direction they come from.

If the most widely believed theory of neutrinos is correct, then this is just a quibble. All it takes is to put your lab behind a lead wall a million meters thick, and the neutrinos would be blocked.


Ah, but bringing a nuclear reactor close to the outside of my imperfectly sealed lab violates the spirit of the sealed lab thought experiment. You can either bring the reactor into the lab, or we wrap the lab in a few cubic light-years of lead and situate it in an intergalactic void far from neutrino sources. The point being that if there are possible reactions that violate time, space, or orientation symmetries, then experiments can be performed in a totally sealed lab that will reveal these broken symmetries and hence can (in principle) be used to measure the lab's location in spacetime and its orientation without having to make any kind of observation of the outside universe.

I can certainly understand the esthetic value of data compression. That's one of the big virtues of poetry. With poetry you can say powerful things in a few words, and the compression is so valuable that it's worth losing precision. In physics often you can have concision and precision both -- you can express fundamental rules in a little space without being wrong. Of course, in most real situations there are endless complications because there's so much going on that interacts in complex ways, but all that complexity could all come from the simple rules, provided you have all the simple rules worked out correctly. But what if the reality turns out to be a little more complicated? Then you really ought to accept reality rather than insist on an artificial simplicity.


Even a very simple rule set soon leads to intractable complexity. Just look at Conway's Life. It's easy to create patterns in Life whose ultimate history is currently unpredictable, patterns whose initial state is small enough to fit entirely on a standard computer screen. The rules of particle physics are much more complicated than those of Life, and it's not easy to make accurate calculations / predictions from first principles. Quantum electrodynamics isn't too bad, but it's still not easy to predict the behaviour of big atoms using QED, and modeling large molecules accurately that way is extremely hard - simplifying approximations have to be made in order to perform useful calculations. The situation in QCD is much worse - just predicting the approximate mass of the proton from the rules governing quarks was a major effort involving a lot of supercomputer time, according to Frank Wilczek.

The upshot of this is that nobody in the field believes that the ultimate goal of attempting to find the simplest possible rule set governing the behaviour of matter and energy is that we'll be able to predict everything from first principles. All we want is to have a set of rules that can be tested empirically and that cover everything that we can observe.

You figure that if conservation laws fail then there must be a preferred inertial frame. Doesn't that depend on the pattern in which they fail? I have the idea they could fail in a way that wouldn't single out one inertial frame as the preferred one, though I don't have an example in mind right now that I'm certain would work. I could be wrong. If I'm right then your idea that if we give up anything we lose it all, is an example of a pitfall with the concise physical theory approach. When it all seems to fit together too precisely, when it seems like a perfect jewel, then it's only natural to believe that losing any single part would result in chaos, the jewel must be smashed to powder. But remove one part and there might be ways to salvage a lot of the rest. Perhaps it could fit together as a different jewel.


If the conservation laws fail in some convenient pattern, then we just reformulate them in a way that takes this pattern into account and we've got a new system with a new invariant that replaces the old one with something a little subtler. Essentially, that's what happened when we discovered radioactivity and merged conservation of mass with conservation of energy to get conservation of mass-energy.

The theory that posits (anti)neutrinos carry stuff to or from from weak interaction events makes the bookkeeping much simpler than a theory without neutrinos. :) One important quantity that neutrinos carry is intrinsic angular momentum (aka spin). They don't have a lot of energy, and they carry no electromagnetic charge, but they have the same amount of spin as an electron or a quark. ( Interestingly, all neutrinos have the same helicity (left-handed), with antineutrinos (of course) having the opposite helicity, leading to the suggestion that the neutrino is its own antiparticle. ) So even if you do come up with a decent theory of weak interactions that does away with the neutrino that somehow explains the tiny amount of energy and momentum currently assigned to the neutrino, you still have to explain what happens to the spin, which is rather significant.

My hope is that the more different theories you can entertain, the better you may think of interesting experiments.


I don't think that anyone in this thread disagrees that there is some value in competition between theories. However, since there are so many potential alternative hypotheses we need to have some way of measuring the relative usefulness of competing theories. Sure, some potentially useful theories may get neglected for philosophical reasons, or even political reasons. But if a theory does have some utility, eventually someone will get around to putting it through its paces. However, some theories have very low utility, since they are extremely impractical to test and/or have very little of use to tell us about the universe (eg the blue bottle theory mentioned earlier).

I've read this whole thread (although I will admit that I did skim some of the posts), but I'm not sure if you really disbelieve in neutrinos, or if you're just playing Devil's Advocate. FWIW, according to Quantum Field Theory, all subatomic particles are convenient fictions: the reality of matter and energy is the fields themselves. Particles nicely encapsulate the quantized nature of quantum fields, and they allow us to make much simpler calculations at the cost of losing some accuracy. Fortunately, the inaccuracies only become significant in non-linear interactions, and can safely be neglected in many practical situations. I'm certainly not an expert in QFT, OTOH, we do have a few people over on the Science forum that work in QFT, including one regular poster who's currently working on a PhD project that involves QFT in curved spacetime. I'm sure he'd happily agree that all particles are BS, but they are very handy BS, since they work so well. Most of the time. When the particle model fails, it does so abysmally. :)

User avatar
eran_rathan
Mostly Wrong
Posts: 1842
Joined: Fri Apr 09, 2010 2:36 pm UTC
Location: in your ceiling, judging you

Re: 1063: "Kill Hitler"

Postby eran_rathan » Thu Jun 14, 2012 2:52 pm UTC

J Thomas wrote:
chenille wrote:What idea have you proposed, J Thomas?


I propose that neutrinos are a good candidate for a science fad, a concept that is taken up without adequate consideration and deeply held beyond the available evidence. They have a number of the hallmarks of that. It's possible that if I learned enough of the details I would think otherwise. On the other hand, if I'm right then I would have little to gain by learning a whole lot of the details -- even if I found where they went wrong and proposed a better alternative, I would be ignored.


have you ever looked at IceCube?

http://icecube.wisc.edu/

if the observed evidence fits the predictions, it is ok to assume that the model is generally correct, until you find something that doesn't work - that's the basis of science.

Yes, we know the Standard Model is incomplete. Yes, there are issues with it - but that doesn't mean the whole thing is wrong (until we find out it is :D)
"Does this smell like chloroform to you?"
"Google tells me you are not unique. You are, however, wrong."
nɒʜƚɒɿ_nɒɿɘ

ewindisch
Posts: 1
Joined: Thu Jun 14, 2012 3:18 pm UTC

Re: 1063: "Kill Hitler"

Postby ewindisch » Thu Jun 14, 2012 3:23 pm UTC

Perhaps someone else has mentioned it in the past 7 pages, and I simply missed it...

He didn't have to use the time machine twice. He might have simply returned to the machine after killing Hitler and before his younger self entered.

J Thomas
Everyone's a jerk. You. Me. This Jerk.^
Posts: 1190
Joined: Fri Sep 23, 2011 3:18 pm UTC

Re: 1063: "Kill Hitler"

Postby J Thomas » Thu Jun 14, 2012 4:37 pm UTC

PM 2Ring wrote:
J Thomas wrote:Say that you are in a sealed laboratory, and you have a neutrino detector. Somebody brings a nuclear reactor close to the outside of your lab and you can detect it. They move the reactor far away and you can detect that. In theory, if you had a way to seal your lab so neutrinos couldn't get in then your neutrino detector would get the same result regardless. But in practice you can't do that. So in reality, conservation of linear momentum fails because you can detect neutrinos, and conservation of angular momentum fails because you can detect which direction they come from.

If the most widely believed theory of neutrinos is correct, then this is just a quibble. All it takes is to put your lab behind a lead wall a million meters thick, and the neutrinos would be blocked.


Ah, but bringing a nuclear reactor close to the outside of my imperfectly sealed lab violates the spirit of the sealed lab thought experiment. You can either bring the reactor into the lab, or we wrap the lab in a few cubic light-years of lead and situate it in an intergalactic void far from neutrino sources. The point being that if there are possible reactions that violate time, space, or orientation symmetries, then experiments can be performed in a totally sealed lab that will reveal these broken symmetries and hence can (in principle) be used to measure the lab's location in spacetime and its orientation without having to make any kind of observation of the outside universe.


Yes? And is there a way to tell whether there are experiments that could be performed in that theoretical condition? Suppose for a minute that it's true that the reactions which are thought to produce neutrinos actually produce violations of conservation laws. Is there a way to go from there to measure the lab's location in spacetime?

I can certainly understand the esthetic value of data compression. That's one of the big virtues of poetry. With poetry you can say powerful things in a few words, and the compression is so valuable that it's worth losing precision. In physics often you can have concision and precision both -- you can express fundamental rules in a little space without being wrong. Of course, in most real situations there are endless complications because there's so much going on that interacts in complex ways, but all that complexity could all come from the simple rules, provided you have all the simple rules worked out correctly. But what if the reality turns out to be a little more complicated? Then you really ought to accept reality rather than insist on an artificial simplicity.


Even a very simple rule set soon leads to intractable complexity. Just look at Conway's Life. It's easy to create patterns in Life whose ultimate history is currently unpredictable, patterns whose initial state is small enough to fit entirely on a standard computer screen. The rules of particle physics are much more complicated than those of Life, and it's not easy to make accurate calculations / predictions from first principles. Quantum electrodynamics isn't too bad, but it's still not easy to predict the behaviour of big atoms using QED, and modeling large molecules accurately that way is extremely hard - simplifying approximations have to be made in order to perform useful calculations. The situation in QCD is much worse - just predicting the approximate mass of the proton from the rules governing quarks was a major effort involving a lot of supercomputer time, according to Frank Wilczek.

The upshot of this is that nobody in the field believes that the ultimate goal of attempting to find the simplest possible rule set governing the behaviour of matter and energy is that we'll be able to predict everything from first principles. All we want is to have a set of rules that can be tested empirically and that cover everything that we can observe.


Yes. So there will be important practical things which can't be predicted, but you want a simple collection of rules which are compatible with everything you observe. Somehow you want to divide up what happens into fundamental things, which the rules can predict, and composite things which can get complicated.

You figure that if conservation laws fail then there must be a preferred inertial frame. Doesn't that depend on the pattern in which they fail? I have the idea they could fail in a way that wouldn't single out one inertial frame as the preferred one, though I don't have an example in mind right now that I'm certain would work. I could be wrong. If I'm right then your idea that if we give up anything we lose it all, is an example of a pitfall with the concise physical theory approach. When it all seems to fit together too precisely, when it seems like a perfect jewel, then it's only natural to believe that losing any single part would result in chaos, the jewel must be smashed to powder. But remove one part and there might be ways to salvage a lot of the rest. Perhaps it could fit together as a different jewel.


If the conservation laws fail in some convenient pattern, then we just reformulate them in a way that takes this pattern into account and we've got a new system with a new invariant that replaces the old one with something a little subtler. Essentially, that's what happened when we discovered radioactivity and merged conservation of mass with conservation of energy to get conservation of mass-energy.


Yes, exactly!

The theory that posits (anti)neutrinos carry stuff to or from from weak interaction events makes the bookkeeping much simpler than a theory without neutrinos. :) One important quantity that neutrinos carry is intrinsic angular momentum (aka spin). They don't have a lot of energy, and they carry no electromagnetic charge, but they have the same amount of spin as an electron or a quark. ( Interestingly, all neutrinos have the same helicity (left-handed), with antineutrinos (of course) having the opposite helicity, leading to the suggestion that the neutrino is its own antiparticle. ) So even if you do come up with a decent theory of weak interactions that does away with the neutrino that somehow explains the tiny amount of energy and momentum currently assigned to the neutrino, you still have to explain what happens to the spin, which is rather significant.


Yes. It isn't clear how simple you can make a theory without neutrinos until you actually make it, though.

My hope is that the more different theories you can entertain, the better you may think of interesting experiments.


I don't think that anyone in this thread disagrees that there is some value in competition between theories. However, since there are so many potential alternative hypotheses we need to have some way of measuring the relative usefulness of competing theories. Sure, some potentially useful theories may get neglected for philosophical reasons, or even political reasons. But if a theory does have some utility, eventually someone will get around to putting it through its paces. However, some theories have very low utility, since they are extremely impractical to test and/or have very little of use to tell us about the universe (eg the blue bottle theory mentioned earlier).


It appears to me that several people in this thread have extolled the value of having one theory which is better than any alternative, and argued that no alternative is worth considering because the one good one is good enough. But let's put that aside. Theories that are impractical to test are not worth much, but you don't find out how impractical they are to test until you consider them. And one thing about new experiments -- when an experiment is designed well, often it gives the opportunity to do more than just choose between two hypotheses. It gives nature the opportunity to show you things that would not be expected from either of the hypotheses being considered. That's valuable!

I've read this whole thread (although I will admit that I did skim some of the posts), but I'm not sure if you really disbelieve in neutrinos, or if you're just playing Devil's Advocate.


I'm an agnostic about neutrinos. I don't say I know they don't exist. I don't say I know they do. I say they are a theoretical construct which fits the known data, and it's valuable to find alternative constructs too.

FWIW, according to Quantum Field Theory, all subatomic particles are convenient fictions: the reality of matter and energy is the fields themselves. Particles nicely encapsulate the quantized nature of quantum fields, and they allow us to make much simpler calculations at the cost of losing some accuracy. Fortunately, the inaccuracies only become significant in non-linear interactions, and can safely be neglected in many practical situations. I'm certainly not an expert in QFT, OTOH, we do have a few people over on the Science forum that work in QFT, including one regular poster who's currently working on a PhD project that involves QFT in curved spacetime. I'm sure he'd happily agree that all particles are BS, but they are very handy BS, since they work so well. Most of the time. When the particle model fails, it does so abysmally. :)


I'm glad somebody is following that approach.

I was once part of an amateur effort to rethink classical physics, using some ideas that are available now that were not available then. The idea was to create a variety of ideas and see which could be refuted classically. So for example there was the idea that perhaps protons and neutrons etc were made of combinations of electrons and positrons, the only elementary "particles". Then mass could be just from charge -- a proton with a whole lot of particles that attract and repel, but one extra that attracts would have much more inertia. The masses don't come out quite even, but then they don't come out quite even for nuclei made of protons and neutrons, either. We didn't disprove that with the data we allowed ourselves.

We started out thinking of electrons as tiny blobs, but they are now considered to have something that's analogous to physical spin, and what does that mean for a blob that's symmetric in all dimensions? I proposed an alternative, that an electron might be composed of a central positive charge with two negative charges rotating around it at a fixed distance. A bar-shaped electron could spin meaningfully, and that gave one asymmetric model to use as an alternative that was fairly easy to play with. I'd make other asymmetric models when we figured out why that one couldn't work.

Classical magnetism didn't make sense for particles. The traditional formula depended on absolute motion by sender and receiver both; whether there was magnetism or not depended on the velocity of the observer and the total amount of force varied accordingly.Absurd. We didn't work out an analogous force using only relative velocity, but it was the obviously-correct way to go.

It was fun, but of course if you got a better classical physics all you'd have would be a better classical physics. But I think trying it out gives some sense of how the process works. And so I notice that operationally "quarks" describe a pattern in the mass etc for a collection of hadrons. It's convenient to think of them as particles that make up the hadrons, but what's observed is the pattern. It's like being in the position of Mendeleev and inventing the periodic table of elements. The periodic table was periodic, and explanations why that was the case needed to come later. Quarks might be real, but if there's some other explanation it may not come to people who assume it's quarks.
The Law of Fives is true. I see it everywhere I look for it.

J Thomas
Everyone's a jerk. You. Me. This Jerk.^
Posts: 1190
Joined: Fri Sep 23, 2011 3:18 pm UTC

Re: 1063: "Kill Hitler"

Postby J Thomas » Thu Jun 14, 2012 4:45 pm UTC

eran_rathan wrote:
J Thomas wrote:
chenille wrote:What idea have you proposed, J Thomas?


I propose that neutrinos are a good candidate for a science fad, a concept that is taken up without adequate consideration and deeply held beyond the available evidence. They have a number of the hallmarks of that. It's possible that if I learned enough of the details I would think otherwise. On the other hand, if I'm right then I would have little to gain by learning a whole lot of the details -- even if I found where they went wrong and proposed a better alternative, I would be ignored.


have you ever looked at IceCube?

http://icecube.wisc.edu/

if the observed evidence fits the predictions, it is ok to assume that the model is generally correct, until you find something that doesn't work - that's the basis of science.

Yes, we know the Standard Model is incomplete. Yes, there are issues with it - but that doesn't mean the whole thing is wrong (until we find out it is :D)


I wouldn't say the Standard Model is wrong unless there was evidence to show it was wrong.
I say that when you pay attention to only a single model you are not likely to do much to advance science -- unless you accidentally find evidence that it is wrong. When you can think of only a single model you are not likely to find evidence that it's wrong on purpose, because you think it's right and you will have no idea what tests it may fail.

But I'm stating that too strongly. There's the possibility you might try to apply the theory to new areas where it has never been tested, and observe it to fail there. And confirming that it does not fail in those areas is also a contribution.
The Law of Fives is true. I see it everywhere I look for it.

chenille
Posts: 430
Joined: Tue Dec 20, 2011 7:25 pm UTC

Re: 1063: "Kill Hitler"

Postby chenille » Thu Jun 14, 2012 5:42 pm UTC

J Thomas wrote:We started out thinking of electrons as tiny blobs, but they are now considered to have something that's analogous to physical spin, and what does that mean for a blob that's symmetric in all dimensions?

Just as a matter of interest, spin works because they're not. A vector field isn't symmetric because it has a distinct direction; well, a spinor does too, so an electron simply looks different from different angles. For orbitals angular momentum has always had to do with the shape of the waves around a point, rather than something actually spinning over time, so this really is an analogous concept. I mention this just because I think it's something a lot of people overlook.

J Thomas wrote:And so I notice that operationally "quarks" describe a pattern in the mass etc for a collection of hadrons. It's convenient to think of them as particles that make up the hadrons, but what's observed is the pattern.

I don't think this is true. My understanding was that quarks were only accepted because of scattering experiments, which gave results much more consistent with the idea that protons and neutrons are genuinely composite particles. Some people have proposed similar sub-particles to explain the diversity of leptons and quarks, but in the absence of anything similar, that doesn't seem to have any advantage over the standard model.

J Thomas
Everyone's a jerk. You. Me. This Jerk.^
Posts: 1190
Joined: Fri Sep 23, 2011 3:18 pm UTC

Re: 1063: "Kill Hitler"

Postby J Thomas » Thu Jun 14, 2012 7:26 pm UTC

chenille wrote:
J Thomas wrote:We started out thinking of electrons as tiny blobs, but they are now considered to have something that's analogous to physical spin, and what does that mean for a blob that's symmetric in all dimensions?

Just as a matter of interest, spin works because they're not. A vector field isn't symmetric because it has a distinct direction; well, a spinor does too, so an electron simply looks different from different angles. For orbitals angular momentum has always had to do with the shape of the waves around a point, rather than something actually spinning over time, so this really is an analogous concept. I mention this just because I think it's something a lot of people overlook.


Thank you! I hadn't heard that explained so clearly.

J Thomas wrote:And so I notice that operationally "quarks" describe a pattern in the mass etc for a collection of hadrons. It's convenient to think of them as particles that make up the hadrons, but what's observed is the pattern.

I don't think this is true. My understanding was that quarks were only accepted because of scattering experiments, which gave results much more consistent with the idea that protons and neutrons are genuinely composite particles. Some people have proposed similar sub-particles to explain the diversity of leptons and quarks, but in the absence of anything similar, that doesn't seem to have any advantage over the standard model.


For a long time there has been reason to think that protons and neutrons might be composite particles. If a neutron separates into other particles, do you suppose it used to be a single undifferentiated particle that changed into other particles, or that it was composite all along? Either interpretation used to work, of course. Atomic nuclei can toss out electrons or positrons, either one. Were they there all along or did they get created just in time to get thrown out?

What you'd need would be scattering experiments that are more compatible with the components being quarks than combinations of other particles.

Without knowing many details, I kind of like the idea of a whole lot of little particles involved. So they could interact more in atomic nuclei. When you have a nucleus with 6 protons and 6 neutrons, that's a powerful repulsion that needs a very strong force to hold it together. When you have some large X number of positive charges and X-6 negative charges, it isn't so many. They could form a crystal (8:8 coordinated if they're close to the same size) with the extra charges separated around the outside. As the crystal gets larger the excess charges get farther away, but then at still larger sizes the surface increases by the square while the volume increases by the cube, so they start getting closer again. Instead of such a large ad hoc strong force to hold them together, you're left trying to explain why such a limited variety of structures are stable, and why they break apart in such a limited number of patterns, and you need some sort of local repulsive force or excluded volume. But I heard there were experiments that indicated atomic nuclei do include some sort of large chunks that might correspond to protons and neutrons and alpha particles, themselves in a sort of crystal. So I like the idea of a crystal with many small components, but I expect it would fail.
The Law of Fives is true. I see it everywhere I look for it.

PFD Studio
Posts: 11
Joined: Mon Sep 27, 2010 2:05 pm UTC
Contact:

Re: 1063: "Kill Hitler"

Postby PFD Studio » Thu Jun 21, 2012 2:43 pm UTC

I think if someone actually managed to travel back and kill Hitler, the results would be more like this:

http://www.techcurmudgeon.com/2012/06/time-after-time.html

User avatar
Max™
Posts: 1792
Joined: Thu Jun 21, 2012 4:21 am UTC
Location: mu

Re: 1063: "Kill Hitler"

Postby Max™ » Wed Jun 27, 2012 6:04 am UTC

PFD Studio wrote:I think if someone actually managed to travel back and kill Hitler, the results would be more like this:

http://www.techcurmudgeon.com/2012/06/t ... -time.html

Love the comic, and I was going to make a joke about plotting a closed timelike curve in a Xeelee nightfighter (i.e. my avatar) here, but an older me showed up and said it didn't go over so well.
mu

DesertPlah2
Posts: 1
Joined: Mon Feb 04, 2013 4:31 pm UTC

Re: 1063: "Kill Hitler"

Postby DesertPlah2 » Mon Feb 04, 2013 4:36 pm UTC

Hey, check out over on Cracked their version of Randall's joke. "How not to kill Hitler"

User avatar
addams
Posts: 10271
Joined: Sun Sep 12, 2010 4:44 am UTC
Location: Oregon Coast: 97444

Re: 1063: "Kill Hitler"

Postby addams » Tue Feb 05, 2013 5:48 am UTC

Djehutynakht wrote:Well, if we get into time travel theory we could say this is a self-fufilling sort of thing. Hitler didn't really die by suicide. Hitler died because BHG went back in time and killed him. It was always meant to be that way. Forever and forever.

That leaves the question of the others though, such as Eva Braun. Did she kill herself before BHG arrived? After? Did BHG just kill her too? Was it entirely unrelated? Was Hitler planning to kill himself before BHG showed up? Would he have ever planned so if he didn't? Could he have had a secret way to come back and win the war if BHG hadn't killed him?

The world may never know.


On the other hand, I notice we're treating a time machine like a teleportation device again (I'm assuming that BHG appeared inside Hitler's bunker, as opposed to going through all of the trouble to get from... wherever he is to Hitlers bunker right before Hitler's death). In that case, even if the time travel part is only one use, can he still use it as a teleporter in the present?


Actually, you know what, we don't know anything.

For all we know, BHG went back in time, killed the struggling artist known as Adolf Hitler, assumed his identity, and was in fact, while disguised as Hitler, the person responsible for creating the Third Reich and starting WWII, only to hide his time machine in his bunker in order to escape there and fake his own death after having caused as much destruction and mayhem as physically possible.

Yeah, I could see him doing that.

Or; Not.

Kill Hitler? That is dark humor. Right?
Hitler: The most evil person, ever. Is that the way you learned it? That is not the way I learned it.

Now Black Hat Guy can replace him. Fine.
Black Hat Guy can take it.
An Evil Hero?
For a fictional 2D creature, it is also charming in its own way.
I was won over when he 'Keyed' her car.
Life is, just, an exchange of electrons; It is up to us to give it meaning.

We are all in The Gutter.
Some of us see The Gutter.
Some of us see The Stars.
by mr. Oscar Wilde.

Those that want to Know; Know.
Those that do not Know; Don't tell them.
They do terrible things to people that Tell Them.

User avatar
PM 2Ring
Posts: 3713
Joined: Mon Jan 26, 2009 3:19 pm UTC
Location: Sydney, Australia

Re: 1063: "Kill Hitler"

Postby PM 2Ring » Tue Feb 05, 2013 6:12 am UTC

addams wrote:Now Black Hat Guy can replace him. Fine.
Black Hat Guy can take it.
An Evil Hero?
For a fictional 2D creature, it is also charming in its own way.
I was won over when he 'Keyed' her car.


What? That was Beret Guy, not Black Hat Guy.
http://xkcd.com/1030/

User avatar
addams
Posts: 10271
Joined: Sun Sep 12, 2010 4:44 am UTC
Location: Oregon Coast: 97444

Re: 1063: "Kill Hitler"

Postby addams » Tue Feb 05, 2013 6:19 am UTC

PM 2Ring wrote:
addams wrote:Now Black Hat Guy can replace him. Fine.
Black Hat Guy can take it.
An Evil Hero?
For a fictional 2D creature, it is also charming in its own way.
I was won over when he 'Keyed' her car.


What? That was Beret Guy, not Black Hat Guy.
http://xkcd.com/1030/

Oops. Beret Guy is charming.
Black Hat Guy, Is! Hitler?!
oh. dear.
Life is, just, an exchange of electrons; It is up to us to give it meaning.

We are all in The Gutter.
Some of us see The Gutter.
Some of us see The Stars.
by mr. Oscar Wilde.

Those that want to Know; Know.
Those that do not Know; Don't tell them.
They do terrible things to people that Tell Them.


Return to “Individual XKCD Comic Threads”

Who is online

Users browsing this forum: No registered users and 50 guests