# Dear [Chat]y

## Posts

• rRegistered User regular
edited September 2010
Arch wrote: »
Winky wrote: »
Winky wrote: »
Winky wrote: »
I was presented with a problem.

There are infinite potential hypotheses. If you falsify a set of these hypotheses, any arbitrarily large set, you still are left with infinite potential hypotheses. So how can you be getting any closer to the truth by doing so?

I'm thinking the answer has to do with bounding the set of hypotheses.

Are you looking for a practical answer like "In the process of eliminating false hypotheses we also gain information with which to start predicting the likelihood of remaining hypothesis to be true"?

I'm looking for it in more abstract terms. Note that I don't actually know if there's an answer here. This was a problem my philosophy of science teacher posed to me, I'm don't know of anyone who has written on this specific subject.

But I am somewhat interested in this concept of cumulative data shaping future hypotheses, because that does put a hard limit on which hypotheses can be posed. Like I said, bounding, there's still an infinite number of potential hypotheses, but only within parameters.

I'd also like to see how 'infinite potential hypotheses' works, unless you're doing something silly like counting duplicates or allowing for an infinite amount of information in a system.

There might be a seemingly infinite amount of hay in a stack, but combing through it still gets us closer to the needle.

He demonstrated to me that there were infinite hypotheses. He drew three points on a board and said that these were three recorded positions of Jupiter over the horizon (for an example, maybe you'd like something harder to measure like the position of subatomic particles or something) at different time intervals. Say you are formulating a hypothesis to account for the motion that Jupiter traveled to be at those three points at those points in time. You can make an infinite number of paths between those three points, so there are infinite possible hypotheses that fit to that data.

but if you know about the movement of jupiter in the past (I.E., you have data collected on the movement of jupiter) you can begin to eliminate the paths that don't make sense

Right, but for any set of data there is still infinite paths.

Winky on
• Registered User regular
edited September 2010
so if you managed to kill your grandfather you can be pretty sure someone lied to you about your family tree, that the neighbour fucked your grandmother, or that you got the date wrong

Abdhyius on
• Registered User regular
edited September 2010
It's not that we haven't observed causality being violated.

It's more like we have no memory of observing causality being violated.

Isn't that right, Josef Stalin III, Prime Minister of the United Kingdom of Great Britain and the Calais Fiefdom?

RMS Oceanic on
• Registered User regular
edited September 2010
Chanus wrote: »
And Earthworm Jim taught us of the speed of lint.

God those games were fun.

Inquisitor on
• I've seen things... Registered User regular
edited September 2010
Chanus wrote: »
If time travel exists then it's inconsequential and anything that can be changed has already been changed.

If time travel doesn't exist then it's inconsequential because nothing can be changed.

So there is no causality, then. We're all just drifting along like a train car on a rail.

its irrelevant.

if time travel exists and you can go back to change what has happened then it stands to reason that those people who wanted to go back and fix something have already done so or have already created a universe destroying time paradox in trying to do so.

the fact that you have free will to choose does not mean that your choice wasn't already made by you in the cyclical future/past that is effecting the timeline you're in. you just don't know about it yet because you haven't done it yet.

This would only apply to a single universe theory, of course.

Chanus on
Allegedly a voice of reason.
• Comedy Gold Registered User regular
edited September 2010
Abdy closed time loops, thus a constant repeating pattern actually ends up being impossible according to current theories. Here, if you like time travel theory, read this article. It is really interesting. Now off to my one day of work a week.

http://discovermagazine.com/2010/mar/02-the-real-rules-for-time-travelers/?searchterm=time%20travel

Mazzyx on

• Neat-o, mosquito! Registered User regular
edited September 2010
Winky wrote: »
Arch wrote: »
Winky wrote: »
Winky wrote: »
Winky wrote: »
I was presented with a problem.

There are infinite potential hypotheses. If you falsify a set of these hypotheses, any arbitrarily large set, you still are left with infinite potential hypotheses. So how can you be getting any closer to the truth by doing so?

I'm thinking the answer has to do with bounding the set of hypotheses.

Are you looking for a practical answer like "In the process of eliminating false hypotheses we also gain information with which to start predicting the likelihood of remaining hypothesis to be true"?

I'm looking for it in more abstract terms. Note that I don't actually know if there's an answer here. This was a problem my philosophy of science teacher posed to me, I'm don't know of anyone who has written on this specific subject.

But I am somewhat interested in this concept of cumulative data shaping future hypotheses, because that does put a hard limit on which hypotheses can be posed. Like I said, bounding, there's still an infinite number of potential hypotheses, but only within parameters.

I'd also like to see how 'infinite potential hypotheses' works, unless you're doing something silly like counting duplicates or allowing for an infinite amount of information in a system.

There might be a seemingly infinite amount of hay in a stack, but combing through it still gets us closer to the needle.

He demonstrated to me that there were infinite hypotheses. He drew three points on a board and said that these were three recorded positions of Jupiter over the horizon (for an example, maybe you'd like something harder to measure like the position of subatomic particles or something) at different time intervals. Say you are formulating a hypothesis to account for the motion that Jupiter traveled to be at those three points at those points in time. You can make an infinite number of paths between those three points, so there are infinite possible hypotheses that fit to that data.

but if you know about the movement of jupiter in the past (I.E., you have data collected on the movement of jupiter) you can begin to eliminate the paths that don't make sense

Right, but for any set of data there is still infinite paths.

But that doesn't change the fact that you can perform further tests and begin to falsify some of these infinite paths, and begin to arrive at the path that, in spite of the myriad of infinite possibilities, is the one that is MOST accurate.

It may not be the literally correct path, but it will be the path that makes the most 'sense'.

This is how chemistry and particle physics work

Arch on
• I've seen things... Registered User regular
edited September 2010
Abdhyius wrote: »
so if you managed to kill your grandfather you can be pretty sure someone lied to you about your family tree, that the neighbour fucked your grandmother, or that you got the date wrong

Sure, a lesson in not changing history from Mr. I'm My Own Grandpa!

Chanus on
Allegedly a voice of reason.
• Registered User regular
edited September 2010
Look, can you guys stop gaying up my [chat] thread with all your science?

Geeeeeeeeeze.

Inquisitor on
• We are only now beginning to understand the full power and ramifications of sexual intercourse Registered User regular
edited September 2010
Inquisitor wrote: »
Look, can you guys stop gaying up my [chat] thread with all your science?

Geeeeeeeeeze.

Mojo_Jojo on
Homogeneous distribution of your varieties of amuse-gueule
• Registered User regular
edited September 2010
Mojo_Jojo wrote: »
Abdhyius wrote: »
do we have any proof that causality can not be violated?

Other than we haven't seen it happen yet?

I suppose we have that antiparticles can just be considered particles that go in the other direction (of time). When you draw Feynman diagrams you end up with arrows in both directions, and then some lines without arrows.

In a way, there's causation, but it doesn't always go the same way in time. And you can have little loops.

I've always thought the assumption that "A happens which causes B to happen" is the only way that it works to be weird since it really has no basis

"A caused B to happen, then A happened" requires more mental gymnastics, but it could work

Abdhyius on
• Registered User regular
edited September 2010
Inquisitor wrote: »
Look, can you guys stop gaying up my [chat] thread with all your science?

Geeeeeeeeeze.

Gay science should totally be its own discipline.

Psycho Internet Hawk on
• Registered User regular
edited September 2010
Chanus wrote: »
Chanus wrote: »
If time travel exists then it's inconsequential and anything that can be changed has already been changed.

If time travel doesn't exist then it's inconsequential because nothing can be changed.

So there is no causality, then. We're all just drifting along like a train car on a rail.

its irrelevant.

if time travel exists and you can go back to change what has happened then it stands to reason that those people who wanted to go back and fix something have already done so or have already created a universe destroying time paradox in trying to do so.

the fact that you have free will to choose does not mean that your choice wasn't already made by you in the cyclical future/past that is effecting the timeline you're in. you just don't know about it yet because you haven't done it yet.

This would only apply to a single universe theory, of course.

if there are multiple universes then nothing you do really matters because you did the opposite in some other universe.

if there are multiple parallel universes then the multiverse has too many backups and needs a better data storage plan.

• Registered User regular
edited September 2010
time travel in this fashion would mostly be an exercise in removing the illusion of free will and thus pretty depressing

Abdhyius on
• Neat-o, mosquito! Registered User regular
edited September 2010
For instance, Winky, there are an infinite number of positions that an electron can occupy in an orbital (You know this).

But that doesn't mean that some of those infinite positions make sense, or occur often enough to be relevant.

The electrons in a Carbon atom MAY be located miles away, but it is far more likely that they will follow the shape of the sp orbital, and even though there are still infinite numbers of possibilities, we can use the data we have collected on the movement of electrons (either through direct testing or mathematical modeling) to make accurate predictions on the location of these particles, even if we can't pin them down to a concrete location.

Arch on
• I've seen things... Registered User regular
edited September 2010
Abdhyius wrote: »
Mojo_Jojo wrote: »
Abdhyius wrote: »
do we have any proof that causality can not be violated?

Other than we haven't seen it happen yet?

I suppose we have that antiparticles can just be considered particles that go in the other direction (of time). When you draw Feynman diagrams you end up with arrows in both directions, and then some lines without arrows.

In a way, there's causation, but it doesn't always go the same way in time. And you can have little loops.

I've always thought the assumption that "A happens which causes B to happen" is the only way that it works to be weird since it really has no basis

"A caused B to happen, then A happened" requires more mental gymnastics, but it could work

Chicken. Egg. Spaceships.

Chanus on
Allegedly a voice of reason.
• We are only now beginning to understand the full power and ramifications of sexual intercourse Registered User regular
edited September 2010
Abdhyius wrote: »
Mojo_Jojo wrote: »
Abdhyius wrote: »
do we have any proof that causality can not be violated?

Other than we haven't seen it happen yet?

I suppose we have that antiparticles can just be considered particles that go in the other direction (of time). When you draw Feynman diagrams you end up with arrows in both directions, and then some lines without arrows.

In a way, there's causation, but it doesn't always go the same way in time. And you can have little loops.

I've always thought the assumption that "A happens which causes B to happen" is the only way that it works to be weird since it really has no basis

"A caused B to happen, then A happened" requires more mental gymnastics, but it could work

It should be noted that my "and you can have little loops" doesn't really refer to loops in time so much as little self-contained casual bubbles of particles magicing themselves into existence, interacting a bit and then fucking off.

Mojo_Jojo on
Homogeneous distribution of your varieties of amuse-gueule
• Is a woman dammit ceterum censeoRegistered User regular
edited September 2010
If you have infinite data would you only have one possible hypothesis?

Elldren on
fuck gendered marketing
• What a lovely day for tea Secret British ThreadRegistered User regular
edited September 2010
Fart

Zen Vulgarity on
• I've seen things... Registered User regular
edited September 2010
Chanus wrote: »
Chanus wrote: »
If time travel exists then it's inconsequential and anything that can be changed has already been changed.

If time travel doesn't exist then it's inconsequential because nothing can be changed.

So there is no causality, then. We're all just drifting along like a train car on a rail.

its irrelevant.

if time travel exists and you can go back to change what has happened then it stands to reason that those people who wanted to go back and fix something have already done so or have already created a universe destroying time paradox in trying to do so.

the fact that you have free will to choose does not mean that your choice wasn't already made by you in the cyclical future/past that is effecting the timeline you're in. you just don't know about it yet because you haven't done it yet.

This would only apply to a single universe theory, of course.

if there are multiple universes then nothing you do really matters because you did the opposite in some other universe.

if there are multiple parallel universes then the multiverse has too many backups and needs a better data storage plan.

RAID-Infinity backup server.

Chanus on
Allegedly a voice of reason.
• on my way to work in a suit and a tie Ahhhh...come on fucking guyRegistered User regular
edited September 2010
Inquisitor wrote: »
Look, can you guys stop gaying up my [chat] thread with all your science?

Geeeeeeeeeze.

Gay science should totally be its own discipline.

Gay science is like regular science, but the atomic orbitals are FABULOUS instead of unpredictable.

Deebaser on
• Neat-o, mosquito! Registered User regular
edited September 2010
Elldren wrote: »
If you have infinite data would you only have one possible hypothesis?

I don't think so, but let me give it some thought.

Arch on
• Registered User regular
edited September 2010
Elldren wrote: »
If you have infinite data would you only have one possible hypothesis?

no, you would have a null hypothesis as well.

• Neat-o, mosquito! Registered User regular
edited September 2010
Deebaser wrote: »
Inquisitor wrote: »
Look, can you guys stop gaying up my [chat] thread with all your science?

Geeeeeeeeeze.

Gay science should totally be its own discipline.

Gay science is like regular science, but the atomic orbitals are FABULOUS instead of unpredictable.

This is wonderful.

Arch on
• Registered User regular
edited September 2010
Deebaser wrote: »
Inquisitor wrote: »
Look, can you guys stop gaying up my [chat] thread with all your science?

Geeeeeeeeeze.

Gay science should totally be its own discipline.

Gay science is like regular science, but the atomic orbitals are FABULOUS instead of unpredictable.

Finally, an excuse to build a large hardon collider.

Psycho Internet Hawk on
• Registered User regular
edited September 2010
question: why can't logical paradoxes occur?

I mean, why can't event A cause event A to happen?

Abdhyius on
• rRegistered User regular
edited September 2010
Arch wrote: »
For instance, Winky, there are an infinite number of positions that an electron can occupy in an orbital (You know this).

But that doesn't mean that some of those infinite positions make sense, or occur often enough to be relevant.

The electrons in a Carbon atom MAY be located miles away, but it is far more likely that they will follow the shape of the sp orbital, and even though there are still infinite numbers of possibilities, we can use the data we have collected on the movement of electrons (either through direct testing or mathematical modeling) to make accurate predictions on the location of these particles, even if we can't pin them down to a concrete location.

Yes, I like this explanation.

As I initially suspected, it has to do with bounding.
Elldren wrote: »
If you have infinite data would you only have one possible hypothesis?

Interesting question.

Winky on
• I've seen things... Registered User regular
edited September 2010
Chanus wrote: »
Abdhyius wrote: »
so if you managed to kill your grandfather you can be pretty sure someone lied to you about your family tree, that the neighbour fucked your grandmother, or that you got the date wrong

Sure, a lesson in not changing history from Mr. I'm My Own Grandpa!

Come on! This was good!

Chanus on
Allegedly a voice of reason.
• Registered User, __BANNED USERS regular
edited September 2010
Hm. I feel pretty much the opposite.

Seems the only logical way to think to me.

Everything that can be or could be, is.

It's certainly logical. But my hangup is that if it isn't some how reflected in the overall theme of the story, then what's the point in telling the story? The protagonist didn't prevent the catastrophe from happening, he just prevented it from affecting his universe.

Basically, it boils down to Ocarina of Time, and how it handled it. Big turn off.

I guess if I read a story and it was done well then I would okay with the concept. But at face value, it's too big of a disconnect for me.

Sheep on
• Neat-o, mosquito! Registered User regular
edited September 2010
Elldren wrote: »
If you have infinite data would you only have one possible hypothesis?

no, you would have a null hypothesis as well.

I think this is my answer as well

Arch on
• I've seen things... Registered User regular
edited September 2010
Abdhyius wrote: »
question: why can't logical paradoxes occur?

I mean, why can't event A cause event A to happen?

Because if science doesn't actually mean anything there can only be God.

Chanus on
Allegedly a voice of reason.
• Is a woman dammit ceterum censeoRegistered User regular
edited September 2010
Arch, what are you doing today?

Elldren on
fuck gendered marketing
• Neat-o, mosquito! Registered User regular
edited September 2010
Sheep wrote: »
Hm. I feel pretty much the opposite.

Seems the only logical way to think to me.

Everything that can be or could be, is.

It's certainly logical. But my hangup is that if it isn't some how reflected in the overall theme of the story, then what's the point in telling the story? The protagonist didn't prevent the catastrophe from happening, he just prevented it from affecting his universe.

Basically, it boils down to Ocarina of Time, and how it handled it. Big turn off.

I guess if I read a story and it was done well then I would okay with the concept. But at face value, it's too big of a disconnect for me.

Well, technically his universe is the only one we are currently interested in.

seems like a slide into nihilism to say that he "didn't prevent it from happening, only prevented it in HIS universe"

I mean if the catastrophe spans ALL universes, then stopping it in one should stop it in all, but generally they are not that large

Okay I lied this is Sci Fi, sometimes they are that large

Arch on
• Neat-o, mosquito! Registered User regular
edited September 2010
Elldren wrote: »
Arch, what are you doing today?

Wasting time before a meeting because the website I was supposed to update apparently doesn't need to be updated because they are hiring someone to directly manage the websites.

Thank god. Or I guess science.

I should stop saying "thank god"

Arch on
• rRegistered User regular
edited September 2010
Arch wrote: »
Elldren wrote: »
If you have infinite data would you only have one possible hypothesis?

no, you would have a null hypothesis as well.

I think this is my answer as well

I'm not so sure about that. How can there be a null hypothesis with infinite data? Your hypothesis cannot possibly be false.

Winky on
• Registered User regular
edited September 2010
Mazzyx wrote: »
Abdy closed time loops, thus a constant repeating pattern actually ends up being impossible according to current theories. Here, if you like time travel theory, read this article. It is really interesting. Now off to my one day of work a week.

http://discovermagazine.com/2010/mar/02-the-real-rules-for-time-travelers/?searchterm=time%20travel

this is super interesting, thanks

also, great game idea in that article
To illustrate this point, imagine that you stumble upon a time machine in the form of a gate. When you pass through it in one direction, it takes you exactly one day into the past; if you pass through in the other direction, it takes you exactly one day into the future. You walk up to the gate, where you see an older version of yourself waiting for you. The two of you exchange pleasantries. Then you leave your other self behind as you walk through the gate into yesterday. But instead of obstinately wandering off, you wait around a day to meet up with the younger version of yourself (you have now aged into the older version you saw the day before) with whom you exchange pleasantries before going on your way. Everyone’s version of every event would be completely consistent.

We can have much more dramatic stories that are nevertheless consistent. Imagine that we have been appointed Guardian of the Gate, and our job is to keep vigilant watch over who passes through. One day, as we are standing off to the side, we see a person walk out of the rear side of the gate, emerging from one day in the future. That’s no surprise; it just means that you will see that person enter the front side of the gate tomorrow. But as you keep watch, you notice that he simply loiters around for one day, and when precisely 24 hours have passed, the traveler walks calmly through the front of the gate. Nobody ever approached from elsewhere. That 24-hour period constitutes the entire life span of this time traveler. He experiences the same thing over and over again, although he doesn’t realize it himself, since he does not accumulate new memories along the way. Every trip through the gate is precisely the same to him. That may strike you as weird or unlikely, but there is nothing paradoxical or logically inconsistent about it.

Dialog-based game. It would be awesome.

Abdhyius on
• What a lovely day for tea Secret British ThreadRegistered User regular
edited September 2010
Chat's too fast again

Will check back later

Zen Vulgarity on
• Not at all confident in his reflexive opinions of thingsRegistered User regular
edited September 2010
Winky wrote: »
Arch wrote: »
Winky wrote: »
Winky wrote: »
Winky wrote: »
I was presented with a problem.

There are infinite potential hypotheses. If you falsify a set of these hypotheses, any arbitrarily large set, you still are left with infinite potential hypotheses. So how can you be getting any closer to the truth by doing so?

I'm thinking the answer has to do with bounding the set of hypotheses.

Are you looking for a practical answer like "In the process of eliminating false hypotheses we also gain information with which to start predicting the likelihood of remaining hypothesis to be true"?

I'm looking for it in more abstract terms. Note that I don't actually know if there's an answer here. This was a problem my philosophy of science teacher posed to me, I'm don't know of anyone who has written on this specific subject.

But I am somewhat interested in this concept of cumulative data shaping future hypotheses, because that does put a hard limit on which hypotheses can be posed. Like I said, bounding, there's still an infinite number of potential hypotheses, but only within parameters.

I'd also like to see how 'infinite potential hypotheses' works, unless you're doing something silly like counting duplicates or allowing for an infinite amount of information in a system.

There might be a seemingly infinite amount of hay in a stack, but combing through it still gets us closer to the needle.

He demonstrated to me that there were infinite hypotheses. He drew three points on a board and said that these were three recorded positions of Jupiter over the horizon (for an example, maybe you'd like something harder to measure like the position of subatomic particles or something) at different time intervals. Say you are formulating a hypothesis to account for the motion that Jupiter traveled to be at those three points at those points in time. You can make an infinite number of paths between those three points, so there are infinite possible hypotheses that fit to that data.

but if you know about the movement of jupiter in the past (I.E., you have data collected on the movement of jupiter) you can begin to eliminate the paths that don't make sense

Right, but for any set of data there is still infinite paths.

I think this is only true to a degree; you can have two paths which appear the same except for a misplaced molecule, or even an atom, or even a subatomic particle. Eventually you can't get any more granular and you're tracing the same path even with theoretical precision, regardless of the fact that it stopped being relevant long ago.

TL DR on
• Neat-o, mosquito! Registered User regular
edited September 2010
Winky wrote: »
Arch wrote: »
Elldren wrote: »
If you have infinite data would you only have one possible hypothesis?

no, you would have a null hypothesis as well.

I think this is my answer as well

I'm not so sure about that. How can there be a null hypothesis with infinite data? Your hypothesis cannot possibly be false.

Infinite data does not mean it is infinite correct data, or data that has any meaning.

I can have infinite data on the movement of an insect, but that doesn't mean I can then use that infinite data to formulate or falsify a hypothesis on the effect of a certain chemical on a cell.

Arch on
• Registered User regular
edited September 2010
Chanus wrote: »
Abdhyius wrote: »
question: why can't logical paradoxes occur?

I mean, why can't event A cause event A to happen?