Proof of Immortality, VII

Status
Not open for further replies.
Dave,
- Do you think that ~H is a reasonable possibility?

One of the problems with your approach is taking subjective judgements like "reasonable possibility" and "unimaginably small" and trying to substitute numerical equivalents.
Dave,
- Sorry. Somehow, I missed your response.
- That's an unfortunate aspect of this particular issue. My claim is that if we can accept that 10-100 s not too large an estimate for the likelihood of my current existence -- given OOFLam -- the prior probability of OOFLam could be .999999999 (say), and if I'm not 'sharpshooting,' the posterior probability of OOFLam would be no greater than 10-80 (or, thereabouts).
 
Dave,
- Sorry. Somehow, I missed your response.
- That's an unfortunate aspect of this particular issue. My claim is that if we can accept that 10-100 s not too large an estimate for the likelihood of my current existence -- given OOFLam -- the prior probability of OOFLam could be .999999999 (say), and if I'm not 'sharpshooting,' the posterior probability of OOFLam would be no greater than 10-80 (or, thereabouts).

This is worthless because those numbers aren't justified.

Even if those numbers were justified it wouldn't make any sense, and would suggest that you are using Bayes's theorem wrong.
 
Sorry. Somehow, I missed your response.

You miss a lot of responses and you're not sorry about any of them. That's too bad, because a lot of people are giving you a lot of important feedback. It's up to you to choose whether you're more interested in fooling people into thinking you're right, or actually being right.

That's an unfortunate aspect of this particular issue.

No, it's an intractable aspect of your argument. You're literally claiming that you can just pull a bunch of random numbers out of your nether regions, run them through a formula, and suddenly have proven something momentous. Bayes' theorem is not a Magical Fact Creator.

My claim is that if we can accept...

Your claim simply repeats what godless dave rightly accused you of. You're asking people to agree in their heart-of-hearts that you've supplied your model with suitable data, in lieu of proving an actual proof or testable rationale for it. I guess by repeating your claim you've conceded that Dave is right. You also go on to emphasize that because of how you've loaded your likelihood ratio (with completely made-up numbers), the posterior probabilities have no reasonable dependence on their priors. You seem to think it's a feature of your model that it inevitably produces the answer you need regardless of inputs. It isn't; that's a bug.
 
Last edited:
Dave,
- Sorry. Somehow, I missed your response.
- That's an unfortunate aspect of this particular issue. My claim is that if we can accept that 10-100 s not too large an estimate for the likelihood of my current existence -- given OOFLam -- the prior probability of OOFLam could be .999999999 (say), and if I'm not 'sharpshooting,' the posterior probability of OOFLam would be no greater than 10-80 (or, thereabouts).

What is your estimate for the likelihood of a particular snowflake landing in a particular spot on Mount Kilimanjaro?
 
My claim is that if we can accept that 10-100 s not too large an estimate for the likelihood of my current existence -- given OOFLam -- the prior probability of OOFLam could be .999999999 (say), and if I'm not 'sharpshooting,' the posterior probability of OOFLam would be no greater than 10-80 (or, thereabouts).

Ok, support your claim, then.
 
Dave,
- Sorry. Somehow, I missed your response.
- That's an unfortunate aspect of this particular issue. My claim is that if we can accept that 10-100 s not too large an estimate for the likelihood of my current existence -- given OOFLam -- the prior probability of OOFLam could be .999999999 (say), and if I'm not 'sharpshooting,' the posterior probability of OOFLam would be no greater than 10-80 (or, thereabouts).
Why are you still arguing about the numbers when it's been explained to you so many times why they don't matter?

For the numbers to matter it would be necessary to demonstrate that the Texas Sharpshooter fallacy is not a fallacy. We know you can't do that so there is no point in trying to establish meaningful estimates for the numbers, even if that was possible.
 
Thermal,

- Keep in mind that "likelihood" and "probability," as they are used in Bayesian statistics, are not the same.
.

Keep in mind that the statisticians you took your proof to soon realised that you either didn't know how to apply Bayes or that you were being deliberately obtuse. Then, rather than be schooled in how to properly apply the theorem, you ran away.

All documented here
 
Why are you still arguing about the numbers when it's been explained to you so many times why they don't matter?

For the numbers to matter it would be necessary to demonstrate that the Texas Sharpshooter fallacy is not a fallacy.

Because Jabba can't or won't understand it. So, given that he can't or won't understand it, there is no reason to perpetuate the thread.
 
Dave,
- Sorry. Somehow, I missed your response.
- That's an unfortunate aspect of this particular issue. My claim is that if we can accept that 10-100 s not too large an estimate for the likelihood of my current existence -- given OOFLam -- the prior probability of OOFLam could be .999999999 (say), and if I'm not 'sharpshooting,' the posterior probability of OOFLam would be no greater than 10-80 (or, thereabouts).

This is worthless because those numbers aren't justified.

Even if those numbers were justified it wouldn't make any sense, and would suggest that you are using Bayes's theorem wrong.
Dave,
- Sounds like I'll have to leave it there.
- So, this illustrates two basic disagreements between us:
1. Whether or not Bayesian statistics actually applies to my usage.
2. Whether or not my numbers should be taken seriously.

- I'll be reviewing our past conversation for other unresolved disagreements between us, but you can probably provide your remaining reservations faster than can I. I doubt that I can be much more convincing than I already have without allowing for another year or two of debate -- and, probably not then either...
- Once I feel like I've given my best shots to your other reservations, I'll take what's left of my syllogism (assuming that I still think I'm right) to at least one statistics site, and see what happens.
 
Well...yeah. If you throw billions of packs of cards in the air, some will land in a perfect straight flush. Some of those cards landed in such a way as to produce the moon and Jabba. And in the vastness of time and space, that happened. Are you are suggesting that it was so unlikely that..something...must have had a hand in making sure those things all happened? If not, I don't see why it should seem so improbable.

Eta: and if we knew enough about those cards, their weight and other forces acting on them, it could be predict where and how they would land right out of the gate. That doesn't mean that the straight flushes are so staggeringly unlikely; they are inevitable
Thermal,
- I doubt that I can do any better at quashing your reservations than I already have...
 
- So, this illustrates two basic disagreements between us:

The only reason why there's still disagreement is that you refuse to listen to your critics.

- I'll be reviewing our past conversation for other unresolved disagreements between us, but you can probably provide your remaining reservations faster than can I.

No, I don't think you can still reasonably expect others to do your work for you, since you have provided ample evidence that you are quite able to do so yourself.
 
Thermal,
- I doubt that I can do any better at quashing your reservations than I already have...

Yes, the English language lacks the capacity to make fundamentally stupid ideas less dumb merely by rewording them. Still, you'll get your reward in imaginary heaven.
 
So, this illustrates two basic disagreements between us....

"Disagreement" implies there is no error on your part. There is, and you are simply unwilling to bear the consequences of it.

Whether or not Bayesian statistics actually applies to my usage.

There is no question on this point. Every single qualified statistician you've consulted has told you that you are wrong. We've explained in exactly what way you are wrong. You're simply uninterested in that information.

Whether or not my numbers should be taken seriously.

Your model is broken at a fundamental level, and you've already admitted you cannot redeem it. In addition, you have simply made up all the "data" you propose to feed that model. Why should you expect that to be taken seriously? It's a joke from beginning to end, and you don't seem to realize that everyone but you can see this.

I'll be reviewing our past conversation for other unresolved disagreements between us, but you can probably provide your remaining reservations faster than can I.

Oh, please. You were given a comprehensive list of your errors. You pretended for six months that it didn't exist and then finally admitted you couldn't answer it. It's far past the time when you can convince anyone you're serious about addressing errors in your proof. You've ever only wanted people to give you a round of applause for allegedly being so clever.

I doubt that I can be much more convincing than I already have without allowing for another year or two of debate -- and, probably not then either...

The reason you aren't convincing is because you're simply wrong in ways that are exceptionally easy for knowledgeable people to discern. The people whose approval you've sought in the past were not very discriminating and let you get away with unchallenged nonsense. You keep admitting defeat in this roundabout way but decline to own it. After five years of peddling the same nonsense and expecting people just to roll over and let you keep committing these simple errors, it's time to admit the problem is with you, not with them. But no, your history shows you won't do this. You're only admitting tactical defeat now in order to soften criticism. I guarantee it won't last.

...to at least one statistics site, and see what happens.

You already did that, and you haven't changed your basic argument or strategy since you did. What did they tell you? They told you that you were wrong and explained why, just as we have. Not only that, they told you that you were delusional to the point of "reinvent[ing] probability theory" to make your theory work. They further reached the same conclusion as we have: that you are entirely impervious to criticism and seem just to want to hear yourself talk and be praised for it.

The problem is not your audience. The problem is that you are objectively wrong. You are wrong in easily discerned, elementary ways, not in some esoteric nuance that you can convince a crowd of sycophants is just the braying of skeptics.

I doubt that I can do any better at quashing your reservations than I already have...

Then you agree that you have lost the debate and that your proof fails.
 
Dave,
- Sorry. Somehow, I missed your response.
- That's an unfortunate aspect of this particular issue. My claim is that if we can accept that 10-100 s not too large an estimate for the likelihood of my current existence -- given OOFLam -- the prior probability of OOFLam could be .999999999 (say), and if I'm not 'sharpshooting,' the posterior probability of OOFLam would be no greater than 10-80 (or, thereabouts).

All the initial numbers in your equation should be 42. That will get you the most accurate results.
 
Thermal,
- I doubt that I can do any better at quashing your reservations than I already have...

XeQvjZk.jpg
 
I think you're showing progress and applaud your admission of defeat. That you can admit that your ideas are indefensible is laudable.

Don't be too hasty. He does this every two or three months. "I guess I just can't convince you people" is what he's saying, and subsequent discussion reveals that what he really means is that we skeptics are just too hopelessly entrenched in our limited modes of thinking to appreciate his special genius. Just wait and see. If I'm wrong, I owe you 10100 fake internet points.
 
Status
Not open for further replies.

Back
Top Bottom