🚀 go-pugleaf

"Sylvia Else" <sylvia@not.here.invalid> wrote 
> On 29/09/2010 3:34 PM, |-|ercules wrote:
>> "Sylvia Else" <sylvia@not.here.invalid> wrote
>>> On 29/09/2010 2:43 PM, |-|ercules wrote:
>>>> "Sylvia Else" <sylvia@not.here.invalid> wrote...
>>>>> On 29/09/2010 12:48 PM, |-|ercules wrote:
>>>>>> "Sylvia Else" <sylvia@not.here.invalid> wrote ...
>>>>>>> On 29/09/2010 12:14 PM, |-|ercules wrote:
>>>>>>>> You've probably heard Sylvia Else backing down from doing an
>>>>>>>> objective test of my mind reading powers, saying again and again
>>>>>>>> that a negative result doesn't prove I DON'T have powers so what is
>>>>>>>> the point?
>>>>>>>
>>>>>>> We started an objective test, but you bailed out after just two
>>>>>>> results were in, since they were not to your liking.
>>>>>>
>>>>>>
>>>>>> I "bailed" on one particular protocol that we introduced mid way
>>>>>> through
>>>>>> one test,
>>>>>> by suggesting a better way of indexing the questions.
>>>>>>
>>>>>> ME : suggest continue testing - make 1 change due to 1 failed result
>>>>>> YOU : 500 excuses to not test = BAIL
>>>>>>
>>>>>> I have enough statistical evidence from 0/6 questions right to know
>>>>>> when
>>>>>> to try something different.
>>>>>>
>>>>>>
>>>>>>>
>>>>>>> It's a no-win situation for me. I fully expect you to fail any test.
>>>>>>> But you won't then accept that you don't have the power. Instead, at
>>>>>>> best you'll want to start the whole laborious process again with some
>>>>>>> variant of the protocol. At worst, and on the evidence, most
>>>>>>> probably,
>>>>>>> you'll accuse me of subverting the protocol in some way so as to
>>>>>>> prevent you from succeeding. You'll also engage in various forms of
>>>>>>> abuse. There are plenty of examples of that stored in Google's Usenet
>>>>>>> archive.
>>>>>>>
>>>>>>> In our recent discussions, it's become clear that you believe your
>>>>>>> power draws a distinction between what people think and what they do,
>>>>>>> being sensitive to the former rather than the latter. Pending the
>>>>>>> construction of real mind-reading machines, this makes your power
>>>>>>> completely immune to any kind of disproof.
>>>>>>>
>>>>>>> In such a situation, why would I, or anyone else, bother? Might as
>>>>>>> well be hanged for a sheep as for a lamb.
>>>>>>>
>>>>>>> Sylvia.
>>>>>>
>>>>>>
>>>>>> Because if you make an offer to test people you have to test people. I
>>>>>> don't
>>>>>> mind you being gutless and seeing nothing in it for you and
>>>>>> expecting me
>>>>>> to fail,
>>>>>> but Peter thinking your scientific methodology is correct protocol
>>>>>> is a
>>>>>> laugh!
>>>>>>
>>>>>> Herc
>>>>>>
>>>>>
>>>>> On the ASS web site, it says
>>>>>
>>>>> "The test must give a definite yes or no result."
>>>>>
>>>>> http://www.skeptics.com.au/features/prize/
>>>>>
>>>>> You want people to accept a test that says "yes" or "don't know". Such
>>>>> tests don't qualify.
>>>>>
>>>>> Sylvia.
>>>>
>>>>
>>>> That's per trial you idiot.
>>>
>>> It says "The test", not "Each test", indicating that there's only one
>>> answer, which is either "yes" or "no".
>>>
>>>>
>>>> You can't disprove a power unless the claim is 100% hit rate.
>>>
>>> You can reach a required probability that the failure is not by
>>> chance, just as there's a required probability that the success is not
>>> by chance (and heaven help me if I've just pointed out a wriggle that
>>> Herc hadn't previously noticed).
>>>
>>> The wording chosen was "yes or no", not "probably or probably not",
>>> but it is clear from the rest of the content that the "yes" answer is
>>> to be determined on the basis that there's a less than one in million
>>> chance that it's achieved by chance. I see no reason to think that the
>>> "no" answer would be considered to be absolute.
>>>
>>> Sylvia.
>>
>>
>> I have no idea why you think I am interested in this line of reasoning,
>> if you need to justify why you bailed on me then maybe you should
>> explain it to yourself.
> 
> I strongly doubted that you were interested, but you're claiming that 
> ASS are reneging on their own protocol. I disagree, so I'm voicing my 
> opinion, which is that they're not, because you're claiming that you 
> possess a power that is not falsifiable, and the ASS requirement that a 
> "no" result be possible means that they require that claimed powers be 
> falsifiable.
> 
> If you don't like it, try for a prize where that's not a requirement.
>>
>> The statement of detecting powers only to be tested is quite clear.
> 
> That sentence certainly isn't.
> 
> Sylvia.

It says objective pass / fail test is required.

If I break 1,000,000:1 AND above my "double expected average" bench mark
then it's a pass.

If not it's a fail.

The maximum number of trials is the LEAST amount of trials to break 1,000,000:1
at my set bench mark ratio of hits.

Maybe this is what you've been going on about all week, which should be evident
and obvious for all binomial rated trials.

I'm not claiming to break 1,000,000:1 by getting average results for thousands of trials
but eventually scoring really high bringing the odds back up.  Minor minor minor caveat
about the binomial odds stated which is trivially implied, if that's what your problem was.

You are making the most inane argument I have every come across,
do all people claiming to break 1,000,000:1 have to define basic common
sense testing protocol like this?

Herc