Paulhoff
You can't expect perfection.
- Joined
- May 1, 2005
- Messages
- 12,512
Then I’m either right or wrong aren’t I. Simple as that.
If they insist on me scanning the other 9 I can but what would be the reason for taking longer?
Then I fail, why would I want to test the rest?
Why would they? The next trial Can begin.
That saves time remember I’m looking for the gold when I find it I have accomplished the mission.
It doesn’t really matter as long as I can take notes from the scale.
As long as they pass on one spot I can dowse each of the containers, that won’t fool me.
I’m not opposed to scanning all of them but what would be the reason if I pick the target?
My way would be quicker, how much I don’t know because it’s random.
Yes I know that’s why I figured this could be beneficial.
Yes isn’t it amazing and the parallel to that is hundreds of years of using it, that dowsing seems to work perfectly well in the field, If I had not gotten such good numbers while mining I would have to think that you are right but seeing is believing.
Especially standing on ground that has been worked to death by other miners and dowsers, that mine.
The task in these times is enormous even at the scale and especially on the scale of a recreational miner.
Now a days you can go broke on the cost of fuel alone.
Then patejdl says,
I'm handsome, I drive an 80's black porsche, run a comfortably successfull company, have a very good software soon to be released so yeah I think you are old
You should learn how to use spell check or Word.
And boy aren’t we full of ourselves?
Some one who runs a company should be able to spell.
What are orbs, is that what you mean?
Or, do you even know what an orb is?
I don’t give a crap what they are right now, Mr. Porsche. With a capital P.
I got two big ones.
It was over your head.
Last edited:
SezMe
post-pre-born
There is a very simple protocol that avoids cannisters, locations, targets, the scale, calibration, etc. Think about the tool (i.e., the dowsing rod) not the target.
Fix the target to be something that edge's rod responds to dramatically. Say, for example, an ingot of gold borrowed from the Franklin Mint.* Set it out in plain sight and leave it there for all to see. Let edge dowse for it and show that his rod damn near breaks as it jerks into the vertical position. Remember, edge, like it did near the creek? Surely a gold ingot would cause the same reaction.
Now take the dime off the end of edge's rod and let him dowse for the ingot. Nothing will happen.
Now, each trial consists of putting a dime or a wooden substitute dime on the head of the dowsing rod in a way that edge cannot tell which it is (this would be very simple). Now let edge dowse the ingot. The rod will either snap downward instantly or it will do nothing. Each trial can be done in a minute or two with the dowsing in each trial limited to 1-2 seconds. This eliminates the fatigue factor on edge's part. And it will be obvious to anyone what the result is.
That said, I'm kinda with Reno here. I don't believe any test is going to take place and all the effort to get an airtight protocol here is just wasted time. I think you, edge, should quit screwing around here and get down to serious discussions with Randi. All this dithering here is just giving you an excuse for not getting on with getting on.
* I'm just exaggerating with the gold ingot. But any large chunk of metal that edge's dowsing rod reacts dramatically to will do.
Fix the target to be something that edge's rod responds to dramatically. Say, for example, an ingot of gold borrowed from the Franklin Mint.* Set it out in plain sight and leave it there for all to see. Let edge dowse for it and show that his rod damn near breaks as it jerks into the vertical position. Remember, edge, like it did near the creek? Surely a gold ingot would cause the same reaction.
Now take the dime off the end of edge's rod and let him dowse for the ingot. Nothing will happen.
Now, each trial consists of putting a dime or a wooden substitute dime on the head of the dowsing rod in a way that edge cannot tell which it is (this would be very simple). Now let edge dowse the ingot. The rod will either snap downward instantly or it will do nothing. Each trial can be done in a minute or two with the dowsing in each trial limited to 1-2 seconds. This eliminates the fatigue factor on edge's part. And it will be obvious to anyone what the result is.
That said, I'm kinda with Reno here. I don't believe any test is going to take place and all the effort to get an airtight protocol here is just wasted time. I think you, edge, should quit screwing around here and get down to serious discussions with Randi. All this dithering here is just giving you an excuse for not getting on with getting on.
* I'm just exaggerating with the gold ingot. But any large chunk of metal that edge's dowsing rod reacts dramatically to will do.
EHocking
Penultimate Amazing
Have you asked JREF if they would be willing to conduct the test in this manner?
EHocking
Penultimate Amazing
Have you asked JREF if they will accept that a trial is "complete" when you have chosen the gold target, and do not wish to dowse the rest of the 10 canisters?
Let me narrow it down for as I give them the option either way to save time.
I’m sure they are beyond that and have asked the last question which had to do with the score deriving from picking out only the metals.
You said save time there it is!
Protocal states,
They surly must understand this as I understand it even more, in this way I save my energy, and can go for a longer set of tests.
Only you are Questioning it.
Oualawouzou
Critical Thinker
- Joined
- Feb 25, 2007
- Messages
- 490
Quoting this as I don't think Edge has mentionned already weither or not it would be an acceptable protocol.
EHocking
Penultimate Amazing
I am not questioning the method. I asked if JREF have/will accept it.
If they will, I am willing to write a clear and concise, air tight, double-blind protocol for you to present to them so that you can get on with lining up a time and place with them.
I am NOT questioning you - I am offering to help you with the protocol process.
The questions I have been asking you for the past week or so have been to make sure that the protocol I present to you satisfies *your* conditions and is fair to *your* claim.
But it's not worth my while if the protocol will be rejected by JREF.
I'm one question away from writing this up for you - and that is.
Is JREF willing to accept this last condition from you?
If you don't want this help - say so now and I'll remove myself from this discussion.
EHocking
Penultimate Amazing
If edge can get JREF to agree to his last condition (that is, once he's picked the gold target, that set of 10 is "complete" and he waivers the need to test the remainder), his original claim of 70% success in 10 trials may be practicable.
If they reject his suggestion, then 1 in 2 trial, with it's required higher success rate is the only reasonable solution.
Just waiting for edge to confirm JREF's acceptance/rejection of his latest condition.
Or edge telling me he doesn't want help to draw up a protocol.
It's up to the JREF to accept or not and so far they haven't responded yet.
I wrote them yesterday,
Hi Alison,
I assume that was the last problem that you had.
I sent an answer.
When you agree to a protocol we can get on with the
next part which is time, location and place.
Mike
I wouldn't exclude my protocol yet, till I get an answer.
The way I will do the test isn't much different than the original design of the test, and if they want to take the time I will scan all of the containers.
My choice will still be my choice.
In other words EHocking wait and see.
My protocol is air tight for a true exhibition of dowsing and I will even if i don't pass get numbers that are higher than their 10% that they expect from all dowsing demonstrations.
I wrote them yesterday,
Hi Alison,
I assume that was the last problem that you had.
I sent an answer.
When you agree to a protocol we can get on with the
next part which is time, location and place.
Mike
I wouldn't exclude my protocol yet, till I get an answer.
The way I will do the test isn't much different than the original design of the test, and if they want to take the time I will scan all of the containers.
My choice will still be my choice.
In other words EHocking wait and see.
My protocol is air tight for a true exhibition of dowsing and I will even if i don't pass get numbers that are higher than their 10% that they expect from all dowsing demonstrations.
Spektator
Is that right?
edge, given that so many people in the forum haven't been able to understand your protocol, it may not be as clear as you think it is.
Also, exactly where does the JREF say that 10% is the norm for ALL dowsing demonstrations? I don't think that would be a tenable position for anyone. I can design a dowsing protocol that is all but guaranteed to give me 50% accuracy, more or less! It still wouldn't mean that dowsing works, though.
EHocking
Penultimate Amazing
... and THIS is why I have offered to help you. You know not of what you write.
10% is not JREF's expectation of success from all dowsing demonstrations.
It is YOUR misunderstanding of the test you are proposing to perform.
Yes, you are attempting to find 1 target in 10, BUT, they have stated that your 60% success rate is not better than that expected by random chance at odds of 1:10,000.
And that is why YOU changed you claim to be 70%.
A 10% success rate is NOT a demonstration of dowsing, just as your 90% success in the demonstration for SezMe was still well within what would be expected from random chance. In this case, not even at odds of 1:100.
EHocking
Penultimate Amazing
In this case, your choice is wrong.
If you halt a trial part way through testing where a target is from (say) 10 canisters, the trial is no longer double-blind.
Double-blind requires that neither the tester nor the testee knows the results until the test is complete.
By halting the test part way, ONE of the recorders will know which cans you are rejecting and which you are claiming to be the target.
And example. Say in a trial you state that the 4th can in the trial is the target and that trial is halted. One person (probably the randomiser) at the test area will know the numbers on the 3 canisters you feel are dummies and the number of the 4th can which you consider to be the target.
Now. Without getting into discussions of the feasibility of that knowledge somehow being imparted to other parties, or yourself during the Challenge, the fact that someone DOES know the results before the completion of the entire test breaches the conditions of a double-blind test.
By definition, your suggested protocol is not double-blinded.
Double-blind protocol is a condition of the Challenge.
In other words edge, your protocol is NOT air tight, and JREF will reject it.
IXP
Graduate Poster
- Joined
- Oct 19, 2004
- Messages
- 1,395
I don't see any problem in stopping the trial early if Edge makes his determination early. The knowledge by the randomizer that Edge has failed or succeeeded on any given trial has no effect on the probabilities and is of no help to Edge. In fact Edge could be told after each set of 10 whether he succeeded or not and this would not affect the outcome.
The randomizer already has the information which COULD effect the test if communicated to Edge, i.e. which box the target is in. His extra knowledge of whether or not Edge succeeded on a given trial does not make the test any less blind.
IXP
Last edited:
EHocking
Penultimate Amazing
Conditions for a blind test are that the person being tested is not given feedback on his success during the test.
Not so in a double-blind test. The canisters will be sealed in front of everyone then put in a covered box. They would then be randomly selected from the box and then the NUMBERING of the canister would be randomly selected (drawing cards, drawing marbles, throwing dice, whatever).
In a DOUBLE blind test NO ONE knows where the target is until revealed at the end of the entire test.
The MDC stipulates that the test will be double-blinded.
Therefore:
1. No feedback to the person being tested.
2. No one knows which of the numbered canisters is the target before and during the Challenge trials.
Last edited:
nathan
Zygoticly Phased
- Joined
- Nov 22, 2004
- Messages
- 3,477
I see that, but I don't see how Edge's desired short circuiting affects this. IIUC Edge's requirement here is that he wishes to:
a) spent 2-4 mins (or whatever it was) dowsing canisters [1..X]
b) spend zero minutes dowsing canisters (X..10]
X is the canister Edge thinks contains the target.
I don't see a requirement for him to be told at that point whether he's right or not. so I don't see how it affects the blindness.
EHocking
Penultimate Amazing
Heh. I knew I'd have this conversation somewhere in the thread.
It's not the fact that he is given feedback, but the fact that those recording the results will have an indication of edge's selections.
As you say, edge dowses 1...X canisters, halting the experiment at X, as he considers this to be the target. Let's say, that X is canister No.4.
In a properly double-blinded test neither he nor the testers know if No.4 (or indeed canisters 1, 2 or 3) contain the target.
But...
The canisters ARE numbered. Therefore, when the test is halted at canister No.4, AT LEAST one of the recorders will know that in that trial, edge believe No.4 is the target and that Nos.1, 2 and 3 are dummies.
Not terribly significant, perhaps, but - if someone knows what number canisters edge selects or rejects, by definition, the test is no longer double-blind. SOMEONE knows the number on the canisters that are being selected/rejected.
Now there IS a slim (very slim, I admit) chance that this knowledge could be imparted to a 2nd tester and even edge.
Let's say in the next trial, canisters come up in the order 4, 3, 2, 1 (first was 1, 2, 3, 4).
Two scenarios.
1. edge halts at 1, 2 or 3. One of the recorders knows that he has not chosen the same canister (No.4) as the previous trial and AT LEAST one of his selections to date HAS to be wrong.
It is *possible* that the during the next trial, that person might indicate, with a smirk or grin or whatever, that he/she knows that edge has rejected or selected that canister from previous trials.
The double-blindness of the test is compromised.
2. edge halts at No.4 again. One of the recorders now knows that edge has selected the SAME canister twice in two trials. The fact that the recorder doesn't know whether it is or is not the target is of no consequence, that person STILL knows the result of edge's choice during the trial.
Again, the double-blindness of the test is compromised.
OK, so the probability that this person's knowledge of edge's pattern of selection can be passed on is slim, I don't deny that, but the fact is that the probability exists.
Just as there is a probability (admittedly 1:10,000) that edge will pass purely by random chance.
It may all sound a little anal - but the fact is, purely by the definition of a double-blind test, this Challenge cannot be performed in this manner.
Jackalgirl
Graduate Poster
- Joined
- Aug 25, 2004
- Messages
- 1,801
I'm a little confused. The recorder is going to know which canisters edge selects or rejects every time; the recorder's job is to record this. Even if edge dowses all 10 containers and picks, say, container number 4, then the recorder knows that edge picked 4 and rejected (1 through 10 minus 4).
But neither the recorder nor edge will know whether that selection was correct until the very end, when the two lists (actual locations vs. dowsed locations) is compared.
Am I missing something here?
But neither the recorder nor edge will know whether that selection was correct until the very end, when the two lists (actual locations vs. dowsed locations) is compared.
Am I missing something here?
EHocking
Penultimate Amazing
Just the protol!
That's because the physical process hasn't been discussed here.
One way to double blind would be:
1 person does the randomisation. They place the target in a sealed container to be taken to the test site.
A 2nd person then takes the coffee tin and places it at the test site.
A 3rd is with edge at all times to make sure he's out of sight of all this.
Person 1 (randomiser) records only the order in which the numbered canisters are taken to the test site they do not record edge's guess, nor do they have access to that record during the running of the test.
Person 2 does not know the number in the tin, but records edge's guess.
Person 3 also records edge's guess - for verification by edge, and checking against person 2's record of the results.
In this scenario - if the test is halted part way through - the randomiser will know both the number of the canister AND edge's selection/rejection.
If all 10 canisters are tested, no one person will have both sets of data during the running of the test.