Global Consciousness Project

[T'ai Chi]

If the zener voltage changes even fractionally from RNG to RNG, what will this mean to the output in terms of randomness? Is the tolerance of voltage specified, and is it the same for each diode? Does Orion use the same voltage diodes? Still pondering those questions? It's OK, take your time.

Your first issue is silly, because if, as you are suggesting, RNG output varied dramatically if voltage changed, we'd expect ALL RNG's to be behaving terribly. But they don't, Claus.

Claus

Why did you attempt to pressure me with your demand of

"Do you admit that the Orion RNG has not passed the DIEHARD test? Just yes or no."

when, in fact, when you asked that you did not know if there was an explanation (there was, as posted) for why they said the Orion passed the DIEHARD test and then the only wrote about using ~2million bytes farther down on the page?

HELLO? It is ok... take your time.

 

[T'ai Chi]

Call it what you like.

Well, that sums up your approach.

You assume global consciousness exists, but only for the purpose of debunking.

I take a cautious and scientific view by just looking at the RNG output, and get slammed for it.

Oh well.

 

[T'ai Chi]

see next post

 

[T'ai Chi]

Let's do this.

10000 times.

Here is my R script
x<-rbinom(10000,200,.5)
hist(x,xlab="# of heads")
c(mean(x),sd(x))

the mean is 99.909100. The standard deviation is 7.080795.

(histogram attached)

Now, are you going to proceed to ask me what happens if a fly lands on the computer while it generating the numbers, or if one resistor has more paint on it than the others?

Just focus on the actual pertinant issue, please.

CLAUS, is the RNG performing statistically different from what is expected?

 

[rwguinn]

Let's do this.

10000 times.

Here is my R script
x<-rbinom(10000,200,.5)
hist(x,xlab="# of heads")
c(mean(x),sd(x))

the mean is 99.909100. The standard deviation is 7.080795.

(histogram attached)

Now, are you going to proceed to ask me what happens if a fly lands on the computer while it generating the numbers, or if one resistor has more paint on it than the others?

Just focus on the actual pertinant issue, please.

CLAUS, is the RNG performing statistically different from what is expected?

you are still ignoring the basic question...
Just how in He!! do you calibrate an instrument while it is immersed in the medium and event it is supposed to be calibrated to (rule 8) measure.
All these side shows mean absolutely nothing if the basic question is unanswered.
period. end of science. back to woo.

 

[T'ai Chi]

you are still ignoring the basic question...

And you are still ignoring my response to your non-issue.

I'm not theorizing about what a hypothetical 'global conscious' can or cannot do. We, you included, don't know if such a beast even exists. I'm asking an empirical question, investigating if the RNGs normally produce results that are not statistically different from what we expect.

end of science. back to woo.

So you won't comment on the actual data either?

 

[T'ai Chi]

Bump for Claus.

 

[Zep]

From WAY BACK then!!!

There are two logical reasons to explain this situation without having to resort to woo-woo ones:

a) The DIEHARD tests are NOT a good test of statistical randomness. I find that difficult to sustain, but others may well know better than I.

b) The "supposed non-random bit stream" was no such thing. The "peaks in the data at critical times" were merely wishful thinking on the part of the people running the EGG experiments based on random fluctuations in the data.

I don't agree with a) at all. It could be b), sure. You also leave out a c) possibility, that there may be something going on that has unlikely RNG output occur on days with events.

Your response clearly shows you are having trouble with English comprehension. Or staying on focus. My statement (a) was my acceptance that the DIEHARD tests were not likely to be the cause of the problems. In saying you DON'T agree, you now contend that the DIEHARD tests ARE bunkum, flatly contradicting some pages of your own posts above.

What can I say, Justin... Don't point that gun at your feet again, OK?

 

[T'ai Chi]

Look - "global conscience"

consciousness... not conscience, Zep.

is what these EGGs are supposed to be measuring - that's what this ridiculous exercise is all about

Please concentrate on what I am saying, Zep. I am interested in looking at the RNG output, not in pondering the assumed qualities of hypothetical global consciousness. I will not assume such things even for debunking purposes.

Your response clearly shows you are having trouble with English comprehension.

I got A's in those classes when I was in school, so nope, that can't be it.

My statement (a) was my acceptance that the DIEHARD tests were not likely to be the cause of the problems. In saying you DON'T agree, you now contend that the DIEHARD tests ARE bunkum, flatly contradicting some pages of your own posts above.

My comments were such that if someone thought a), that the DIEHARD tests were bunk, I disagree with them, because the DIEHARD tests are quite good. I'll remind you that the first part of your a) was

a) The DIEHARD tests are NOT a good test of statistical randomness.

and that is the part I disagree with.

What can I say, Justin... Don't point that gun at your feet again, OK?

Aren't you curious what Claus's answer will be regarding if the RNG is performing statistically different from what is expected? Will he commit to any analysis whatsoever?

What do you think of the graph I made, Zep?

 

[Zep]

consciousness... not conscience, Zep.

Yeah, whatever. When it is proven to exist, I'll start to spell it right.

Please concentrate on what I am saying, Zep. I am interested in looking at the RNG output, not in pondering the assumed qualities of hypothetical global consciousness. I will not assume such things even for debunking purposes.

Fine. Then do you agree that to calibrate ANY instrument so that you can later determine if there's been any change in its readings under experimental conditions, you need to NOT have those conditions present while calibrating it? Yes?

I got A's in those classes when I was in school, so nope, that can't be it.

That depends on your school, I suppose...

My comments were such that if someone thought a), that the DIEHARD tests were bunk, I disagree with them, because the DIEHARD tests are quite good. I'll remind you that the first part of your a) was "a) The DIEHARD tests are NOT a good test of statistical randomness." and that is the part I disagree with.

Ah, isn't selective quoting a WONDERFUL thing, Justin? Pity you do it so openly though, or you would have fooled us all. My FULL line was: "a) The DIEHARD tests are NOT a good test of statistical randomness. I find that difficult to sustain, but others may well know better than I." So now you are agreeing with my statement??? Please make up your mind, OK...

Aren't you curious what Claus's answer will be regarding if the RNG is performing statistically different from what is expected? Will he commit to any analysis whatsoever?

What do you think of the graph I made?

Nope, I'm not curious at all. I'm waiting for YOU to come to your senses and realise you are making a goof of yourself here. The again, you MAY be the sort of idjit pilot who sets his altimeter to zero at 10,000 feet in the air and then claims to be "on the ground".

And it's a nice graph - it's got numbers and lines and everything! So what?

 

[CFLarsen]

you are still ignoring the basic question...
Just how in He!! do you calibrate an instrument while it is immersed in the medium and event it is supposed to be calibrated to (rule 8) measure.
All these side shows mean absolutely nothing if the basic question is unanswered.
period. end of science. back to woo.

Precisely.

 

[T'ai Chi]

Then do you agree that to calibrate ANY instrument so that you can later determine if there's been any change in its readings under experimental conditions, you need to NOT have those conditions present while calibrating it? Yes?

But in RNG-land, calibrate just means it passes those tests. Their RNG's pass those tests.

My FULL line was: "a) The DIEHARD tests are NOT a good test of statistical randomness. I find that difficult to sustain, but others may well know better than I."

Yes, and you were informed that I disagreed with the first part of it.

And it's a nice graph - it's got numbers and lines and everything! So what?

So will you avoid adressing the pertinant issue here? Is the RNG that produced it producing it as we'd expect, or is it all out of wack?

 

[T'ai Chi]

Precisely.

Claus, you said if I presented a graph you'd need to know how many times it was done before commenting.

I presented a graph that I made by doing it 10000 times.

Where's your comment?

The mean is 99.909100. The standard deviation is 7.080795.

Just focus on the actual pertinant issue, please. CLAUS, is the RNG performing statistically different from what is expected?

 

[Zep]

But in RNG-land, calibrate just means it passes those tests. Their RNG's pass those tests.

OK, *sighs* we will run with your silly notion, shall we?

Let's accept that these RNGs were calibrated "successfully" in "RNG-land". But then these RNGs were not JUST "in RNG-land". "RNG-land" is necessarily a subset, if you will, of Globalised Conciousnicity (or whatever...). And therefore the RNGs have been under the influence, such as it is, of GC and have been at all times, including when they are calibrated. They cannot escape it - it is not "turn-off-able" or capable of being shielded from, ever.

So in making your claim about "RNG-land", you are actually agreeing with Claus and I that the RNGs WERE affected by GC, even when they were being calibrated.

Whew - I'm glad we sorted THAT out! Now we are all back at square one and facing the right way.

Further, it is now clear, when they were being used in the experiment, there should have been NO variation in the conditions under which the RNGs would have been operated from when they were being calibrated. Same mode of operation, same sensing of data, same effect from GC operating. Do you agree with this, Justin?

Yes, and you were informed that I disagreed with the first part of it.

Good. We agree that DIEHARD is a valid test of RNG calibration. One step forward for mankind, I'm sure.

And having agreed that the RNGs WERE being affected by GC while being calibrated (see your own assertion above), this leaves the only alternative as to why there were any "results" obtained during the PEAR experiment itself:

Zep: b) The "supposed non-random bit stream" was no such thing. The "peaks in the data at critical times" were merely wishful thinking on the part of the people running the EGG experiments based on perfectly random fluctuations in the data.

So will you avoid adressing the pertinant issue here? Is the RNG that produced it producing it as we'd expect, or is it all out of wack?

Please explain to us how the operating conditions for a real live data run of an RNG for PEAR was in any way different from the DIEHARD calibration run. Then we can talk more.

 

[T'ai Chi]

Let's accept that these RNGs were calibrated "successfully" in "RNG-land". But then these RNGs were not JUST "in RNG-land". "RNG-land" is necessarily a subset, if you will, of Globalised Conciousnicity (or whatever...).

Again, you assume Globalised Conciousnicity (or whatever...) exists, Zep, to make your criticism. You are making a HUGE leap here.

Further, it is now clear, when they were being used in the experiment, there should have been NO variation in the conditions under which the RNGs would have been operated from when they were being calibrated. Same mode of operation, same sensing of data, same effect from GC operating. Do you agree with this, Justin?

You still are assuming GC exists, Zep.

 

[Zep]

Again, you assume Globalised Conciousnicity (or whatever...) exists, Zep, to make your criticism. You are making a HUGE leap here.

You still are assuming GC exists, Zep.

No, I'm not making that assumption, PEAR is. That is the whole point of them setting up all these EGGs in the first place - to measure GC. (Although how they manage to explain the effect of GC on RNGs is a matter of laughable speculation, but that's a subject for another discussion). I'm simply allowing for the purposes of argument that GC does exist with the properties they claim it does and that it will affect RNGs.

But, to humour you, let's suppose that GC does NOT exist. Therefore there is no influence by GC in the RNGs or anything else for that matter. And yet PEAR supposedly pulled "valid" data out of these devices. Oh dear. Two options then appear:

a) Some OTHER phenomenon affected the RNGs, perhaps something quite mundane, perhaps not. This would require further careful investigation.

b) The "supposed non-random bit stream" was no such thing. The "peaks in the data at critical times" were merely wishful thinking on the part of the people running the EGG experiments based on perfectly random fluctuations in the data.

And we can go on from there, and as we do, the chances of option (b) being the best answer continue to increase.

 

[T'ai Chi]

No, I'm not making that assumption, PEAR is.

In your criticisms, you certainly are. For example, you won't even look at RNG output, because you say, 'because the GC exists, the RNG output is worthless'.

That is not a skeptical approach, Zep.

Or, am I wrong and you have looked at the RNG output?

 

[athon]

This would be a comedy if it weren't so tragic.

Let me see if I can have a shot at grafting together an understanding of this:

We have a random number generator (RNG). We'll assume that it will generate random numbers with complete randomness on the assumption it has passed a test designed to accomplish this. We'll trust the output of this machine for the purposes of this evaluation.

Right.

Said RNG runs for a determined amount of time. When analysed, there appears to be strings of data which are not considered 'random'.

Point 1: How do we know they are not 'random'. Is there a pattern? What is a random vs. non-random pattern? If I use an analogy of a coin flicked completely randomly, coming up heads and tails, a hundred heads in a row could still apply to being random. The curve of data which fits the nice 'bell curve' expectation is a mathematic ideal, not always the rule for practical randomness.

Not all patterns have meaning. In fact, most don't.

OK, so maybe you can show they are not 'random' strings (again, let's assume). What does this then show evidence of? Non-random indicates an influence, which is what?

Point 2: If these non-random strings are attributed to an influence that cannot be shielded, the evidence for that influence will be impossible to gather. It becomes purely speculative. This is where the confusion over calibration comes into it; the project fails to provide evidence in support of their claims in that there is no negative control demonstrating that 'non-randomness' does not occur at all when GC is shielded.

Now I've either muddied the waters completely or hopefully underlined the main sticking points.

Athon

 

[T'ai Chi]

Said RNG runs for a determined amount of time. When analysed, there appears to be strings of data which are not considered 'random'.

When the output is looked at on the days there was a formal hypothesis made, the results are generally significant, the GCP site says.

Point 1: How do we know they are not 'random'. Is there a pattern? What is a random vs. non-random pattern? If I use an analogy of a coin flicked completely randomly, coming up heads and tails, a hundred heads in a row could still apply to being random. The curve of data which fits the nice 'bell curve' expectation is a mathematic ideal, not always the rule for practical randomness.

If you get those results (100 heads in a row), you'd start, I hope long before, to think the coin was not a fair coin, that is, that there is something going on. The probabiltiy of that is .5^100, which is darn tiny, and is strong evidence against the coin having probability(heads) = .5.

The probability, as of July 2004 of seeing the RNG output on days associated with the formal hypotheses is .000076, which is fairly tiny. If the RNG 'coin' was fair, we'd expect this probabiltiy to be larger than .05 from what I understand.

 

[CFLarsen]

When the output is looked at on the days there was a formal hypothesis made, the results are generally significant, the GCP site says.

Wrong.

What GCP is doing is picking some events that could pass for "global consciousness", and make predictions for those days. However, they leave out some events: E.g. I see an analysis of Earth Day 2001 (which is claimed to show GC), but not any analysis of Earth Day 2002, 2003, or 2004.

That's plain dishonest.

If you get those results (100 heads in a row), you'd start, I hope long before, to think the coin was not a fair coin, that is, that there is something going on. The probabiltiy of that is .5^100, which is darn tiny, and is strong evidence against the coin having probability(heads) = .5.

The probability, as of July 2004 of seeing the RNG output on days associated with the formal hypotheses is .000076, which is fairly tiny. If the RNG 'coin' was fair, we'd expect this probabiltiy to be larger than .05 from what I understand.

The RNGs are not calibrated. The output is worthless.