On Oct 5, 2009, at 9:40 PM, William Beaty wrote:
Hey, SSE just published a new magazine on alt-science, edited by
Patrick
Huyghe of The Anomalist. (I'm buying the paper version, so they'll
actually earn a few dollars.) http://www.scientificexploration.org/
edgescience/
---
EdgeScience Magazine
New from SSE!
Why EdgeScience? Because, contrary to public perception, scientific
knowledge is still full of unknowns. What remains to be discovered
- what
we don't know - very likely dwarfs what we do know. And what we
think we
know may not be entirely correct or fully understood. Anomalies,
which
researchers tend to sweep under the rug, should be actively pursued as
clues to potential breakthroughs and new directions in science.
ISSUE 1, OCT 2009
http://www.scientificexploration.org/edgescience/
THE OBSERVATORY
* Doing Science Means Exploring: An Editorial by Henry Bauer
NEWS NOTEBOOK
* "Surely There's Nothing Left to Discover"
* Just Off By a Factor of 1,000
* A Mysterious Variation in Radioactive Decay Rates, By Peter
Sturrock
FEATURES
* Is the Global Mind Real?, By Roger D. Nelson
[snip]
Cool! And the first issue is free in PDF form.
From the "Is the Global Mind Real?" article I was surprised to find
the noosphere or "Global Consciousness Project" is still alive and
kicking. I posted the following in their statistics forum:
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
The noosphere FAQ states under the topic "The XOR":
"To eliminate biases of the mean that might arise from such
environmental stresses as temperature change, electromagnetic fields,
or component aging, an exclusive or (XOR) mask is applied to the
digital data stream. This is either an alternating 1/0 pattern (this
is applied by the GCP software for the Orion devices, and in hardware
for the PEAR devices) or a more complex mask comprising an array of
all bytes with equal occurrence of 1/0 (this is in firmware in the
Mindsong devices). Both types of XOR exclude bias of the mean, in
principle, and the latter also excludes all short-lag bit-to-bit and
byte-to-byte autocorrelations. Finally, data for the GCP are recorded
as "trials" that are the sum of 200 bits drawn from the primary bit
sequence. This sum across bits further mitigates any residual short-
lag autocorrelations or other time-series predictability. The result
is a data sequence of random numbers that conform to the appropriate
theoretical binomial distribution and to its normal approximation."
I posted some comments in 2000 and 2003, in an anomalous science
newslist, regarding the methodology employed in US Patent 5,830,064,
the Mindsong Inc patent describing methods used in a REG. See:
http://www.mtaonline.net/~hheffner/RandPad.pdf
I think a superior method of data whitening, compensating for device
bias, was suggested for your REG application.
If the sum of bits means the XORing of 200 bits to make one trial
bit, and only trial bits are used for data, then the need for initial
device bit XOR whitening is drawn into question. It raises the
question of whether whitening of device data is actually desirable at
all for your application. The whitening might actually be destroying
useful data. It may more useful (sensitive) to record and develop
baseline statistics for raw (non-whitened) data and then compare live
data sample statistics with baseline statistics using student-t tests
or similar.
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Best regards,
Horace Heffner
http://www.mtaonline.net/~hheffner/
