linkstream2 microblog

Do Police Interrogation Techniques Produce False Confessions? Yes, from too much of the Reid approach
http://www.newyorker.com/reporting/2013/12/09/131209fa_fact_starr MT @MMLipinski

DEPT. OF CRIMINAL JUSTICE
THE INTERVIEW

BY DOUGLAS STARR

DECEMBER 9, 2013

Key points of the the Reid Technique:
– anxiety as a hallmark of lying
– focus on non-verbal cues – e.g. face touching, averting eyes, arm crossing, &c.
– downplaying moral consequences without mentioning legal ones – story construction via picking from multiple choices
– subtle lies from the interrogator

vs. PEACE Approach
– focus on content and logical contradictions, which give rise to a cognitive load, instead of non-verbal cues
– open ended Qs instead of multiple choice

Reid’s approach was called in question for yielding false confessions by psychological experiments such as the “alt key test” .

Romantic Insights From Maps of Facebook #Networks: Many mutual friends don’t necessarily indicate a good match
http://bits.blogs.nytimes.com/2013/10/28/spotting-romantic-relationships-on-facebook

The Decline Of #Wikipedia. Current issues: fewer editors, biased coverage & hidden wysiwyg interface
http://www.technologyreview.com/featuredstory/520446/the-decline-of-wikipedia via @mhsantosa

http://chronicle.com/article/Still-Hunting-Medical/143221/

Paul Basken
The Chronicle of Higher Education

Genome-Wide Comparison of Medieval & Modern M. #leprae, reveals ~1600 #pseudogenes, w/ slightly more in modern strain
http://www.sciencemag.org/content/341/6142/179.abstract

QT:{{”
Most bacteria will have some pseudogenes in their genome, maybe, you know, in Mycobacterium tuberculosis, for instance a close relative, there are 4,000 genes
present and perhaps 20 pseudogenes. In Mycobacterium leprae, there are 1,600 real
genes and as many pseudogenes. For me, this has always been puzzling, because bacteria
generally tend to, once a function has been lost, the corresponding genes are usually
eliminated and we see the genome shrinking. This hasn’t happened in Mycobacterium
leprae, because there’s still such a huge number of pseudogenes present. And that makes
me think that maybe Mycobacterium leprae emerged in this form only very recently and
that there hasn’t therefore been sufficient time for these pseudogenes to be lost.
However, this is clearly speculation, and it needs to be tested by further experiment. For
instance, looking at older samples might be helpful because the analysis described in the
recent Science paper shows that there are more, a few more pseudogenes present in
modern strains of Mycobacterium leprae than there were in medieval European strains.
So if we could go back a few thousand years more, we might find that actually there were
a couple more functional genes at that particular point.
….
Now is there any evidence that the successive number of pseudogenes contributes to
either its slow growth or its resistance to growing in the lab or its just kind of long
standing plague on humanity?
….
Yes, I think while there’s no experimental evidence to prove that the pseudogenes are
responsible for the slow growth, I think it’s highly likely that they do contribute because
lots of very essential functions have been lost, and this means that M. leprae, for instance,
has difficulty in acquiring iron because it’s lost the genes required for iron uptake.
“}}

https://discussions.apple.com/thread/2599717

http://www.salon.com/2013/12/08/war_from_afar_how_the_pentagon_fell_in_love_with_drones/

.@randal_olson @thatdnaguy Lots of interesting #network datasets available from http://snap.stanford.edu/data . Thanks for pointing out this site!
http://snap.stanford.edu/data

This is the URL for ccle with WGS data:

https://cghub.ucsc.edu/datasets/ccle.html

They have an XML describe all the datasets and there are WGS as well. (This should be publicly available WGS.)

QT:{{”

The Cancer Cell Line Encyclopedia (CCLE) project is a collaboration between the Broad Institute, the Novartis Institutes for Biomedical Research, and the Genomics Institute of the Novartis Research Foundation. CCLE will conduct a detailed genetic and pharmacologic characterization of a large panel of human cancer models, develop an integrated computational analyses that link distinct pharmacologic vulnerabilities to genomic patterns and translate cell line
integrative genomics into cancer patient stratification. The CCLE provides public access to genomic data, analysis and visualization for about 1000 cell lines.

NOTE: these data sets do NOT require any special access or authorization status. However, you will need to use the public access token/key URL, or your secure key file that you may have downloaded for projects like TCGA that use a secure key file.

“}}

http://www.michaelnielsen.org/ddi/how-the-bitcoin-protocol-actually-works/