| Wednesday,
August 2, 2006 Please check your Tarnkappe
at the door!
"Any sufficiently advanced technology is
indistinguishable from magic." Sir Arthur C. Clarke
http://news.yahoo.com/s/nm/20060731/sc_nm/science_invisible_dc
It's unlikely to occur by swallowing a pill or donning
a special cloak, but invisibility could be possible in the not too distant
future, according to research published on Monday.
Petronius
==========
Subject: Eight words that explain a corrupt concept,
Jerry
Political correctness is affirmative action for lousy
ideas.
Sue
==========
Subject: Global Warming Link to Hurricane Intensity
Questioned
Thought you might find this interesting. It looks like
some are looking at the data, and not just trying to generate grants by
proclaiming the groupthink du jure.
Jim Riticher
http://news.nationalgeographic.com/news/
2006/07/060728-hurricane-warming.html
Excerpt:
Global Warming Link to Hurricane Intensity
Questioned
John Roach for National Geographic News
July 28, 2006
An expert with the U.S. National Oceanic
and Atmospheric Administration (NOAA) is questioning the connection
between climate change and the appearance of more intense hurricanes in
recent years.
Historical data on hurricanes is too crude
to determine long-term trends in intensity, says Christopher Landsea, a
science and operations officer with NOAA's National Hurricane Center in
Miami, Florida.
Extreme hurricanes like Katrina were likely
as common around the world 30 years ago as they are today, Landsea says.
But since satellite imagery was poorer, storm intensities were
underreported.
============
How soon the
Singularity?
Subject: This Morning's Stories
Vernor Vinge suggests that we will know the
Singularity is approaching when we start seeing *large* software projects
being developed successfully, particularly those involving a lot of
parallelism. Right now--based on the UK Government's experience with IT--I
think we're safe, but I think I should comment--I was a software architect
for many years, and my current research is exploring biological ways of
generating intentional behaviour using parallelism. (I'm using a large
Beowulf cluster and biomimetic robots for part of this work.)
To me, the problem with large projects appears to be
that software development currently involves finding a large number of local
solutions. The architect ensures that these local solutions work well
together--it often takes a good deal more intelligence than your average
manager possesses to do that. I suspect we won't solve the problem until we
can provide the architect with suitable tools to visualise and control the
design. Most of the tools I've seen are unable to manage just those aspects
of the system design that I have to manage in a successful solution. The key
interactions that cause problems are deep down in the details, and tools
don't help you identify or control them. If we want to build larger systems,
I think we will have to move away from the algorithmic and specification-
driven approach to solving problems and come up with a more robust approach
to combining subsystems. That would make the job easier for the architect,
too, because it is the sheer number of detailed interactions that limits the
size of a system that one man or woman can design. It will also allow
evolutionary design to work, since a refactoring capability is just what I
want at the architectural design level. We experimented with architectural
teams at TRW, based on some ideas of John Gormally, and they seemed to be
successful for large systems, but it still helped to have one or a few
persons on the team providing style guidance and controlling specific
system- wide aspects. So the first need is an architectural assistant in the
architect's laptop. The Eclipse SDK provides a taste of some of the things
needed. Give me an architectural tool like Eclipse, and I could build some
amazing things.
We're only just beginning to understand how the brain
solves these kind of problems. Here is a report of a recent paper on how the
brain turns on innate behaviour <http://cognews.com/1154442364/index_html>
that suggests one solution might involve a timed sequence of actions turned
on by hormones, rather than by synaptic interactions. Since innate behaviour
can be as complex as a new-born colt or calf standing up and walking, it
suggests that most behaviour involves such timed sequences. Learned (as
opposed to innate) behaviour is probably scheduled relative to internal
timing signals--perhaps gamma and beta rhythms. The fact that behaviours can
be interleaved suggests various behaviours are associated with specific
phases of the internal timing sequences--Christo Panchev has done work
suggesting how this might take place <http://www.his.sunderland.ac.uk/
ps/panchevwermter05.pdf> . So a second need is some way of scheduling
timed activities so they work together compatibly. The solutions I see the
engineers coming up with here are too narrow, but I haven't had the time to
point them in what I suspect is the right direction. (Sue Denham has
commented that neuroscience is too hard for most engineers!) My current
grant may give some insight here in about two or three years.
There are some other hard problems to be solved before
we will make much progress. One major one is understanding how the brain can
do algorithmic things non-algorithmically. For example, bats are able to
predict the future positions of *accelerated* targets, yet there's no
evidence in the brain of any mechanisms for vector arithmetic. Also,
planning seems to involve decision tree representations, but we don't see
evidence for the switching networks we would expect. That's why I speculate
that real brains may have some capabilities for working with a continuous
external reality that digital computers will always find computationally
*very* demanding. Real AIs may have to be hybrid entities, with both silicon
and biological components. I know I already interact with my laptop that
way, but the user interface needs a lot of work. So the third need is a more
intuitive GUI. I already use my laptop as medium and long-term memory, but
if it could act as an extension of short-term memory, I would allow me to
work on bigger systems. I'd also like it to compute off-line doing tasks,
searches, and algorithmic calculations while I work on other aspects of a
problem. Perhaps that would be the next key application--it would certainly
use all the compute power available and more!
On to the news:
Blair says we need to completely rethink the war on
terror: <http://www.timesonline.co.uk/article/0,,3-2295604,00.html>
MI5 assessment: <http://www.timesonline.co.uk/article/0,,2-2295582,00.html>
Curfew orders on terror suspects held illegal: <http://www.guardian.co.uk/terrorism/story/0,,1835296,00.html>
<http://www.timesonline.co.uk/article/0,,2-2295588,00.html>
NHS trusts divert sexual health funds to cover
their deficit: <http://news.bbc.co.uk/1/hi/health/5234938.stm>
<http://www.guardian.co.uk/medicine/story/0,,1835308,00.html>
<http://www.timesonline.co.uk/article/0,,2-2295495,00.html>
Another IT disaster: <http://www.guardian.co.uk/conservation/story/0,,1835280,00.html>
A hospital fined for being too efficient: <http://www.timesonline.co.uk/article/0,,2-2295840,00.html>
To save money, NHS trusts are only providing one
cochlear implant to deaf children: <http://www.timesonline.co.uk/article/0,,2-2295387,00.html>
(This is one area where I work as a researcher, so I have a strong negative
opinion of these policies.)
Hermitage hit by thieves: <http://news.bbc.co.uk/1/hi/world/europe/5235148.stm>
<http://www.guardian.co.uk/russia/article/0,,1835338,00.html>
School workload: <http://news.bbc.co.uk/1/hi/education/5235146.stm>
-- Harry Erwin, PhD, Program Leader, MSc Information
Systems Security, University of Sunderland. <http://scat-he-g4.sunderland.ac.uk/~harryerw>
Weblog at: <http://scat-he-g4.sunderland.ac.uk/~harryerw/blog/index.php>
=============
The other shoe drops.
http://www.washingtonpost.com/wp-dyn/content/
article/2006/08/01/AR2006080101334_pf.html
-- Roland Dobbins
Not unexpected.
========
Dr. P.,
I just finished listening to your guest appearance on
"This Week in Tech" with Leo Laporte. I very much enjoyed your input. I
listen to the show weekly and I have occasionally felt that the viewpoint is
too heavily skewed toward the 'insider' or 'geek' viewpoint. You provided an
exceptional grounding for the conversation and it made the show much more
valuble.
Thanks for all you do,
Brian K. Moynihan
Thanks!
==========
Subject: 1 Aug 1971: Apollo 15 finds Genesis rock
Jerry:
With everything else that's going on, this is worth
remembering:
http://news.bbc.co.uk/onthisday/hi/dates/stories/august/1/newsid_4101000/4101579.stm
<http://news.bbc.co.uk/onthisday/hi/dates/stories/august/1/newsid_4101000/4101579.stm>
Steve
*****************
It is lack of confidence, more than anything
else, that kills a civilisation. We can destroy ourselves by cynicism and
disillusion, just as effectively as by bombs.
---Lord Kenneth Clarke
============
I asked Colonel David
Couvillon, USMCR, to comment on the Iraq article:
Subject: Re: Even
Churchill Couldn't Figure Out Iraq.
Of course the Sunni's are afraid the Shiia won't give
them a fair deal. After all, they didn't give the Shiia a fair deal the
whole time they had control.
One question they avoid, is whether a Shiia/Sunni
conflagration wouldn't be better for the Western world than peace? Saudi,
Kuwait, Jordan, Sunni Iraq, possibly Turkey against the Shiia Iraq and Iran,
The Taliban and Hezbollah (I'd bet a buck that Syria tries to sit out the
overt warfare). Pressure is certainly relieved against Israel. Iran is
preoccupied with war - diverting attention away from developing nuclear
weapons; well, certainly slowing it (AND, it provides cover for western
espionage [as poor as they are in that part of the world] to sabotage the
Persian nuclear program). Palestinians lose their funding. The Western world
(nee - the US) would have to blockade the Persian Gulf and Mediterranean
east coast.
Downside is the disruption in the flow of oil from the
region - turn that around, though and give the Russians opportunity to sell
the world their oil. They'll reluctantly support the Western effort.
Col Hammes is absolutely correct in his analysis about
the Iraqis believing that we are leaving. Everyone is moving to grap the
power left by that vacuum. We saw that immediately after Baghdad fell, while
in the early months after the war - everyone skeptical but jockeying for
position. Certainly the Shiia where deathly afraid that we'd leave and the
Sunni would again renew their pogroms. We really blew that - the Iraqi Shiia
were ready to be on our side AND to ignore the Persians (despite the
presence of the Badr Brigade/SCIRI) - The Iraqi Shiia hated the Persians
from the Iran/Iraq war. We could have denied the Mullahs their reach in
Iraq, if we wouldn't have been so afraid of their money and organization. We
were so afraid, in fact, that we stood by and let them organize and use
their money to put themselves in the position they're in now.
Sadr used to be a joke to ALL Iraqis. Now, because we
didn't have the guts confront him, he has strength - and the Sunnis/Al-Queda
saw that we were a paper tiger - not willing to use the power we had. Now
it's squandered.
A colleague of mine, serving in Iraq as an advisor to
an Iraqi general, was present at a meeting with the senior American
commander. After the meeting was over, the Iraqis were not satisfied. My
colleague's counterpart, commented to him about the American commander,
(interpreted & cleaned up a bit) "Nice guy, but he has no balls." Too many
of the Iraqis, Arabs, Islamists, etc., believe we have no balls.
Couv
Machiavelli: Never do your enemies a
small injury.
Our problem was that the neo-conservatives
in the White House really believed that Chalabi would be welcomed with
flowers and cheering, and he would take over Iraq and be our ally and build
a pipeline for the oil, and everything would be just peachy. When that
didn't happen they didn't have a real plan, and they distrusted all those
who did have plans because we had all told them, "Don't invade. If you do
invade, this is what you must do. But don't invade." They never heard a word
past the advice about not invading.
Competent imperial wars are more expensive
than most believe. Incompetent imperial wars are even more expensive, but
one needs to know more history than most to know this.
============
Continuing the discussion on domestication
of humans:
Subject: domestication
I'm afraid Mr. Thompson and Mr. Cochran are greatly
misinterpreting the trends in human social interactions. Not all
hunter-gatherer groups are or were completely egalitarian. True egalitarian
communalism is a rarity, both ethnographically and archaeologically.
Evidence suggests that in even largely unstratified egalitarian societies
there would often be at least some informal status differentiations (whether
merit-based or ascribed by lineage) through which a group's activities are
organized.
Further the general trend in group structures is to
maintain the status quo, even when that is not necessarily to the best
advantage of a large portion of the group as a whole. This is why
archaeologists are very much interested in the transition between small band
social systems and nascent state societies (antiquated terminology these
days, but I figured most people would be familiar with these terms).
In the former, subsistence and resource allocation are
relatively homogenous across the population thereby being much to the
advantage of even the lowest status members of the group. In the latter,
resources are typically very unevenly distributed across the population and
there is a great deal of differential access to subsistence and resources
between status group members. Therefore the latter system is much less to
the advantage of the majority of the population. Even still the general
historical trend is towards greater stratification and complexity of social
systems despite its disadvantage for the majority of the social members.
The logical conclusion from this is that the trend as
population increases and societies become more complex in structure
resources and subsistence becoming unevenly distributed is supported by the
majority disadvantaged population, and resource availability _as a whole_
becomes greater although the lower strata have less access than they had as
small groups (e.g. typical nutrition and health in agricultural societies is
far worse than that of hunter-gatherers - agriculture is just a bit more
predictable).
Does this mean they've become domesticated? Not
really. In making individual choices within a cultural system, the
probabilistic trend is toward the norm maintaining the status quo and the
status differentiations. Only extreme pressures of subsistence and survival
are likely to break those choices away from the normative trend, and is
likely to be motivated by an outlier individual or individuals (incipient
elites) making choices that can rally and organize the lower strata
population. Thus the rarity of (successful or not) peasant or slave revolts.
Most civilizations have collapsed from the top down instead.
None of this suggests domestication, only that the
probabilities involved in human agency within a social structure operate to
maintain a system rather than being a destabilizing force. Yet another
reason why the Jacobin philosophy of waiting for the "people" to rise up
against oppressive regimes is unfounded. People more often than not will
choose AGAINST their own self interest to maintain the devil they know.
A somewhat simplistic explanation, but covers much of
the observable evidence without going into too many of the complexities of
large scale social interactions and cultural propagation.
Just my 2c as a reluctant voodoo scientist...
--J. Scott Cardinal
==============
Steve Goldberg on natural law.
This came from another conference. Dr.
Goldberg had made an argument that elicited the comment from me, "No natural
law for you, then." Another participant then answered,
>>(moral philosophy)...refers to the logic,
science (=experience), and art of designing public policies that actually
have some realistic chance of increasing the general welfare<<
to which he replied:
Meme 071: A virtues approach to personality
Hi,
This is no doubt desirable, and entirely concordant
with science, but it has nothing to do with moral philosophy. It is
scientific in the sense that it addresses empirical causes and effects. But
"moral philosophy", virtually by definition, attempts not merely to find
effective ways of implementing moral assumption--determining the moral
assumptions not being the province of science. Moral philosophy attempts to
determine which moral assumptions are "right" (I.a., desirable). Science has
nothing to say about this.
To the extent that one wishes to answer the empirical
questions and take the moral assumptions as given (without claiming that
they are justified in any scientific sense) then economics and political
science would be the venues.
Which is not to say that these, or any other social
sciences, are, in fact, capable of well answering the empirical questions.
After forty years in sociology I've become persuaded that, save those
aspects of social life for which hereditary factors play a role, such
questions are for the most part unanswerable. The problem is not primarily
the extent to which ideology has captured social science, but the fact that
most social questions are effectively infinite-dimensional manifolds.
I *would* ease up and go along with a moral philosophy
that saw societal survival as the initial moral assumption. The problem is
that cross-cultural evidence gives us pretty much the set of moral precepts
that would be obvious to a seven-year-old. All else varies from society to
society and can't be seen as required for societal survival.
It may well be true--as I have written--that the view
that separates science from morality is disastrous for many reasons. But
this is no argument for it's not being correct.
Best,
Steve Goldberg
I haven't time to write a long essay on
the subject. I do refer interested readers to C. S. Lewis, The Abolition
of Man, all three parts.
================
Need surgery? How about going overseas?
Dear Jerry,
An interesting response to high health care
costs-outsource it:
"After going overseas to outsource everything from
manufacturing to customer services, American businesses - pressed by rising
health-care costs - are looking offshore for medical benefits as well. A
growing number of employers who fund their own health insurance plans have
begun looking into sending their ailing employees around the world for
surgeries that in the U.S. would cost tens of thousands of dollars more..."
http://www.concordmonitor.com/apps/pbcs.dll/article?
AID=/20060731/REPOSITORY/607310329/1013/48HOURS
Perhaps the hidden costs of Malpractice lawyers and
illegal immigration are coming home to roost.
Cheers, Rod Schaffter
-- "A good scientist, in the absence of data, simply
says, "we don't know" rather than proposing an unfalsifiable, untestable
explanation." --Robert Bruce Thompson
TOP
CURRENT
VIEW Wednesday
|
An Important warning about domain hijacking
