| Wednesday,
September 24, 2008 Jerry
Here's how the Mexican handle disasters:
http://www.fredoneverything.net/KatrinaSouth.shtml
Reminds me of us, before FEMA.
Ed
We used to have Civil Defense. And when I was
growing up in Tennessee that's how we dealt with tornadoes and such. That's
what de Tocqueville described as uniquely American. Cry for what we have
lost to the liberals and socialists who do not believe in government by the
people. They believe in government of the people by elites.
==========
GPS accuracy for Train Control
Dear Doctor Pournelle,
An unidentified correspondent wrote in Mail for
09/17/08:
"However, GPS is not precise enough to tell the
difference between two tracks, side by side, or to tell whether a train is
clear of a switch by only a hundred feet."
I am not an expert, but I am gifted with some common
sense, and a ten minute bout of research gave me the following facts/data
indicating that GPS actually could either do that job either today, or if
not today, then very soon with upgrades to the GPS satellite system that are
being implemented right now:
Civilian C/A Coded signals used by civilian GPS
receivers have an accuracy of:
http://www.annadelpress.com/GPSAccuracy.pdf
:The DoD estimates the accuracy of civilian GPS to be
better than 50 feet at least 95% of the time, anywhere in the world. This is
what’s called the 2drms (twice the distance root-mean-square) accuracy. It
is based on statistics. Don’t worry about the math, just remember that the
2drms number means that over an entire day, a receiver’s accuracy should be
better than 50 feet at least 95% of the time."
BUT, the CA code is lightly encrypted and easily
"spoofed" (think hackers, psychopaths and terrorists jamming a trains' "CA"
code GPS system). Any secure system would use the current heavily encrypted
military "P" code. When the upcoming (2011-2013 expected) Block III
satellites are available, there will be a more accurate military "M" code
available.
The current military "P" code is much more accurate
than the current civilian CA code. How much more? It's a secret. the exact
numbers. Here's an anecdotal report of how accurate they can be in practice:
http://forums.groundspeak.com/GC/index.php?showtopic=50480
"The mil GPS has 2 modes. In the uncoded mode they are
a bit more accurate than the civilian handhelds. However, in the coded mode
they are really accurate. The 2nd night out we were to locate a .30 cal ammo
can hidden in the desert that had a message in it. The purpose was to
simulate hooking up with a covert agent who had left us instructions. Again,
moonless night, we were moving tactically which means absolutely no lights,
and it was a 7+ kilometer hike thru some pretty tough ground. Using the mil
GPS in the coded mode I stepped on the can."
A .30 cal. ammo can, IIRC, is about the size of your
average lunch box. "I stepped on the can", that's fairly decent accuracy.
Sounds as if, judging by that account, a military "P" coded GPS signal is
likely accurate enough to distinguish between two trains side by side, and
almost certainly able to distinguish between two trains on the same block of
track.
Further information available at:
http://www.oreillynet.com/pub/a/wireless/2000/12/08/gps_intro.html
"Using both P-code frequencies, PPS position accuracy
has been published by the US military as being able to provide 9 meters CEP,
16 meters SEP, or 22 meters R95. UTC time is accurate to 200 nanoseconds.
This is quite impressive accuracy, particularly considering that the
position can be calculated after only a few seconds of observation from a
complete "cold start." Methods for improving accuracy down to the meter
level have also been proposed.
For non-privileged SPS users, the accuracy is about
half as good. Before May 1st, 2000, SPS accuracy was significantly worse
(approximately 76 meters SEP) because of something called *selective
availability* (SA). SA is an operational mode of GPS, which introduces an
intentional degrading of the CA-code signal. While now disabled, users
should be aware it can be reactivated in (wide) areas during times of
hostilities."
So the military will admit to being twice as accurate
as the civilian GPS.
But it gets better (I love this part!): From later in
the same article linked above:
"In a classic case of one branch of an organization
directly working against another, the US Coast Guard installed the Maritime
Differential GPS (DGPS) service. This works by having GPS receivers placed
at fixed, known locations, and having them transmit information about each
satellite's error.
DGPS-equipped receivers, when getting good signals
from both the GPS satellites and one or more nearby DGPS stations, can
resolve their position with an accuracy of 10 meters (R95) or better,
depending on how close the receiver is to the DGPS station."
So if your "Train Locating System" sets up a ground
network of DGPS stations, you get even more accuracy! As a bonus, those same
DGPS Stations would make even the CA code civilian systems more accurate.
You get door to door accuracy for cell phone GPS and automobiles as a bonus
for your investment. Think of blind people having "P" oir eventuall "M" code
GPS "glasses" that guide them step by step. Children with similar GPS
bracelets or GPS in their clothing that allows them to be tracked in real
time, locating lost or kidnapped children in minutes (sort of a "LOJAC" for
kids).
By the way, for those worried about CA code systems
being used by Bad Guys: DOD can flip a switch at any time that degrades the
CA code signal, making it only half as accurate. I would imagine they could
also disable selected DGPS ground stations if there was a localized problem.
It's a system feature I would certainly insist upon.
If these sources are accurate, then GPS can handle the
train job rather handily.
I cannot imagine any train, or other transportation
mode, anti-collision system not using GPS as a Very Important Component of
the location sub-system, with non-GPS components (track sensors, inertial
guidance, terrestrial radio signal positioning and even celestial
navigation) playing roles.as verification/redundancy components. At the
least I would put GPS right next to track sensors, or right behind track
sensors in order of importance. Since a track sensor is actually easier to
spoof or sabotage than a "P" signal GPS, I think GPS is likely more reliable
and "fail safe". But use both. Satellites can fail too.
Petronius
PS IIRC, didn't your GPS system in the Death Valley
Rollover Incident (as related in Chaos Manor Reports by you) a few years
ago, actually show that your altitude had changed after the Bronco rolled
one-eighty?
===========
A couple of favorite iPhone apps
Jerry,
For a calculator, you really can't beat the oddly
named i41cXp, which is a PERFECT emulation of the old HP 41c calculator
(circa 1980, and what I used in graduate school). You can even write RPN
programs with this thing. Obviously a labor of love and I recommend it.
You can also have Pandora radio (http://www.pandora.com/)
on your iphone. If Pandora survives the increased song licensing fees, this
will continue to be a great thing. Your music, selected by you to your
tastes, streaming to you wherever you want it. It doesn't get much better,
and it's free.
My overall impression of the App store is that it is
90% dreck, typical calculators and multiple versions of old (and the same)
games, but the gems are interesting and some are opening up interesting new
territory.
The iPhone has the computing power of a mid-late
90's(?) PC. It will be interesting to see how this platform evolves.
Chuck
==========
Embezzling broker invests proceeds in 419 scam.
<http://www.nytimes.com/2008/09/19/nyregion/19broker.html>
-- Roland Dobbins
==========
About Bad Fannie & Freddie...
New York Times article from September 11, 2003:
The Bush administration today recommended the most
significant regulatory overhaul in the housing finance industry since the
savings and loan crisis a decade ago.
Under the plan, disclosed at a Congressional hearing
today, a new agency would be created within the Treasury Department to
assume supervision of Fannie Mae and Freddie Mac, the government-sponsored
companies that are the two largest players in the mortgage lending industry.
The new agency would have the authority, which now
rests with Congress, to set one of the two capital-reserve requirements for
the companies. It would exercise authority over any new lines of business.
And it would determine whether the two are adequately managing the risks of
their ballooning portfolios.
The plan is an acknowledgment by the administration
that oversight of Fannie Mae and Freddie Mac -- which together have issued
more than $1.5 trillion in outstanding debt -- is broken. A report by
outside investigators in July concluded that Freddie Mac manipulated its
accounting to mislead investors, and critics have said Fannie Mae does not
adequately hedge against rising interest rates.
Source, Instapundit. Apparently, Congress did nothing.
Julie
Of course they did nothing.
===========
Hand Held Computers like in the Mote in God's Eye.
In your column today you talked about how your IPhone
was getting close to becoming like the hand held computers you wrote about
in the Mote In God's Eye.
Big deal, I'm waiting for the computer implants you
promised were coming in the Oath of Fealty!
Love your work! I was sure glad to hear that you beat
your tumor or what ever it was.
Wade
===========
Jerry,
Passing thoughts on the mail you posted overnight
Monday...
1) From personal observation, the comment about US
shipbuilders building for the military affecting their commercial
competitiveness also applies to US aerospace, particularly after the
consolidations that eliminated McDonnell-Douglas as a competitor and seems
to have closed Lockheed's commercial aircraft capability. Locally, the
surviving US commercial aircraft manufacturer is a laughing stock for their
government contract work due to low-quality product delivered late with
every deliverable parsed for legal consequence (e.g. only the minimum
necessary contractually required deliverable, etc.) and with complex,
expensive "quality controls" that are impossible to apply due to their
complexity.
2) Autocad owns the entrenched market because it was
there first -- in my (limited) experience, BRL CAD did not have a
competitive capability until recently. Certainly if BRLCAD is competitive
now it should be used as a viable alternative (and I'll be exploring it for
CAD work on some of my current projects as required).
3) Black swans. The current financial crisis is only a
surprise to those who haven't been paying attention (though it is a shock in
the sense of "happening now"), and while its probably impossible for McCain
to successfully lay all the blame at the feet of the Democrats, the fact
remains that the Democrats contributed more to the regulatory and political
environment which created and nurtured the crisis, and lead the charge to
delay investigations particularly of the federally-chartered organizations
that contributed to their campaigns at such high levels. I've never been
comfortable with the concepts of short selling, hedge funds, and investment
reinsurance (capitalize at $10 million. $2 millions goes to an insurance
company that repays the $10 million to the investors if the company fails),
and despite the pro-short selling essay in yesterday's Wall Street Journal
-- which I must admit I didn't read -- shorting done honestly may be a
viable method for regulating "overheated" markets but is far too prone for
abuse. Similarly, reinsurance eliminates the incentive to honestly assess
risk in investments -- hence the current crisis, and the reinsurers leading
the charge to the bottom. The two reforms I would make: (a) federally
chartered corporations are prohibited from lobbying and donating to
Congressional candidates as an obvious conflict of interest; (2) a large
fraction of executive bonuses should be tied to three-year performance (e.g.
options at market which mature in three years, or cash bonuses placed in
escrow for that period with the deferred bonus pool -- vice Lehman's bonus
pool -- available back to the corporation to repair financial problems
during that period. Unfortunately, it looks like the solution will be
taxpayer infusions of money and Congressionally-appointed scoundrels
replacing the shareholder-appointed scoundrels who created the problems.
(And what happened to the responsibilities of the corporate boards?)
It's a mess, but it's a mess that the government
created and perpetuated by messing with the free market. Unfortunately, the
solution will be more socialism (government ownership of corporate
institutions) and fascism (government control of privately-owned corporate
institutions).
And -- dare I say -- Palin is the only candidate with
both the experience and the isolation to resolve the problem. Which has lead
to some thoughts that I would rather not have thought of and prefer not put
into writing explicitly (think of conspiracy theories and the "down side"
elements of Mr. Heinlein's "Through the Looking Glass" from Expanded
Universe).
Jim
Every dollar removed from the economy in taxes reduces
GDP by at least a dollar.
on shorting: he who sells what isn't his'n
must buy it back or go to prison...
=============
Climate Science: Is it currently designed to answer
questions?
Paper by Richard S. Lindzen, 19 Sep 2008.
http://arxiv.org/ftp/arxiv/papers/0809/0809.3762.pdf
The abstract ends with: "....When an issue becomes a
vital part of a political agenda, as is the case with climate, then the
politically desired position becomes a goal rather than a consequence of
scientific research. This paper will deal with the origin of the cultural
changes and with specific examples of the operation and interaction of these
factors. In particular, we will show how political bodies act to control
scientific institutions, how scientists adjust both data and even theory to
accommodate politically correct positions, and how opposition to these
positions is disposed of."
============
Re: CNU Update: Bricks 'N Mortar AutoCAD 2008; A
Second Derivative Ripoff
It would be easy to sustain the Great CAD War debate
with Mike Powers. I have QuickCAD 7 & 8, AutoCAD 2000i, DeltaCAD and
TurboCAD Deluxe r14. And now AutoCAD 2008 Inventor/Mechanical.
And he left out a primary file interchange format, the
Drawing Exchange File, *.dxf.
I've used CAD programs since 2000 to make drawings for
CNC cut parts. Elena was having trouble with 3d parties in her work. I
bought QuickCAD 7 at Office Depot and discovered it was very similar to
navigation with a map. Points and vectors in a Cartesian coordinate system
after all.
All irrelevant.
He's right about the multiplicity of programs. Which
further highlights the question of why AutoCAD is the default program at
Lemon Bay HS, the Charlotte Technical Center, Manatee Community College, CNU,
Penn-Foster...
You caught my real point, which is radical capital
cost reduction for people retraining and new productive enterprises. And not
just with CAD programs, which after all are only one tool in the tool box.
Mark
=============
Energy capital
Jerry,
Note on the radio news this AM, that a substantial
majority of Obama supporters and about half of McCain supporters believe
energy companies need to be investing in renewable energy even if it drives
energy prices up.
Uh..hasn't most of our capital to support such
projects recently evaporated?
Jim
Indeed
===========
NASA - Solar Wind Loses Power, Hits 50-year Low
Jerry
According to NASA, the solar wind has been losing
power, and has hit a 50-year low:
http://science.nasa.gov/headlines/y2008/23sep_solarwind.htm?list1067857
"[T]he speed of the million mph solar wind hasn't
decreased much—only 3%. The change in pressure comes mainly from reductions
in temperature and density. The solar wind is 13% cooler and 20% less
dense."
Hmm. I guess now is not the time to move to Canada.
Ed
=============
'The GPO officials estimated that the GPO amassed as
much as $180 million in profits by charging an additional $1.83 per
passport on its sales to the State Department.'
http://www.washingtontimes.com/news/
2008/sep/24/passport-printer-repays-51-million/
-- Roland Dobbins
The Iron Law in action.
============
China building reactionless drive
Dr. Pournelle,
I remember seeing an article about a reactionless
drive based on relativistic theory a few years ago. Apparently China's
decided to see if it works:
http://blog.wired.com/defense/2008/09/chinese-buildin.html
It's not theorized to work well, but then it doesn't
have to work well. It should be interesting to see what happens.
Thanks,
Ryan Brown
It doesn't have to work well. It just has to
work at all to revolutionize physics. The way to bet is that it won't
work, but the payoff is enormous if it does.
===========
Spraypainted Swans
I've been seeing the term "black swan" bandied about
quite a bit on the internet these days, in reference to the whole subprime
mess.
This is not one of those cases. Black Swans much, much
rarer than people think. A Black Swan is a healthy twenty-three-year-old
having a massive aneurysm. A Black Swan is a gas bubble in a cast-aluminum
structural frame that makes your car break in half at fifty-five miles an
hour.
What happened in this market was that everyone decided
that they just would stop worrying about risk. Securities were securities,
and a 3/1 interest-only nothing-down ARM was exactly as risky as a T-Bill.
Now everyone is realizing that most of those 3/1-IO's aren't ever going to
be paid off; but instead of saying 'well, we screwed up', they're blaming it
all on a Black Swan. Nobody could have predicted this or prepared for it,
right? This certainly isn't OUR fault, right?
This is not a Black Swan. What happened here was that
a bunch of sheep wandered into a field, and instead of looking out for
wolves they tucked right in, because that grass was just SOOOOO TASTY. And
now the wolves are here, and the sheep are trying to claim that there was no
need to watch out for them, because who would have thought that wolves would
come?
-- Mike T. Powers
Come now. No one is "blaming" anything on a
Black Swan which is a simple phrase use in Taleb's book by that name to
indicate that normal Gaussian statistics do not give sufficient attention
to the really far out outliers, and do not apply in very complex
situations where we do not know the distribution. That's not blame, it's
an explanation intended to teach something to those who will take the
trouble to understand.
Black Swans were thought to be
impossible, because in Europe all swans are white. "All swans are white"
thus became an irrefutable assumption. Now there were no fortunes made or
lost because of that assumption and its refutation by the existence of
black swans in Australia; but it is very much a warning that the map is
not the territory. Your model of the world is not the world.
There are realms where there are no
Black Swans. When you shoot craps in Las Vegas, the rules don't change in
mid game; the probabilities are known to many decimal places. The pass
line gives the house 1.41% advantage; but if you get a number on the first
roll, you are permitted to back up your bet at the true odds and be paid
at the true odds (6:5 on 6 & 8; 3:2 on 5 & 9; 2:1 on 4 & 10) if you
win. This dilutes the house advantage (for
full explanation, see this posting). The statistical model of craps at
Las Vegas is exact, and you can calculate the expected value of what you
do at the crap table. You can choose the proper strategy (pass line and
back up your bet) or you can choose a strategy that gives the house a much
better chance to win (Field, Hard Way, and various other proposition
bets). A Black Swan in this case would be a raid on the house with the
police confiscating all the money on the table, or a robbery by Southsea's
Nine, or a hole opening in the floor that you fall through. But note that
the model of what happens at the table is exact because this is a game. It
also assumes that the house will not bring in loaded dice.
That was the point of Taleb's essays and
books, and that is the point I am trying to get across to my readers. It's
an expansion of the General Semantics axiom that the map is not the
territory; and there are realms in which we have no decent maps, and in
those regions are territories that on the map ought to be noted "Here
there be Monsters" or some such. We do not have good models of a market
place, nor are we likely to have them, because there are intelligent
opponents in the game, and they have choices. No battle plan survives
contact with the enemy.
The experts and stock analysts know what
they are doing in some parts of the realm in which they operate because
their maps do correspond to the territory; but as the real world
repeatedly has shown, they have no better idea of where their models fail
than anyone else does. They also have strong temptations to change
the rules just as the house can be tempted to bring in loaded dice.
I spend a good bit of time on this
because it's something I want my readers to understand. After all, you
can't subscribe to this web site if you've been wiped out.
The proper way to think about Black
Swans is, "Don't bet all you have on a proposition whose only proof is
empirical observations and confirming instances." That's an
oversimplification, but it's worth remembering. And it has nothing to do
with "blaming" Black Swans.
==
Of course if your point is that those
who made this mess are now blaming a Black Swan that suddenly and
unexpectedly appeared from nowhere, then of course that is "blame"; but I
haven't heard that this is happening. Indeed, I would be happy to learn
that the Masters of the Universe have finally learned that in this world
there are Black Swans, and that their maps are not the territories.
Note also that the concept of a Black
Swan event applies in climatology, and in much of our everyday lives.
Highly improbable disasters do happen.
"In the last analysis, luck comes only
to the well prepared." Helmuth von Moltke the Elder
===========
"The findings indicate that even
those people who appear to be very committed to environmental action find
it difficult to transfer these behaviours into more problematic contexts."
http://www.guardian.co.uk/environment/2008/sep/24/ethicalliving.recycling
- Roland Dobbins
Most green activities are irrelevant to
the real problems, because the green map is not the territory. Mind you, I
don't say we have adequate maps; indeed most of what I write is on the
subject of getting better maps (better data and theories of climate
change) instead of pouring resources in what may or may not be useful
remedies.
If we are truly concerned with CO2 in
the atmosphere we need do develop ways to remove it; If we are truly
concerned that there is excess solar heat entering the Earth's climate
system, we need to find ways to reduce the amount of solar heat that come
into the climate system and stays here. (Google albedo for some helpful
suggestions.)
=========
Fred's Experience in Mexico
Ever read this about Los Angeles?
McPhee, John, 1990, The Control of Nature. One of the
articles is Los Angeles against the Mountains. It describes how the city
deals with similar problems.
-- 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>
Read it years ago, have forgotten the LA
part (it's mostly about the Mississippi and New Orleans). I'll look it up.
It's somewhere around in the swim here.
===========w
TOP
CURRENT
VIEW Wednesday
|
