Archive for the ‘Technology Politics’ Category

Communitarian in Chief

Monday, July 27th, 2009

The brother of White House chief of staff Rahm Emanual is named Ezekiel J. Emanuel. Ezekiel is a medical doctor, who recently has worked in the Office of Bioethics at the National Institute of Hearlth, which focuses on issues like when medical care should be withdrawn and the patient “allowed” to die.

Dr Emanuel is a self-described Communitarian. What is a Commitarian and what do they believe?

Probably the short handed explanation is the quote from Star Trek: “Some times the needs of the many outweigh the needs of the few”, or more to the point – individual rights must give way to the needs of the collective.

Dr Emanuel is a key advisor to President Obama on the proposed National Health Care policy.

Dr Emanual’s thoughts are written down for posterity in The Ends of Human Life: Medical Ethics in a Liberal Polity. Cambridge: Harvard University Press, 1991.

This is definitely a person you want to keep an eye on.

Book Deleting

Monday, July 27th, 2009

From time to time the subject of “book burning” comes up when some parent objects to the content of a book in a library or a school.    Setting aside that there are two types of banning – the librarian who refuses to buy or accept a book, or quietly disposes of a book that they have a personal or political objection to, or liberal patrons who “forget” to return all 37 copies of Rush Limbaugh’s books they borrowed and the library never replaces the book…

Well, that subject is now out in full view.   With a paper book in your possession, the words printed on the page stay there as long as you have the book.    But what about the new book reader from Amazon called Kindle that lets you download and read books electronically?

The “stuff” is hitting the fan.   According to the New York Times, Amazon was selling copies of the book “1984” by George Orwell, from a publisher who did not have copyright permission to sell the book. Amazon remotely deleted the copies of the book from People’s Kindle.

Putting aside the irony for the second, this raises the question of just how far Amazon or other publishers might go in the future (or be forced to go by governments) to delete… or even worse…. modify the text… of books for political or other reasons.

What do you think?

Here we go again – 47 injured in Muni Crash

Sunday, July 19th, 2009

The Muni system in San Francisco was the scene of a collision at 2:30 PM Calfornia time, with 47 people injured, 20 of them seriously
http://www.sfgate.com/cgi-bin/article.cgi?f=/c/a/2009/07/18/MN3C18RJGG.DTL

Some background – because of its geography and dense population, San Francisco was one of the last major cities to have a substantial streetcar system. They used the PCC type streetcars that were invented just before World War two, and deployed widely after the war was over. Most of the lines ran on the very wide Market Street and looped at the Montgomery Station. Prior to BART, the Oakland Bay Bridge had a streetcar line that ran across the bridge.  

When BART was being designed, the decision was made to take the streetcars off Market Street and bury them in a tunnel, and the BART line in the tunnel a level below the streetcars. The streetcars were replaced with more modern LRV cars. (Interestingly, the PCC cars were brought back to Market Street as the “F” line as a heritage line, and has been wildly successful).  Being a transit geek, I’ve travelled to San Francisco several times just to ride BART and the PCC cars.

Here is some home movies of what Market Street was like when it was filled with Streetcars, Trolley buses, Diesel Buses and cars.

With the large number of trains running during the weekday, and the small lead times, MUNI decided to install automatic train operation systems from the Canadian company Altacel to allow the computers to control the operations in the tunnel, allowing 48 to 60 trains per hour instead of the 23 when done by human operators. Once the cars emerge from the tunnel, they run on city streets among cars, so need human operations.

The first few weeks of the MUNI operation were a totally disaster, with cars refusing to move, frustrated operators trying to override the system, and passengers fuming. Eventually, the bugs were worked out and it has operated pretty flawlessly since then – until today.

The reports indicate that one moving train rear-ended a stopped train at high speed. The driver of the moving train is one of the most seriously injured. It’s worth noting that this happened on Saturday afternoon, when trains would be few and far between and operating with large headways. The accident occured at the West Portal of the tunnel under Twin Peaks That’s the point at which the Train Protection ends and street running begins. Muni System Map
The Twin Peaks Tunnel is 2.27 miles long… plenty of time for a driver to be distracted and not watching the space in from of them since there is nothing to do in the tunnel for maybe 3 minutes and in the tunnel the train is still under automated control, and on a Saturday it’s not like there might be train right in front of them.

A blogger in San Francisco caught the damaged train being hauled away Pictures Here

Video of the general area of the crash is on Myspace here. The accident happened outside the tunnel underneath the roofed in area.

There was a similar accident in June 2008, when a Muni train hit the train in front going 17 mph, in a 3 mph zone. Accident Account

In August 2008, a Muni driver rammed the historic streetcar on Market Street, an hour after being warned she was following the streetcar in front too closely. Story

Early reports (usually wrong) suggest the driver of the moving train gestured that he wasn’t able to stop the train, or fainted, or the train in front of him stopped unexpectedly

Maybe President Obama should propose a Smarter Public Transit System.

“New GM” recalls workers – to build luxury SUVs

Tuesday, July 14th, 2009

Well, not technically SUVs.   The new term is “Crossover” vehicles.   It’s a luxury vehicle with all the bells and whistles, but with a hatchback and room to carry lots of stuff.

The Detroit News reports that the factory in the Lansing Michigan area will recall 900 people to restart the second shift. If you haven’t heard of the Builck Enclave, here is their website. The price starts at $35,000. Standard features include:

  • 4975 pound curb weight
  • 4 Wheel Drive, 18″ wheels
  • Seating for 7
  • 288 horsepower V6 engine
  • Bose Stereo with XM Radio
  • Mileage – 16 City, 22 Highway
  • Remote keyless entry
  • Onstar

Optional features include:

  • Rear parking assist

You people just haven’t got the message yet!

On a slightly related note, the LA Times reports that US Government owned Bank of America has announced it will finances purchases of the Tesla Electric car, a $109,000+ sports car.

Abusing technology

Tuesday, June 9th, 2009

The Air France crash (and my current belief that the pilot flew the plane right into the storm on autopilot – and possibly a problem with the airspeed detector caused the plane to stall) has provoked me to tell a related life story of a very similar “accident waiting to happen”.

I’m going to overexplain this (which I’m got at) because you probably don’t know or don’t care about the technology, even if it affects you personally.  I’m a public transit geek – or at least I was.   For about a year, I worked in Chicago and rode the “El” to work every day (about an hour each way).  Unlike many other transit systems, on the CTA’s “El”, if you get in the first car, you can sit opposite the motorman/driver and watch him drive the train and look out the front window.   Since I got on the train at the end of the line, it was usually easy to grab that seat.

Chicago’s subway/elevated lines have what is called “cab controls” on much (but not all) of its lines.   Over the years, there have been lots of accidents involving driver errors (in Chicago and elsewhere – most recently in Boston), where the driver did not obey a stop signal, driving too fast, being distracted by a pretty woman, etc..  and plowing into the train ahead or going around a curve too fast.   So the obvious “fix” to the human problems is to solve it with technology.

The Cab controls prevent the train from going faster than the permitted speed – the train will disengage the motors and put on the brakes.   A separate older system protects the trains from running through a red light.  There is a mechanical trip – if the train proceeds through a red signal, the arm trips the emergency stop on the train and brings it to an immediate halt.  What could possibly go wrong?

Chicago still uses 2 man (oops, person!) operation, which is somewhat rare.  Part of it is probably union rules, part of it is that Chicago’s system is very old.  The cars have doors on both sides because some of the stations have the platform on one side, and some on the other.  The motorman’s job is to drive the train.   The conductor’s job is to open and close the correct set of doors at each station.  The conductor is typically in about the middle of the train.  Late at night after the stations close, the conductor also collects fares if necessary.  They also (in theory) are a deterrent to crime.  They aren’t armed, but they do have a radio.

By splitting one job between two people, that creates a real safety problem.  What happens if the driver starts to move the train and the doors are still open?   What if the train starts moving and at the same time the conductor decides to reopen the doors as someone is running toward the train.   Someone dying is a very probable outcome.  

So technology has the answer – the safety interlock.   To make sure this can never happen, the trains have a built in system to prevent it.   The train will not move if the doors are open, and you can’t open the doors if the train is moving.  Problem solved.

Until you realize that anyone who has ever designed an idiot proof system has not met very many idiots.   The problem is that virtually all of the drivers now let the safety system drive the train.    They just push the power control to max and then let the cab controls push back and slow the train if it is going too fast.  Because the drivers are not in control, they make no judgments about speed (that’s not bad by itself), but that can lead them to become more and more careless about the trains operations.  The only way the train is under manual control is if the operator puts it in manual.  That triggers an alert back at the operations center, and the train is limited to only about 5 mph.  It’s intended to be used only to move trains in and out of the storage yard, or where cab control is not in effect or has failed for some reason.

One day in particular, an operator was doing an incredibly stupid thing – no doubt against the “rules”.     As the train would stop in each station and the doors opened, he pushed the power control to max while in the station with the doors open.   When the doors would close, the safety interlock would disengage, and the train immediately take off at maximum power.  He had taken a safety device meant to protect against a human mistake and converted it into a way to be lazy – introducing a huge risk that if the interlock ever failed, he was likely going to kill some people.   The interlocks frequently failed, but usually the other way – the doors would be shut, but the system thought they were still open and wouldn’t let the train move – the conductor would have to cycle the doors open and closed to get the train to release.

So if you realize how human behavior adapts to not being responsible and in control, it isn’t a big leap to imagine a pilot on plane that can fly itself seeing a thunderstorm on radar and saying “the plane can do a better job than me, it’s flown itself through storms before”…  let’s sit back and let it handle it.

To be clear, I’m not necessarily blaming the pilot.   Had he overridden the computer navigation, he may have had to fly 100s of miles out of the way.  He might not have had enough fuel to land at his destination with an adequate safety margin (extra fuel = more weight = less efficient).  It could be that he would face punishment for doing so.  The Captain is always in charge unless someone else tells him he isn’t.

*** Washington D.C. update ***
6 killed today as one Metro train plows into another that is stopped in a station.   Media reports are blaming a failure of the signaling system, rather than the operator not noticing the impending crash and going into emergency stop.

Just a guess, but the most likely cause of this type of accident is when an engineer overshoots a station, then backs up the train on manual.   That puts the lead train into a “block” that is supposed to be empty.  The following train doesn’t know that because they already went past the green “all clear” signal.   The witness reports all agree that the moving train did not apply brakes at all prior to the crash.   I’ve ridden the Metro, and its cab control is very thorough – the only thing the operator does is to close the doors.  The train controls the acceleration and braking.   It’s very plausible that the operator could have been playing video games on her cell phone and letting the train make all the decisions.

A detail as you look at the pictures and read the accounts – you’ll see 4 tracks, with the Metro train on the inner track.   The outside tracks are NOT metro – they are the Amtrak line that runs into Union Station.   That probably greatly complicated the crash scene and the risk that an Amtrak train could have plowed into the wreckage – fortunately that didn’t happen.

A different theory about Air France crash

Wednesday, June 3rd, 2009

The current generation of large commercial jetliners barely need a pilot, other than to reassure the passengers.  The savings in fuel costs of the latest generation of planes are because most of the decision making is taken out of the hands of the pilot.  the plane decides based on the route, amount of fuel on board, weight of the passengers and cargo, and the wind conditions, etc… how fast to climb, how much throttle to apply, etc… so the plane arrives at the destination right on time and with the minimum amount of fuel consumed. 

The Airbus is a “fly by wire” aircraft – the controls in the cabin are nothing more than what a joystick and mouse do on your computer.   They are just control inputs into the computers – they aren’t actually connected to anything.  (your car probably has at least several controls that are also like that – your car doesn’t have a rod from the gas pedal to a carburetor – your “gas pedal” is just an input to the computer that controls the engine and the transmission.

So here is my crazy theory – the pilots were asleep – and the plane flew itself right into the thunderstorm.  By the time the pilots realized they had a problem, it was too late to solve.  The plane was 4 hours into its flight, cruising at 35,000 feet – it was beyond the range where ground radios would be pestering the pilots, and it was the middle of the night.   The plane flies itself, and will even prevent pilots from taking actions that the computer decides are unsafe.  Large thunderstorms in the middle of the night are a bit unusual.  The stories of pilots falling asleep are more common over the Pacific Ocean where the flights are even longer and more boring.

If you have cruise control on your car, you’ve had a small dose of this.  With a car with cruise control, you can become very inattentive to the journey…  you snap out of it and think…. wow, I just drove for 2 hours, and I don’t remember a single thing that happened.    Probably if some danger was apparent, you would have become more alert and reacted, but maybe not.

The unintended consequence of too much technology is you expect it to always work, and lose the ability to cope when the technology fails.

*** Update ***

My theory is looking really good at this point.  This report mentions that Air France does have the final 4 minutes of flight data from the satellite link.   The pilots took autopilot off the same minute that the plane started to break up.   They let the plane fly itself right into a fatal thunderstorm without taking corrective action until the plane was already doomed.

Google Earth Street View

Wednesday, April 15th, 2009

If you haven’t tried Google Earth, or haven’t looked at it recently – you really need to take a look.   What you will see should scare the hell out of you.

Google Earth’s “Street View” seemed innocent enough – it started out by having someone in a car driving around San Francisco taking pictures of the buildings downtown and making those pictures available integrated in with the satellite imagery and maps.

What’s changed is the scope of what they have done.   It isn’t just a few big cities with a high curiousity level and interest from tourists – they have now sent their cameras to pretty much every street and rural road in America, taking 360 degree pictures every couple hundred feet and placed it all online on the internet.

Back in the heyday of the Soviet Union, they produced deliberately wrong maps as disinformation to prevent the United States (and possible future and past enemies like Germany) from having an accurate understanding of where cities and roads were and identifying targets of military or economic significance.  Entire cities (like where their nuclear weapons programs were conducted) were not even shown on the maps.

Google Earth has essentially put pictures of every road, every house, every business in the Entire United States on the internet for anyone in the world to access without restriction.  With 2 pictures of the same object from slightly different locations, it is possible to create accurate 3d measurements.   

Google is not the government (yet), but this should cause you to have some thoughts about the conflict between the freedom of the first amendment and the government’s need to provide security.   Some countries in Europe are starting to push back on Google Street View.

What do you think the United States should do?  (if anything)

Ethanol raises cost of Government Food Programs

Friday, April 10th, 2009

AP Story Here

Who could have seen this coming?  By converting farm land to making ethanol from corn, the price of food has gone up – requiring the government to spend more money on food stamps, WIC and the school lunch program.

Of course the executives from “Big Ethanol” say that isn’t true.

Arab Emirates to buy Textron – any problem with that?

Thursday, April 9th, 2009

Bloomberg

On a post over on the radio blog, I detailed how Textron got itself entangled in the Radio business.   Textron’s main products are Cessna airplanes and Bell Helicopters.  Textron is also involved in designing control systems for other military weapons.

Bell Helicopters are used by the US military and they were involved in building the new fleet of presidential helicopters that was just cancelled.  Textron was already obviously in finanical distress before that announcement.   The “war on private airplane travel” particularly hit Textron hard.

So today’s news is a group in the United Arab Emerites wants to buy Textron.  

Anyone have a problem with folks in the middle east owning the company that makes aircraft and weapon control systems for the US military, and  airplanes?

After all, we’re not at war with Islam – there is no risk here – we’re all part of the same community.

Why there is no going back now

Thursday, April 2nd, 2009

A few of you are aware that in a previous life, I worked for a major equity trading venue (not the NYSE – think #2) doing programming and software design of electronic order matching (trading) systems.  People who personally knew Bernie Madoff from when he ran that market in the early 1990s made it clear they would rather I leave and stop pestering them about planning for the future – things like in 2000 and 2001 having a real disaster recovery plan in the event of a terrorist attack that destroyed our facility) – so it made my decision to leave in 2002 that much easier.

Many politicians – and radio people like Bob Brinker and Michael Savage – believe one of the root causes of the past year’s collapse of equity prices is to the end of a quaint rule called the “Short Sale Uptick Rule”.  

Without getting completely boring, the purpose of this rule in the 1930s was to allow the person who manages a stock on the NYSE (call the Specialist) to prevent a sudden fall in the stock of a price – by people who wanted to bet that the price will go down further by creating a stampede.   70 years ago, this was a really great idea to avoid market instability and prevent 1929 from happening again. 

If you wanted to sell IBM stock 40 years ago, your order was sent to New York on a teletype, printed out, hand carried to the specialist on the floor, and he would find a buyer or if nobody was buying – the specialist would directly buy your stock at a price that HE set.   Pricing information was delayed 15 or 20 minutes to the general public, so he had a lot of time to “work things out”.   Your sale reported on the “stock ticker”, a machine at your local stock broker so they could keep somewhat current on the day’s trading.

By controlling what traded when and at what price, the specialist (one person) could block a surge in selling using the “tick” when he did not want to be the buyer  (because he also believes the price is headed down and wants to get the prices adjusted lower first and get out of the way in the 15 minute window before the entire world found out).  

So if this rule was working so well to prevent market panics, why did the SEC allow exchanges to get rid of it in 2007?  The main answer:  Electronic trading.   Only the most inefficient markets now involve human decision making to routinely manage the trading of a market.    Note that “program trading” and “electronic trading” are different things.  Program trading is the computers for a customer making automated decisions to buy and sell.  Electronic Trading is the actual matching of buy and sell orders.   You don’t need electronic Trading to do Program Trading – but it does make it faster and potentially more dangerous to market stability.

With automated trading in 2009, the last sale price (which controls the uptick rule) can go up and down 100s of  times a second.  That crawl you see on the cable TV channels is no more real than the Pirates of the Caribbean ride at Disneyland.   It’s nothing but a snapshot of some recent stocks at a speed the human mind can absorb.

A TV picture only updates 30 times a second (computer monitors typically are 70-90 times/second). You just can’t enforce a rule like the uptick rule – even if you still thought it was a good idea.  A human being cannot absorb information at 100 times a second and certainly can’t click a mouse at the “right” time to not break the rule.  

Just how is an electronic market so different in 2009?  Here are the specs for the Universal Trading Platform  that NYSE Euronext is rolling out in Europe, and recently replaced the ARCA trading system that trades NYSE stocks in Chicago (not New York, Andrew Cuomo)

“Customers trading on NYSE Euronext European cash markets will realize reduced latencies […] of 150-400 microseconds per roundtrip on the Universal Trading Platform . In addition to delivering exceptionally fast transaction speed and system-wide reliability, the Universal Trading Platform sets new industry standards for capacity with the ability to handle 100,000 orders per second.“   (The system my group designed in 2001 was shooting for a target of 150 matches per second)

To put this in context, light (and electricity) travel at 186,000 miles per second.   In 400 microseconds, electricity travels only 74 miles.  400 microseconds is their “worst case” total time for the Euronext UTP to get an order, match it and give the answer back to the buyer and seller. 

There are other reasons the Uptick rule also makes no sense in 2009 – because stock options, single stock futures, ETFs (like SPDRs and contra ETFs that bet on an index going down) and a variety of other financial products allow people to “bet the market will keep going down” without any regard to whether the last sale that happened 572 microseconds ago was at a better or worse price than the 2nd older sale.  

Making things even uglier – there is no such thing now as “the exchange” where a stock is traded – no more than there is an “American Car”.  There are multiple trading platforms all trading the same stock at the same time all over the world.    By the time each trading venue “hears” about the last trade at another exchange, it is probably no longer the most recent trade.

This is why I described the NYSE trading floor as nothing but a TV studio living out the quaint fiction for the masses that there is still a trading floor.  Any humans you see on the floor are there for very little reason in 2009, and with each passing day there is less and less reason for humans to stand around on “Wall Street” in New York City.  

Mayor Bloomberg, Governor Paterson and Andrew Cuomo have no leverage over “Wall Street”to keep them under the jurisdiction of any part of New York State goverment – they just are too blinded by their own desire for power to realize it yet.   That’s especially odd in the case of Mayor Bloomberg – since his fortune was made on selling information to people on Wall Street, and knows every detail of the business. 

The Federal Government only has nominally more control because it controls the creation of US dollars by the US Treasury and Federal Reserve.  That power card has now been put on the table for the last time.  Globalism is going to win this fight, and that’s not a good thing if you believe in the values of the United States, individual freedom and representative government based on our national interests. 

But to finish on a positive note, Congress did solve that problem with Baseball Players taking steroids and Michael Vick is in jail for those dog fights.    We need to keep the inportant issues in focus.   Who do you think is going to win American Idol this season?