* Entries include: railroad infrastructure, Florida road trip, Chinese corpse brides, African micro-hydro, astronaut attitude problems, natural house construction, new San Francisco Bay Bridge, algorithms as business tools, the law and electronic discovery, Ethiopian commodities exchange, RFID introduction delayed, Boeing 787 Dreamliner, third-world economic benefits of cellphones, shady DNA analysis by mail, immobots, and Chick tracts.
* CORPSE BRIDE: In the "news of the bizarre" category, THE ECONOMIST reports ("Wet Goods & Dry Goods", 28 July 2007), on the Chinese custom of "ghost marriages". The Chinese do not like to see a son die unmarried, but such things do happen -- no problem, the parents just bury a dead woman alongside the son after a postmortem marriage ceremony.
The Communist regime generally suppressed ghost marriages, but now that things are gradually relaxing, the custom is reappearing, particularly in the rural areas of northern China. The practice is not necessarily corrupt, it being possible to make arrangements with families whose daughters have died unmarried and have a common interest in making arrangements for the afterlife -- but it has led to an upsurge in grave robberies. Recently dead or "wet" female corpses can fetch the equivalent of thousands of dollars on the black market, a hundred times more than a long-buried "dry" corpse.
Given the financial incentives and the difficulty of digging up corpses, it was no great surprise when early this year, the Chinese authorities arrested a Song Tiantang for strangling six women and selling their corpses. He had been arrested a few years earlier for grave robbing, and had decided to move on to more efficient tactics. He was caught after losing his cellphone at the scene of a crime.
* This story reminded me of another that I had heard decades ago, about two Britons named William Burke and William Hare. Early in the 19th century, the only legal way to obtain human corpses for medical dissection was from executed murderers, and since the supply of such was too small, researchers often paid off grave-robbers to obtain specimens. Hare ran a boardinghouse in Edinburgh, and in 1827 Burke took lodgings there. The two got on friendly terms, and when one of the other tenants died while owing Hare money, Burke and Hare sold the corpse to a University of Edinburgh medical researcher named Dr. Robert Knox, who asked no questions.
Burke and Hare then went into the business of supplying corpses to Dr. Knox. The two would bring in beggars or prostitutes, get them passed-out drunk, and then Burke would suffocate them, leaving no marks. However, people started noticing the disappearance of local characters; then some of Knox's students recognized one of the murdered prostitutes, apparently having made use of her services, and in 1828 Burke, Hare, and Knox were arrested. The evidence against them was not very strong and so Hare was encouraged to "cop a plea", as they say these days, providing evidence against Burke. Burke was hanged while Hare and Knox went free, though they had to leave town in fear of the wrath of the citizenry. The story survives in an obscure and creepy nursery rhyme:
Up the close and down the stair, In the house are Burke and Hare. Burke's the butcher, Hare's the thief, Knox, the man who buys the beef.
The story of Burke and Hare gets recycled in horror stories every now and then, and apparently there was even a DOCTOR WHO radio show in which the time-traveling Doctor met up with the infamous duo. In any case, an "Anatomy Act" was passed by Parliament in 1832 that allowed dissection of unclaimed corpses from workhouses and the like. The supply was plentiful and there was no longer any financial incentive to rob graves, much less commit murders, to supply the needs of anatomical research. Sometimes I wonder if we sleep better for not knowing how the real world works.
BACK_TO_TOP* MICRO-HYDRO: While developed-world technologies tend toward the clever and sophisticated, developing-world technologies tend to have an ingenious and resourceful fascination of their own. A case in point, as described in an article in IEEE SPECTRUM ("Thirst For Power" by G. Pascal Zachary, May 2007) is a minuscule dam in the mountains of Uganda that provides all of 60 kilowatts of power.
The dam is a simple concrete wall above a natural waterfall, with the wall diverting the stream flow into a reservoir. The reservoir drains through a pipe into a little bungalow where the generator resides, with its power run off directly to a local hospital. The generator is run off an "impulse" turbine, a relatively crude device in which a jet of water spins a set of buckets on an axle -- it can be made by local artisans and doesn't need a housing. Electric power makes all the difference in the world to the hospital. Says Sabuni Seezi, who takes care of the little "micro-hydro" system: "The government has promised and promised to bring electricity to this village and never has. So we did it ourselves."
All of sub-Saharan Africa has about 30 gigawatts of electrical generation capacity, about as much as Poland. Traditionally, African governments have liked big hydropower schemes, going back about half a century to the Aswan Dam in Egypt on the Nile, which produces 2.1 gigawatts, and the Akosombo Dam in Ghana on the Volta, which produces 768 megawatts. Uganda has a large dam on the Nile at Owen Falls, producing 200 megawatts, and is planning two more.
However, the big dam projects have major limitations. They are expensive and troublesome to build, and corruption tends to considerably aggravate both the expense and the difficulty. They may be linked to grossly inadequate power grids, with the capital city of a country getting all the power and nobody else getting any. Smaller dams, providing 15 megawatts or less, may well be more cost-effective, providing power to people where they need it, but central governments have not generally been enthusiastic about the idea.
That is the beauty of micro-hydro projects: they can be implemented by locals without greedy and inept government officials getting into the loop. Micro-hydro systems are already popular in Vietnam and China. Enthusiasts believe Uganda alone could support thousands, with a potential for tens of thousands across Africa. With backing from aid organization, Africans may be able to give themselves the power the governments can't.
BACK_TO_TOP* FLORIDA ROAD TRIP (4): I left the motel in Mobile, Alabama, not too early on the morning of Sunday, 16 September. Since I had to make a stop at the battleship USS ALABAMA memorial on Mobile Bay, and it didn't open until 8:00 AM; I could afford to sleep in a bit.
The ALABAMA memorial was definitely a marginal stop. The battleship itself was interesting enough and I got lots of shots, particularly of its radar gear and its floatplane catapult. However, the park was littered with other military equipment -- tanks, artillery, aircraft -- in rusty and dilapidated shape. In other words, it was mostly just a "boneyard", an infuriating sight to enthusiasts: "Damn you, why let all this irreplaceable gear go to hell?!".
This seems to be typical of museums with armor and artillery displays -- usually the stuff's just rusty and beat up, hardly worth the bother to visit. However, there were a few aircraft in a hangar in good condition that helped make up for that, including a YF-17 -- the prototype for the F/A-18 Hornet, a very unusual find since only a few YF-17s were built -- and some nicely preserved target drones.
That done, I drove down I-10 towards Pensacola, which was about an hour's drive away. On leaving Mobile I noted a ruined gas station, which brought back memories of the dilapidation I had seen in parts of Kansas -- but on a second glance I realized that the station hadn't been the victim of slow decay, a force of nature had torn it apart. It was a reminder that I was in hurricane country, and I would occasionally spot other victims of storms during my travels in the region -- billboards that had been ripped away, leaving their supports twisted. "Nah, I don't want to be here when the wind's blowing."
I found the naval air station in Pensacola easily enough, and to my relief access to the base was open -- there are a number of public attractions in that area of the base and so there's little restriction on getting in and out. I had been worried about being given the third degree at the gate; the usual greeting at a base gate for a civilian these days is, not without good reason: "Who the hell are you and what the hell do you want?" However, all I encountered was a very polite and well-spoken Marine guard who handed me a pass and waved me through. I was relieved.
The museum was very much worth the visit. It was relatively small, compared to the big Air Force museum in Dayton, but it had an excellent collection in superb shape. The big problem was that there was so much crammed into it that it was hard to get good shots, but I canvassed the place methodically with my camera and tripod for about two hours -- I was done just precisely at that moment my camera batteries went dead. I had spares but the timing was amusing anyway. There was a tour of the flightline where aircraft were awaiting restoration, but that would take up much of the afternoon and I didn't have time to wait -- I had other stops to make on my way to central Florida. In hindsight, I should've taken in the flightline tour.
I had planned to make a stop at the Armaments Museum at Eglin AFB, not far to the east, but while I was plotting out the trip with my detailed road atlas, to my surprise I found a zoo on the coastal highway to Eglin. I checked up on it and it turned out to be the "Zoo of Northwest Florida", and since I would be driving right past it there was no way I could have avoided stopping. As it turned out, it wasn't an attraction I would recommend: I like to visit zoos, but of all the many zoos I've visited this was the saddest, the most dilapidated and uninspired I'd ever seen.
I did manage to get some good shots of a peahen, something I'd been after for a while, and some very nice shots of alligators. They had a little pond of gators, none of them bigger than large dogs, and for a price visitors could dangle bits of food down to them on a line. The gators would snatch the food with the nasty sound of a trap snapping shut -- the things were fascinating, little creepy monsters. Still, it was a pathetic establishment overall, and from the appeals to "keep the zoo alive" posted here and there, it was clearly not in good financial health. My feeling was that they should put the thing out of its misery, or at least sell off half the property and set up a pocket-sized zoo, like the cutesy mini-zoo I saw in Idaho Falls, that they could properly maintain.
I didn't stay long and got back on the coastal highway -- a congested strip, interrupted by many stops, lined with beach houses and the like. I did spot something interesting just offshore, a very tall framed tower with a dome structure on top. I stopped to get shots, though I didn't have any clue as to what it was. Later surfing on the web suggested it was a radar tower, supporting the offshore target range areas used by Pensacola NAS and Eglin AFB. It must have been very sturdily built to withstand hurricane-force winds.
I finally made my way to the Armaments Museum, where I had my first real bungle: it was closed on Sundays. I was so used to thinking facilities like this were open all week that I hadn't checked carefully. It wasn't a complete loss, since they had extensive exhibits of aircraft and munitions all around the building -- the "gate guard" for the museum was a old British 10-tonne (22,000-pound) "Grand Slam" bomb (built by the US as the "T-12"), and I got good shots of an F-111 and a B-57, both of which I had been wanting to get into my picture collection for a while.
The fact that the museum was closed was something of a disappointment, but there was nothing I could do about it, and I had a six-hour drive to central Florida ahead of me. My first target in central Florida was Tampa, but since none of the attractions there opened up until 9:00 AM I didn't see any reason to drive all the way, and I spent the night in Ocala, about 90 miles short of Tampa. I could drive down from Ocala in the morning and balance the time load a bit. [TO BE CONTINUED]
START | PREV | NEXT* INFRASTRUCTURE -- RAILROADS (10): Railroads work on a hub-and-spoke scheme, with cars carrying various cargoes consolidated into a long train for hauling between hubs, and then resorted at hubs for delivery to the final destinations. The resorting is done at the "freight yard", also known as the "classification yard".
There are actually two types of classification yards -- "flat yards" and "hump yards". A flat yard is the straightforward approach: a switching engine shuttles the cars around and links them up together. A hump yard is a bit cleverer: a switching engine pushes a train of cars to the top of a low hill, with the cars being disconnected one by one to roll downhill and fan out among a set of switches. A hump yard requires good coordination in switching, and also requires some means to keep the cars from building up too much speed as they roll downhill.
Speed control used to be performed by a brakeman riding on each car, who adjusted its speed as other workers turned the switches to divert the car appropriately. Playing brakeman was a very dangerous job, since the brakes didn't always work properly. These days, the braking is done by "retarders", which are jaws on the inside of the rails that apply friction to the wheels of cars as they roll past. The retarders used to be controlled by workers who eyeballed the speed of the car and actuated the retarders to provide the appropriate braking. These days, it's automated under computer control, with radar units in the yard to gauge the speed of the car and scales to determine its weight, and the computer actuating the retarders appropriately. About the only manual feature in the process is decoupling the cars.
The computer system also handles activating the switches. Watching a classification yard in operation can be fascinating, with the cars released one by one to fan out to their appropriate designations. It's all very silent, except for the occasional "BOOM" when a humped car slams too hard into a stationary car. Humping, incidentally, is too dodgy for some types of cars or cargoes, and one can occasionally see the notice DO NOT HUMP stenciled onto a car.
The computer system keeps track of the cars and the cargoes they are carried through "radio frequency identification (RFID)" technology, using what are known as "Automatic Equipment Identification (AEI)" tags, which consist of a module mounted on each side of the car. AEI ID tags have been mandatory on freight and tanker cars since 1995. The tags are read by trackside sensors that relay the location of the car and its ID code to the computer system; the sensors also scan the wheels of cars and perform other observations to see if something is wrong. The computer system uses the ID code to index a database and get the relevant information for the car's cargo and its destination.
Some old cars will also carry a multicolored coded pattern on the side. In the 1960s, even before barcodes were introduced to supermarkets, there was an attempt to implement a barcode system based on these patterns. Unfortunately, trains tend toward the grimy and are also victimized by spray-can wielding taggers, and the barcode system just didn't work. The lingering barcodes are fossils of an extinct industrial technology.
The classification yard is associated with a hub node in the network; sorting out cars for delivery to end users at a spoke node is a different process. A local freight engine shuttles a train of cars around to different end-user sites, dropping off cars at the appropriate locations and picking up unloaded cars. Trying to figure out the most efficient scheme for mapping out the deliveries and pickups is complicated, though software is now available to help.
* Classification yards are highly automated and are becoming more so, with switchyard engines now being introduced that don't have crews. The rationale for this is safety, to get people out of the switchyard so they won't be crushed. The traditional system, with switchyard controllers talking to switchyard engineers over two-way radios, is very prone to misunderstandings and dangerous errors. With automated locomotives, the process is under the sole direction of the controller. One suspects that controllers were people who really liked playing with toy trains when they were kids.
That opens the door to the idea of fully automating cross-country train operations as well. Although the US Federal Railroad Administration is now authorizing automated locomotive operation over certain regions of track, nobody imagines that the engineer is going to be replaced any time soon. Innovation is driving interesting improvements in America's rail system, but few can foresee when the day will come that it will operate so smoothly and seamlessly that humans won't be needed to ride the freight-haulers rolling down the rails. [TO BE CONTINUED]
START | PREV | NEXT* GIMMICKS & GADGETS: THE ECONOMIST had a little sidebar article ("Opal Fruits", 4 August 2007), on an interesting new technology: plastics as iridescent as opal, synthesized by a collaboration of the University of Southhampton in the UK and the German Plastics Institute in Darmstadt, led by Jeremy Baumberg.
Opals are iridescent because they have a "face-centered cubic (FCC)" crystal matrix. All that means is that the basic element of the crystal is a cube defined by an arrangement of 14 atoms or other particles, with an atom at each of the eight corners and in the middle of each of the six faces. This orderly arrangement allows the opal to reflect light in a way that leads to optical interference effects, resulting in the shimmering appearance.
Plastics are long-chain polymers, and it would seem there would be no way to force them into an FCC crystal configuration. It turns out there is, using a bit of nanofabrication. Baumberg's group fabricates polystyrene beads about 200 nanometers across, hardens them with a blast of heat, then coats them in a sticky polymer to be heated again. They naturally adopt an FCC configuration. The wavelength of the light can be adjusted by changing the size of the spheres. The material can be applied as a film.
The researchers do not see the technology as particularly expensive and see it as useful for paints, wallpaper, and the like. The material could also be used to build cheap sensors to check for food spoilage by fabricating plastic films that feature an FCC matrix "doped" with molecules that react to spoilage products, changing the optical properties of the film.
* A BUSINESS WEEK article ("Mom, Let's Talk" by Louise Lee, 13 August 2007) surveyed the latest offerings in cellphones for kids. DisneyCorp is one of the leaders in cellphones for gradeschoolers, with a range of products that functionally resemble adult cellphones, with a bare-bones unit at the low end of the price range and a fancy unit with camera and sound player at the high end, with a comparable range of calling plans. They do come with Disney-decorated cases and the like and the software for the Disney kiddie phones is distinctive, for example providing a "call control" that restricts who the kids can call during school hours. There's also a "family monitor" feature that allows parents to track and constrain the calling activities of their kids.
Modeci's TicTalk cellphone takes a different approach, lacking even a keypad. It's preprogrammed by parents with a menu of numbers, the kids using a thumbwheel to select the desired numbers. The Firefly Mobile's glowPhone is a similarly simplified unit, built in gaudy dayglow colors -- making it a bit more difficult to misplace -- and with only five buttons: pick up, hang up, a menu cursor controller, mom, and dad. There are fears that kids with cellphones may be targeted by thievish bullies, scammers, or pedophiles, but there are also clear benefits for both kids and their parents.
* As reported in THE ECONOMIST in the Nouabale-Ndoki National Park in the Congo, a handful of park rangers have been fighting a losing battle with poachers after elephant ivory, animal meat, and lumber. Now they are setting up a radio-enabled network of sensors called "TrailGuard" to help keep an eye on the jungle. The network includes metal-detector sensors to detect poachers wandering around with rifles, and smoke detectors to sense them smoking meats.
When activated, a sensor sends out an alert along with its location using a treetop antenna, with the rangers using mobile phones to read the alarm and respond. The rangers carry an identification friend or foe (IFF) module to keep them from tripping the alarms themselves. Many people in the Congo believe in magic; the rangers intend to do nothing to disabuse people of the idea that they have some supernatural ability to find wrongdoers. TrailGuard is the invention of Steve Gulick, an electrical engineer turned biologist who left the State University Of New York (SUNY) to form a startup named Wildland Security for development of the concept.
There are a number of such systems in development, some of which relay alarms through messaging satellite networks, that are being tested in other wildlife refuges. Rangers in Costa Rica's Osa preserve are using one to stop hunting of exotic birds and jaguars; wildlife conservationists in the Galapagos want to use them to stop the slaughter of the famous turtles in the island group; and the handful of rangers patrolling the Shavla Wildlife Refuge in Russia hope to use a heat-sensing system to track intruders hunting the preserve's rare snow leopards.
BACK_TO_TOP* ATTITUDE PROBLEM: The Bush II Administration has declared a long-term goal of sending humans to Mars. This is obviously a difficult task; as reported in THE ECONOMIST ("Cosmic Mood Swings", 30 June 2007), one of the difficulties is just being able to stand being cramped into a spacecraft for the duration of the mission.
A minimum-endurance Mars mission would take about 17 months, and in the worst case the round-trip communications time with Earth would be about 44 minutes -- conversations would be out of the question, though email would work fine. Obviously, being holed up in a can flying between worlds is going to be hard on morale. To figure out just how hard, the Russian Institute for Biomedical Problems and the European Space Agency (ESA) are preparing a simulation, in which three groups of six "crew" will inhabit ground-based spacecraft mockups for two years, being observed while they go through the tasks involved in a Mars mission. The crew will have to be fluent in Russian and English and will need to monitor consumption of rations and other supplies for the duration of the experiment.
The Russians have a high level of experience in manned spaceflight; two-thirds of the overall flight years of humans in space have been performed by Russians, mostly through their Salyut and Mir space stations. Russians are perfectly aware that cosmonauts on long space missions can become listless and irritable, a condition the Russians call "asthenization". In some cases cosmonauts have exhibited signs of dementia, panicky hypochondria, and paranoia -- believing that ground control "had it in for them" and blaming the controllers for problems. Russian controllers learned to monitor cosmonauts for signs they were heading for the edge, and made sure that the spacefarers got surprise gifts, as well as phone calls from family and celebrities.
Studies of the emotional well-being of spacefarers on the International Space Station are consistent with the Russian experience. The isolation of a six-month stay on a space station is not going to compare to the more distant isolation over three times that duration of a Mars mission. Those who believe that Mars might be better explored by robots are only encouraged by such considerations to ask: Is this trip really necessary?
BACK_TO_TOP* GREEN HOUSES: An article from BBC.com ("Harvesting Houses For The Planet" by Patrick Jackson) examined efforts to devise more environmentally-friendly housing. Housing construction obviously has an environmental impact, for example contributing heavily to the consumption of concrete, the production of which overall is estimated by some to contribute 10% of the world's CO2 emissions. Along with the production of materials, the costs of materials transport and building assembly have to be factored in, along with the operational costs of the structure in use.
Some "out of the box" thinking on the matter was displayed at the "Think 07" trade fair in London's dockyards district. The trade fair demonstrated various ideas for reducing the "footprint" of housing. Some were fairly common notions, such as energy-efficient lighting and "light pipes" to direct sunlight into dark rooms. Some were a bit less familiar, for example paneling made of hemp.
According to Tom Woolley, a professor of architecture at Queen's University Belfast: "Sustainable rotation crops like hemp are the cost-effective future of building. One hectare of land can produce enough hemp stalk to build a house," he told the BBC News website, "and using about 12% of the UK's set-aside land, you could grow enough hemp to build the 200,000 new houses the country needs. Then you have the fiber and oil for other products." As far as existing buildings go, Woolley believes that the crucial thing is to improve insulation, for example applying a mixture of hemp and lime on old brick buildings, a technology used in France.
Such low-cost, environmentally friendly technologies seem particularly important for the developing world, with 80% of the Earth's population, but with access to less than 20% of the world's construction materials -- as estimated by the UN's Industrial Development Agency (UNIDO). UNIDO's technology promotion unit tracks down cheap, energy-efficient construction technology and introduces it to some of the poorest regions on Earth, searching for imaginative ways of using local materials to cut the financial and environmental costs still further.
According to Vladimir Kozharnovich, who manages the technology promotion unit: "The owners of the technologies often do not know how to market them while those looking for the technologies don't know where to find them. We seek to provide people with technological options which can be adapted to their specific environment."
In Herat, Afghanistan, UNIDO has planned a model village of 100 energy-efficient homes, designed by Indian and Chinese architects in consultation with the local authorities. Each home costs a projected $3,500 USD and is equipped with bathroom, toilet, and solar-powered electricity. Building costs are estimated to be from 30% to 50% cheaper than existing dwellings. However, the plan is at a standstill while UNIDO awaits approval from the donor, Japan. UNIDO is moving forward on a similar project in the Afghan province of Baghlan, using EU funding.
UNIDO promotes portable brick factories made in India as one answer to cheap construction materials. Another project, now under discussion with Namibia, is a Russian technique for manufacturing building blocks out of sand and seawater. Says Kozharnovich: "The precision is very good -- it's like Lego. It is a proven technology which cuts production costs five-fold, and can be used in both hot and cold regions."
Back in the UK, Woolley registers his frustration with what he sees as an excessive focus on finding new energy sources: "Somebody has very cleverly got the vast majority of politicians and the public to think that sustainable buildings is about sticking extremely expensive renewable energy equipment on the roof of the building, which is actually the last thing you should be doing," he says. The first thing is to reduce the demand and produce buildings which are breathable and well insulated and airtight." Woolley admits that pioneering projects with environmentally-friendly materials can be expensive, but believes that costs should fall once the technologies go mainstream.
BACK_TO_TOP* FLORIDA ROAD TRIP (3): I woke up on Saturday, 15 September to continue on I-70 towards the sunrise and Saint Louis, en route to Mobile, Alabama. I wanted to make a photo stop there to get some shots of the Saint Louis Gateway Arch -- the map showed I-70 looped right past it, how hard could it be? I knew it would also be fairly easy to spot, though Saint Louis is a sprawling town and I kept looking for it as I cruised down the freeway. Finally I went around a bend and it popped out into my line of sight: "Ah, there it is."
The site is indeed right off the highway but I had to do a bit of squirrel-caging to find a place to get shots -- I ended up stopping in a church parking lot, though after I had done so I realized that I had gone past a roadside stop set aside for people who just wanted to stop for photos. Got the pix, back on the highway immediately, cruising on south toward Cape Girardeau to the south.
I was a bit surprised to see cotton fields in Missouri, assuming that I would encounter them further south. I was also pleasantly surprised when I went past the Cape Girardeau municipal airport and found it had a very clean-condition Douglas TA-4F Skyhawk two-seat jet trainer as a "gate guard". Although it was a long drive to Mobile, I still had some time to make short stops, so I picked up some pix of the Skyhawk and also of some sports fliers prepping a little gyrocopter for flight. Incidentally, the US Federal Aviation Administration maintains an online database of tailcodes, so I knew I would be able to look up the gyrocopter from its tailcode number to figure out what make it was and even who owned it.
I finally crossed over the Mississippi River into Tennessee and continued on down into the state of Mississippi. Historically, it was an interesting region since there'd been so much fighting along the border region during the war -- Shiloh, Corinth, Brice's Crossroads -- though I didn't bother to stop at battlefields, the terrain being informative enough as far as I was concerned. The drive was all back highways and I worried that it would be a pain to have to go slow through every podunk town along the way, but though that was true in the northern part of the state, the roads were generally open as I went farther south.
I'd come up with a new idea for navigation on this trip. I had been using a detailed, relatively high-budget road atlas for a few years, but though it made a great reference for plotting out a trip segment before departure or figuring out where I was when I got lost, it was far too cumbersome, downright unsafe, to use as a navigation reference when actually driving. As a solution, I went and bought the cheapest and thinnest USA road atlas I could find, and marked out my planned path with blue highlighter pen. I'd put a big paper clip on the page for the states I was going through; it only really covered the major highways and the larger towns, but that was enough to allow me to keep my orientation and track my location while cruising down the road.
I was a bit surprised at the landscape. Mississippi is in the cotton belt and so I figured I'd be seeing cotton fields all the way, but in southeast Mississippi the major agricultural industry is clearly forestry -- there were groves of pines for pulp and maybe plywood, with a little cattle-raising going on in clearings. It was actually very pleasant and green, as well as surprisingly unpopulated -- not something I expect to see when I go "back East". There were obviously towns here and there, but they were not at all obtrusive.
I was also keeping an eye out for that symbol of the modern South, the kudzu vine, a Japanese import that took over. I'd been down South for a short time in the early 1970s and not bothered to notice it, but this time I was curious and alert. I figured from what I heard that it dominated the landscape, but that wasn't so much true as it was just an unavoidable part of it. In some cases, it would overgrow a tree, turning it into what looked like a topiary of a DOCTOR WHO monster or something similarly grotesque.
It was interesting to watch it progressively take over wire fences -- just snaking along the wires where the fence was new, completely burying it and making it look like a sloppy hedge where the fence was old. I was reminded a bit of the sprawling blackberry thickets in western Oregon, what we called "Oregon barbed wire" -- blackberry bushes don't climb and strangle, but kudzu doesn't have thorns. It is said, not entirely as a joke, that kudzu can be fertilized with used motor oil sludge, and cultivated by setting a concrete block on top of it: kudzu likes a challenge, you see.
I finally crossed over into southern Alabama and made my way into Mobile as darkness set in, where I checked into a Super 8 motel for the night. I was feeling pleased with myself -- from Colorado to the Gulf of Mexico, no navigation problems. In fact, however, navigating cross-country is usually not a big issue, it's getting around in cities that's troublesome. [TO BE CONTINUED]
START | PREV | NEXT* INFRASTRUCTURE -- RAILROADS (9): Given that there's so many different types of railroad cars, including some very specialized types, and it's not possible to provide more than a quick survey here:
One car not seen much any more is the caboose. It was once used to give the crew a place to stay warm, as well as store tools. The cupola on top allowed the crew to inspect the train ahead, and a brake pressure gauge was also usually fitted. In an era of more automated trains, the caboose is irrelevant; it has been replaced by the "end of train device (ETD)", which is simply a little box with a blinking light attached to the final coupler. The ETD is hooked up to a brake air line and relays the brake pressure back to the locomotive over a radio link. Some ETDs also have a radar unit to report speed and direction, allowing the train crew to make sure they know where the rest of the train is and where it is going. The ETD is also more informally known as a "FRED", for "flashing rear end device".
Some classes of cars, such as those carrying hazardous chemicals or perishable goods, are also fitted with black boxes that can relay the GPS location of a car back to a tracking center over wireless. These black boxes may include sensors to monitor the health of the car or check for leaks of hazardous chemicals. [TO BE CONTINUED]
START | PREV | NEXT* BAY BRIDGE REVISITED: On 17 October 1989, a magnitude 7.1 earthquake hit San Francisco. Its effects included shifting the 3.5-kilometer (2.2-mile) long Bay Bridge, causing a structural collapse that killed a motorist. The bridge was back in operation within a month, but as reported in POPULAR MECHANICS ("Six Mega Engineering Projects", June 2007), everybody knew something better was needed. The original idea was to simply build a stronger version of the existing bridge, in the form of a truss span supported by piers, but the ultimate decision was to build a "Skyway" span supported by piers leading to a "self-anchored suspension span" -- a cross between a suspension bridge and a cable-stay bridge. In a suspension bridge, two cables are strung over two towers and anchored on the shores of the waterway, with the roadbed held up cables strung down from the main cables. In a cable-stay bridge, cables are strung directly from a tower to hold up the roadbed. The self-anchored suspension span uses two cables strung over a single tower and connected down the span in both directions, with the roadbed held up by cables strung down from the two main cables. Such a self-suspended scheme is less vulnerable in a quake.
The 160-meter (525-foot) tall tower of the 567-meter (1,820-foot) self-suspended span is designed as four columns linked together by steel beams that act as "shear pins" in a quake, soaking up the energy by breaking. The beams can be easily replaced. The sections of roadbed are supported by sixty 6-meter (20-foot) long "hinge pipe beams" with a soft steel core that yield gracefully in a quake, and are also easily replaced. The new Bay Bridge will be able to shift 90 centimeters (one yard) and survive. The 2.1 kilometer (1.3 mile) Skyway span sits on 160 pilings driven 90 meters (300 feet) into the soft Bay bottom to reach a solid footing.
Construction has not gone entirely smoothly, with estimated cost rising from $1.3 billion USD to $6 billion USD. However, when the ten-lane bridge opens in 2013, it will be a structure built to last in one of America's most geologically treacherous cities.
* As per the title of the article, it also discussed five other big engineering projects. Some of them were arrests of the "usual suspects", such as the Burj Dubai skyscraper and CERN's new Large Hadron Collider accelerator ring, but the others were interesting:
* DOING IT BY THE NUMBERS: We tend to think of technology in terms of machines or software, but as discussed in an article in THE ECONOMIST ("Business By Numbers", 15 September 2007), technology can also be embodied in the logical or numerical procedures known as "algorithms". Algorithms are also a surprisingly pervasive technology, being widely used in getting things done. One example is the algorithm used to confirm the validity of credit-card numbers. Credit-card numbers are long and it's easy to make a mistake in entering them; since so many credit-card transactions are being performed at all times, it's important to figure out a quick and easy way of determining if a bogus number has been submitted, instead of trying to access an account only to find out it doesn't really exist. The solution was developed by an IBM researcher named Hans Luhn and is unsurprisingly known as the "Luhn algorithm".
The numbers on a credit card define the card type, the card issuer, and the card account number. The last number is defined to support the Luhn algorithm, which works as follows:
The algorithm is easy to implement and has a low rate of false positives. This is a simple example of the use in algorithms in business. For a more elaborate example, consider the problems faced by United Parcel Service (UPS), which has to efficiently plan the delivery of vast numbers of packages every day. Suppose a UPS delivery truck has to deliver to 25 separate locations. The number of possible routes in this case is very large, since there are 25 possible first stops; 24 possible second stops for each first stop; 23 possible third stops for each second stop; and so on. This gives the number of possibilities as:
25 * 24 * 23 * ... 3 * 2 * 1 = 1.55E15
The issue is further complicated by real-world factors such as specified dropoff and pickup times. UPS has put a lot of money into solving the "traveling salesman" problem, as this puzzle is broadly known. To plan the movements of aircraft, the company worked with the Massachusetts Institute of Technology (MIT) to developed a program named VOLCANO, for "Volume, Location, & Aircraft Network Optimizer". It was introduced in 2000, and is used by three UPS groups -- one to plan for deliveries at the peak season between Thanksgiving and Christmas; one to plan schedules for the next two to four months; and one to plan the resources required -- personnel, facilities, aircraft -- to support business over the next two to ten years. Getting scheduling wrong can be very expensive, either in terms of dissatisfied customers and lost business, or in terms of half-empty leased aircraft; although VOLCANO was expensive to write, it has more than paid for itself in practice.
Many other businesses also rely on such "optimization" algorithms. Telecom companies use them to route phone calls and help retrieve web pages efficiently. Manufacturers and retailers use them to optimize supply chains. Call centers use them to prioritize incoming calls, based on such variables as the reason for the call, the location of the caller, and the length of the queues that the staff have to deal with.
One of the complications of optimization is that the problem may feature several "dimensions". For example, a package-delivery scheme must not only take into consideration the distances between delivery "nodes" but also the speed of traffic along each "vertex" between the nodes -- which may vary considerably at different times of day, for example going to a crawl at rush hour. Where such calculations become particularly tricky is when they have to be adapted in "real time". UPS is now working on a system that will permit changes in delivery schedules while the driver is making deliveries. The concept is much along the lines of that of the way an automotive satellite navigation system adjusts its course recommendations if a driver doesn't take a suggested turn.
Such considerations are particularly important to internetworking. Routing algorithms have traditionally simply focused on connections, without much consideration of congestion -- meaning that some connections may be overwhelmed while others remain relatively idle. Telecom operators are now working on improved optimization software that can provide "dynamic" routing, changing paths as congestion rises and falls to ensure better use of the communications network. Airports also have an interest in such algorithms, since they can help relieve the waits of passengers in airliners queued up to take off. This is a complicated problem since safety requirements demand a delay between takeoffs, and faster aircraft can take off more rapidly; there is also the issue of how many passengers an airliner is carrying. Traditionally, controllers have solved the "departure problem" manually, but now software is being developed to tackle the job.
* Optimization is one of the big domains of algorithms in business. Another major domain is statistical analysis. UK supermarket leader Tesco, for example, has a "Clubcard" scheme that monitors purchases by 13 million members across 55,000 product lines. The analysis system was developed by a company named Dunnhumby, and uses what the engineers there call a "rolling ball" algorithm. Each product is assigned a set of "attributes", such as "easy to cook" or "value for money" or "adventurous" or "fresh", with a product such as ostrich burgers getting a high score on the "adventurous" scale. However, the scale does not merely rate the product, but also rates the customers who buy the product. The system tracks what other products customers willing to be "adventurous" are buying, even if they are bland items such as milk or margarine, and ups their rating on the "adventurous" scale. As associations between products grow weaker on one scale, they necessarily obtain a higher rating on a complementary scale, with the "ball" rolling from one scale to another. The ultimate product is a dynamic map of customer segmentation that Tesco can use to target sales pitches.
Statistical analysis is also used to monitor fraud -- for example, some insurers used fraud-detection software that checks claims against a list of items such as the day of the claim -- frauds are most common on Mondays, it seems because people like to hatch plots on weekends; the age of the object of concern in the claim -- old cars are much more often the subject of frauds than new ones; and the stridency of the claimant -- people trying to pull off frauds have a tendency to raise a fuss in hopes of snowing the insurance agent. All credit card companies have systems to monitor for fraud.
Yet another major application of algorithms are internet search strategies, the techniques by which Google, MSN, and Yahoo! attempt to obtain the best matches to queries. One of the reasons that such statistical analysis has become more important is because for the first time businesses have the means to obtain data at a level of detail undreamed-of 50 years ago, and trying to sort out the flood is a full-time job. The availability of such torrents of data has also tempted businesses to do everything possible to exploit the information, in particular to concoct sales pitches tailored for specific customers.
* While algorithms are important to businesses, they are not necessarily important to all businesses. Retailers, telecom firms, and utilities have a big need for them, while home insurers and manufacturers of high-margin, profitable products do not. It also isn't cheap or easy to implement algorithms in software; not only is there the issue of coding them reliably, but also the issue of ensuring that users don't have to be rocket scientists to interface with the system. There is ultimately the fact that algorithms are blind: they can make recommendations, but it is up to the humans in the loop to make the ultimate decision, for example on whether a customer making an insurance claim is a fraud. However, with such a flood of data to deal with humans can use all the help they can get, and algorithms are being continually improved in hopes of providing that help.
BACK_TO_TOP* ELECTRONIC DISCOVERY: The computer revolution has had all sorts of implications that few gave much thought to beforehand. One example, as discussed in an article in THE ECONOMIST ("Of Bytes & Briefs", 19 May 2007), is the legal notion of "electronic discovery." What's this about? Suppose Alice and Bob apply for a divorce, and it leads to a nasty child-custody battle. Alice's lawyer can then demand to obtain Bob's PC to show the judge that Bob is fond of downloaded pornography.
It used to be that "discovery" meant handing over a few boxes of papers, but with volumes of information being downloaded and stored on a PC, it means examining sets of files on a hard disk or backup media. The issue with "e-discovery" is that much more information is recorded now than it was in the days when it had to be printed, which makes the process of discovery much more far-reaching and intrusive. For example, if a company is sued for sex discrimination, the plaintiff's lawyers can ask to see all the email sent by company management to check for sexist remarks.
Not surprisingly, e-discovery can also be extremely expensive, since it can mean obtaining archives of email, voicemail, Powerpoint slides, and so on. The sheer cost of e-discovery gives companies an incentive to come to a settlement with lawsuits even when they are blatantly frivolous. The fact that the rules are unclear only aggravates matters. The volume of e-discovery data also bogs down the courts.
In May, the Institute for the Advancement of the American Legal System, a think-tank at the University of Denver in Colorado, issued guidelines for e-discovery, starting with the simple suggestion that the rules need to be clarified, particularly at the state level. The guidelines also suggested that judges need to be more familiar with the relevant technology, so they can determine when a request for disclosure is reasonable. In addition, the courts should know about software tools that can streamline e-discovery, and make sure the litigants know about these tools; the courts have too much to do to spend all their time sorting through a sea of bits and bytes.
BACK_TO_TOP* FLORIDA ROAD TRIP (2): After considerable logistical effort -- getting the car to the shop for a lube, waxing it shinily to help protect it from bug strikes, packing things up -- I was finally ready for my Colorado-to-Florida road trip. Since I was going to move two time zones ahead, I started getting up an hour earlier a week before to adjust my schedule for Eastern time.
I took off on the morning of Friday, 14 September. The first leg of the drive east was from Loveland, Colorado, to Columbia, Missouri. It was straightforward -- drive south on I-25, dogleg through I-205 in Denver, and then cruise east on I-70. I was on the road well before sunup, so it was still dark while I passed through the I-205 dogleg; it runs through industrial parts of the city, and though refineries are grungy and workmanlike in daylight, in the darkness they are spectacularly lit up.
I never had driven out I-70 to the east before and was thinking it would be dull and desolate. It was to a degree, but it was more somewhat discouraging. The highway was dotted with a sequence of farm towns, one much like the other, marked by grain silos and the McDonald's golden arches. In some places things were well-kept, in others the towns seemed to be dying out, marked by dilapidation and giving an impression of poverty. I was thinking cornfields would be the norm, but it turned out sorghum was more popular, and many fields were planted with sunflowers.
I was a bit surprised not to see a single serious wind turbine there -- it would seem to be a good place for them -- though I did see billboards lobbying against them. Since I like wind turbines I wasn't particularly sympathetic to the resistance, all the more so because the area seemed in need of a little economic development. Eh, don't really have a dog in that fight, they'll work it out on their own. [ED: As of 2018, they did, with farmers all over the Great Plains becoming enthusiastic for wind power, and the naysayers going quiet.]
I was also a surprised to see signs for the "Rolling Hills Zoo" as I approached Salina, which I thought was a pretty isolated place for a zoo. The signs said it was only two minutes off the highway, and since I wasn't on a real tight schedule I figured what the heck, give it a shot. I was thinking it was going to be some tacky operation, but when I got there, the zoo turned out to be a very neat, well-laid-out, and reasonably well-stocked.
I went through it rapidly and found most of its animals familiar, but most zoos have a few distinctions; I was interested to find it had an aardvark exhibit but was disappointed to see the aardvarks holed up in a box taking a snooze, showing off little more than their naked tails. I was more lucky in getting a shot of a capybara -- which, if the beast isn't familiar, is a South American rodent, something like a muskrat but about as big as a mid-sized pig. Although I didn't get many pix there, it was still a worthwhile stop just because the place was so pleasant. I think that's why the place exists, it's a pleasant refuge from the surrounding farmlands.
The pace of the drive changed at Kansas City -- the drivers seemed to get abruptly more surly. Getting through KC was a bit tricky because my plan was to stay on I-70 and I assumed that I would just keep driving straight on the main highway. However, as it turned out to stay on I-70 I had to take an exit, with the main highway changing into another -- and I had to do the same again. If I hadn't been alert, I would have got lost quickly. [ED: Much later I found out that I-70 did a loop, with a short alternate highway segment linking the ends of the loop; if I'd just ignored the exit, I would have gone straight through.]
The terrain also changed as I went east, consisting of forested rolling hills. I don't recall it as being all that distinctive; it was getting dark by that time and I was tired, just cruising on to my night stop at the Super 8 motel in Columbia. I can't recall a single thing about the stop. I just parked, checked in, and went to bed. [TO BE CONTINUED]
START | PREV | NEXT* INFRASTRUCTURE -- RAILROADS (8): While locomotives are the stars of a train, the cars or "rolling stock" have their own interest as well. There's quite a variety -- flatcars of various sorts, boxcars, hopper cars, refrigerated cars ("reefers"), tank cars, and so on. They generally share a common undercarriage -- eight wheels on four fixed axles, with four wheels on a swiveling bogie at each end of the car.
For a long time, the cars had journal bearings, with the axles rolling on lubricated sleeves. They tended to seize up on occasion and could cause fires; modern cars use race bearings instead, which are much less troublesome. Cars with journal bearings still linger, and they can be recognized because the bearing is contained in a metal box into which lubricant is poured. The cars ride on the bogies using a spring suspension. Some boxcars also have pneumatic shock absorbers.
The couplers joining US freight cars are like two hands clasping, with the fingers curled; they join automatically when the cars are shoved together, with a pin dropping down to hold them in place. The couplers are called "knuckles", though the formal name for them is the "Janney coupler", after a 19th-century engineer named Eli H. Janney, who designed them. A pneumatic hose for the brake system has to be coupled alongside by hand. A worker pulls a handle on the side of the car to release the coupler and the brake hose.
There's a bit of slack in the couplers, which is why a train getting rolling starts off with a rumble that moves down the train in a wave. The slack is actually more or less intentional. It's harder to get a vehicle rolling from a dead stop than it is to accelerate it while it's moving, and the slack it allows the cars to overcome this "static friction" one at a time. There's a similar rumble when the train slows down abruptly. European cars use a different coupling scheme, consisting of a horizontal loop on each car linked by a pin inserted by hand. European cars also have spring-loaded buffers between the cars to maintain spacing, and they don't bunch up like US cars do.
The pneumatic braking system goes back to George Westinghouse in the 19th century. The engine has a compressor system, with hoses linking each car and a pipe running under each car to carry the pressurized air; each car also has a reservoir to provide adequate pneumatic pressure to each brake, with the reservoir charged up before the train leaves the station and a set of valves under control of the primary pressure system chain controlling the discharge of the reservoirs. The system is "active off": the pressure is required to keep the brakes off, while releasing the pressure turns the brakes on. This was Westinghouse's clever idea: the brake system is "failsafe", in that if the pressure system fails the brakes go on automatically and bring the train to a halt.
The system works well and reliably, but it has a fault: for a long train, propagating the pneumatic brake signal from the front to the back can take several seconds, meaning the front of the train is braking while the rear is piling up. This can cause the train to "jacknife" off the tracks at high speeds. A new scheme retains pneumatic brakes, but controls them with electrically-operated valves that allow the brakes of all the cars to be applied at the same time, cutting stopping distance by half or more. The problem with the "electronically controlled pneumatic (ECP)" braking scheme is that all the cars in the train have to be wired for ECP, and so the approach is in very limited use at this time. [TO BE CONTINUED]
START | PREV | NEXT* MARKET ACCESS: Ethiopia is usually associated in the Western mind with famine, and indeed about 7 million Ethiopians are on food aid. However, as reported in THE ECONOMIST ("Get The Gangsters Out Of The Food Chain", 9 June 2007), Ethiopia actually is Africa's second-biggest maize producer, and is also a big wheat producer. There is no real shortage of food; the problem is getting the food to the people who need it, since Ethiopia's food distribution system is very broken, suffering from gangsters who drag down the flow of food with extortion and robbery. Even if the gangsters weren't around, the system would be very inefficient.
Now the UN World Food Program (WFP) and its allies are trying to use information technology to break the roadblocks by establishing the "Ethiopian Commodities Exchange (ECEX)". The ECEX is consciously modeled on the pioneering Chicago Board of Trade of 1848, one official associated with the project commenting: "In fact, conditions in modern Ethiopia are similar to those in 1848 Chicago."
The ECEX is expected to be operational by December 2007. The exchange will deal in wheat, maize, teff (a grass along the lines of millet), coffee, peabeans, and sesame. The exchange will provide control over warehousing and inspection of goods, and an electronic trading system will ensure that transactions are completed before the end of the day. There will be a trading pit linked to regional trading centers, which will link in turn to local traders via cellphones. WFP officials have a vested interest in the exchange, since the operation is a big buyer for Ethiopian foodstuffs. Concepts for exchanges elsewhere in Africa are starting to emerge.
BACK_TO_TOP* RFID DELAYED: Over the last few years, an enormous fuss has been raised over "radio frequency identification (RFID)" systems, with advocates claiming they would be the next technological revolution, and critics claiming they were the work of the devil. As reported in THE ECONOMIST ("Radio Silence", 7 June 2007) the revolution appears to have been postponed.
In 2003, the era of RFID seemed to be dawning when giant retailer Walmart and the US Department of Defense mandated that their biggest suppliers provide RFID-enabled product shipments. Other retailers, such as Target and Best Buy, followed suit and established similar requirements.
Then reality intruded. One of the problems was that at the time Walmart and others were demanding RFID from suppliers, there was no comprehensive standard in place for the technology. Nobody could figure out how to implement product RFID that could be uniformly read by RFID scanners, which were based on competing high frequency (HF) and ultrahigh frequency (UHF) technology. A "GEN2" standard specifying UHF RFID technology was finally established in late 2004 -- to be followed by an extended effort to implement and test RFID systems for compatibility to the spec. It turns out that implementing a fully interoperable global RFID spec and coming up with an RFID system that works cleanly is a nasty challenge that hasn't been completely fixed yet.
There's also a question of the business case for RFID. Big operations like Walmart and the Defense Department saw it as a win from the beginning, but the vendors supplying product to such organizations weren't so enthusiastic. An RFID tag only costs about a dime these days, but vendors were already being squeezed to thin profits, and a dime could cut badly into that margin.
Some retailers and vendors of high-value products -- designer clothes, prescription drugs, consumer electronics -- can afford to add RFID and have been enthusiastic about it, since it reduces theft and piracy. Highway toll systems are also now widely RFID-based. Big companies find they can make a good business case for RFID in internal or "closed-loop" processes. RFID enthusiasts such as Walmart are also realizing that simply dictating RFID to vendors wasn't workable and are approaching the matter as a partnership issue. It is clear that RFID is still a coming thing, but given the painful teething problems over the last few years, nobody's quite as willing to predict when it will really take off.
BACK_TO_TOP* RIDE THE DREAMLINER: Boeing Corporation is riding high on the company's new 787 Dreamliner, and according to an article in BUSINESS WEEK ("Better Living At 30,000 Feet" by Stanley Holmes), it's not just the airlines who have reason to look forward to the new, cost-efficient jetliner: passengers will also find the 787 a much more comfortable ride.
The 787 gets some of its "user-friendliness" simply from the fact that it is mostly built from composites. The composite fuselage can withstand greater pressures than an aluminum fuselage, so the cabin pressurization is set at 1,830 meters (6,000 feet), not the usual 2,440 meters (8,000 feet) -- a plus for passengers who are sensitive to thin air. Composites don't rust, either, and so the humidity can be set higher than in a traditional jetliner.
Other features include a flight-control system that smooths out bumps; windows 65% bigger than traditional jetliner windows; a new air filtration system to get rid of odors; blue LED lighting to give either day or night lighting conditions; and even arches to separate the aircraft sections, as well as a fake skylight to give an impression of spaciousness.
Boeing engineers spent a good deal of time with customer focus groups to assess passenger likes and dislikes, then made tradeoffs accordingly. Windows on aluminum-bodied aircraft tend to be small because the window framing adds weight; it still does on a composite aircraft, but the penalty is smaller. The concept of increasing cabin pressure was a bit of a surprise, since there wasn't much data on how passengers handle low pressures; Boeing ended up sponsoring an investigation at Oklahoma State University that showed increasing the pressure reduced the number of passengers who complained from 12% to effectively zero. (ED: 1,830 meters is only slightly higher than the normal altitude here on the Colorado "mile high" -- 1,600-meter -- plains.)
Some critics comment that while all the new whizzy features are very nice, the airlines will continue to load in coach passengers using high-density seating, and ergonomic changes can't do very much about being packed into a sardine can. The more things change, the more they stay the same.
BACK_TO_TOP* FLORIDA ROAD TRIP (1): After my road trip to South California in May, I figured I had enough of traveling for a while, but over the summer I started to reconsider that notion. I had long been wanting to go to Florida, the goal mainly being to get pix at the US Naval Aviation Museum in Pensacola, and I knew it was necessarily going to be an extended trip because of the distance and the bigtime attractions in what I like to call the "Central Florida Tourist Complex", centered on Disney World (DW) and Orlando. I'd been there in early 1998, and things had been updated and improved considerably since then. As long as I was going to drive that far, I'd have to take in the tourist sights as well.
I was considering all the things I had on my worklist for 2007 and 2008, and realized that in fall 2007 I had something of an open space in my schedule. Why not take the trip now and get it off my plate so I could focus on other projects in 2008? Money had been accumulating mostly undisturbed in my travel slush fund; fuel prices were relatively low; if I went in the off-tourist season, that would mean short lines at DW and the like, allowing me to see more in the time available -- and helping me stay off my feet, which seem to be chronically sore these days. I went and bought a guide book and a map of the central Florida area to get some ideas of what my agenda might be.
I was thinking of going in early November 2007, but then I found out there was going to be an event in the Orlando area at that time. That led me to wondering just how I could make sure I visited when nothing in particular was going on. Googling on "Central Florida Events" didn't give me anything very useful -- but then I had a brainstorm: high-end hotels often vary rates with expected traffic, so if I could find the week in the fall when rates were minimal I could expect to miss the crowds. I checked a number of hotels I and it was clear that mid-September was the best time was to hit the road. I also checked around and found out another, less welcome indication of the slow season: I didn't think of it, but to no surprise in hindsight, the theme parks tend to run shorter hours. Duh.
* Since it was mid-August when I figured this out, that meant I had to get planning immediately. Four weeks was not too long to be piecing together such a complicated trip -- eleven days total, with planned visits to six museums, four theme parks, and four zoos (if you want to call a dinosaur park a "zoo"). I had to start thinking about packing up immediately.
Staying at one of the hotels in DW was out of the question. Disney believes in giving customers their money's worth, but also ensures that the customers pay up for it appropriately. Not only are the hotels in DW relatively expensive, but guests are captive to the high-priced DW restaurants -- if I were made of money I'd give it a shot, but I'm not. It is possible to get a fairly good deal through, say, a full-week package from them, but I wasn't going to be there that long. Besides, I find fancy hotels irritating, not just because I'm cheap but because I hate being waited on, and really fancy hotels can come across more as stuffy than luxurious. As long as the hotel room's clean and the water's hot, I've pretty much got all I need.
International Drive in Orlando offered cheap hotels and convenient access to restaurants for any taste and budget. I finally settled on the Best Western Plaza motel -- I prefer motels, and particularly like getting a ground-floor room since it makes loading and unloading the car so simple. It's centrally located in the International Drive area.
* All that decided, I got online and made hotel reservations, bought tickets, and so on. I think we're still a long ways from having a good handle on internet commerce -- trying to wade through the overstuffed Disney website was painful and the Super 8 motels site seemed to have real problems with trying to do two tasks in separate windows -- "OK, I guess I don't do that." I particularly wondered about the Kennedy Space Center (KSC) Visitor's Center website, which charges three bucks for "electronic delivery" of advance tickets -- which means that it's three bucks more than at the door. Hmm, sounds like a government operation at work.
Incidentally, a two-day multi-park / auxiliary attractions ticket to DW cost over $250 USD! Professional as always, Disney mailed me a nice little plastic magstripe card that had all the details printed on it -- though also typically they charged me a few bucks extra to standard-mail it to me. Like I said, Disney is very conscientious about delivering what the customer pays for, but Disney managers are apparently penalized if they offer anything that looks like a deal, even a little one. I got an unlimited access ticket for both Busch Gardens in Tampa and Busch Sea World in Orlando for $90 USD -- Busch, I suppose because they aren't in Disney's league, doesn't have problems offering their customers a pretty good deal. A one-day ticket to the twin Universal Studios parks in Orlando was about $80 USD.
I ended up printing out a pile of maps and directions for the trip. Planning always seems to pay off, since even fairly minor bungles on long-range road trips can do a lot to shift the spirit of the exercise from "fun" to "nightmare". Since my brain tends to become increasingly nonfunctional the longer I'm on the road, it's important to make sure that as much is thought out in advance as possible. I tend to print out maps until I finally think: "OK, I know where I'm going now." Alas, in practice that doesn't always turn out to be the case -- yet further evidence, if any were needed, of the difference between theory and practice. I end up doing as much as I can with the maps and then admitting that I'll have to deal with the rest as it happens. [TO BE CONTINUED]
NEXT* INFRASTRUCTURE -- RAILROADS (7): To those who like railroads, the rail system infrastructure is interesting, but it's the trains themselves that are the main attraction.
Steam trains have the most appeal to railroad buffs, but they are long obsolete, being inefficient and hard to maintain. The first diesel-electric locomotives, which used a diesel engine to drive an electric generator that in turn drove the wheels, were introduced in the 1930s, primarily as "switching engines", shunting cars around in a switchyard. The old steam locomotives had the cab behind the engine, for the simple reason that the locomotives had to haul a "coal tender" behind them to supply fuel. The early switchers unimaginatively retained the rear-cab configuration despite its poor view forward, even though they carried their diesel fuel in a tank under the engine. The rear-cab configuration was less an issue in a switchyard, since a switching engine shuttles back and forth all the time anyway.
The first broadly successful diesel-electric locomotive was the "F-Series", introduced by the General Motors Electromotive Division (EMD) in 1945. It was a relatively sleek machine, with the cab in front, and about 3,000 were built. The F-Series was a stereotype of trains in 1950s media, and still is commonly seen modeled in toy train sets. However, the typical modern diesel-electric locomotive is a much less stylish beast, essentially derived from switching engines and with the cab placed up front. They're very boxy, with few apparent considerations of aerodynamics -- it seems that at typical railroad cruise speeds, the sheer mass of a train makes aerodynamic drag a fairly trivial consideration. Some of the latest locomotives do make some concessions to aerodynamics.
All diesel-electric locomotives work on similar principles. They have a big low-RPM diesel engine with 12, 16, or even 20 cylinders -- the cylinders are huge, I once saw a picture of an engine block with a little boy playing inside a cylinder bore -- that drives a DC generator. The DC generator drives electric motors to turn the wheels. Why the roundabout approach? Because it's a lot easier than trying to build a gearbox that can handle such power.
Switchyards can often be observed from a traffic overpass. From the top, a diesel-electric locomotive has an engine exhaust stack and a set of cooling fans. Some also have a set of fans to help deal with "regenerative braking", in which the electric motors are operated backwards as generators to help brake the train. At present, this is done just to save wear on brakes, with the electricity produced burned up in big wire-wound resistors on the roof of the locomotive, with the fans carrying off the heat.
The latest diesel-electric locomotives feature advanced diesel engines and microprocessor control systems, permitting greater efficiency and lower emissions. Work is underway on improved "hybrid" locomotives -- though the term is a bit odd since it is broadly applicable to all diesel-electric locomotives. The trick is that they actually try to recover the energy lost through regenerative braking, instead of just throwing it away. Given the mass of a train, this is a lot of energy to waste, but it's tricky to figure out how to store it all: it would make no sense to require a bank of batteries as big as the locomotive itself. Some hybrid switching engines are now on the market -- they're relatively small and don't move very fast, so the energy storage requirements are not so severe. [TO BE CONTINUED]
START | PREV | NEXT* CELLPHONE EMPOWERMENT: People in developed societies tend to think of cellphones as tools for chat, forgetting how important communications are for just getting things done. An article in THE ECONOMIST ("To Do With The Price Of Fish", 12 May 2007), reported on a paper written by Robert Jensen, a development economist at Harvard University, that showed how powerful a tool the cellphone can be for the economically underprivileged.
The paper focused on fishermen in the southern Indian coastal state of Kerala. Traditionally, if a fisherman hauled in a good catch of sardines, he could assume that neighboring fishermen would haul in a good catch as well. That meant a glut at the local market, not only depressing the price the fisherman could get for the catch but even preventing him from selling all of it, and then he'd have to throw it away when it started to stink. For all the fisherman knew, a market a few kilometers down the shoreline might be starved for fish -- but he had no way of finding out, and simply puttering around to check would be time-consuming and ineffective.
Jensen's paper focused on the operation of 15 fish markets along the Kerala coast. He found that in 1997, from 5% to 8% of the catch was discarded at saturated markets, even though there were buyers available at markets not far away. However, in 1997 cellphone systems were introduced in the region, and within a few years almost all the fishermen had them, which they used to call different markets to see which were saturated and which were buying. Now the wastage of fish is insignificant, prices are uniform across the region, fishermen have 8% higher profits, and consumers pay 4% less for fish. The higher profits mean that the cellphones quickly pay for themselves.
Jensen's paper is significant because it focuses on "microeconomics", while other studies of the impact of cellphone in developing nations focus on "macroeconomics". One such macroeconomic study showed that when 10 among 100 people in a developing country obtain cellphones, GDP rises by an increment of about 0.5%. However, the problem was trying to trace how the cellphones did this magic -- obviously they helped by reducing the need for travel, widening markets, and permitting exchange of price data, but the specifics were unclear. Jensen's research helped fit in the pieces.
Jensen's paper also pointed out that the cellphones allowed the citizens to improve their lot without government intervention. Cellphone system operators build and operate their networks at a profit, and the citizens find that it is profitable to make use of those networks. All the government needs to do is provide a rational and benign regulatory environment for the cellphone operators, and the people take care of the rest. This is by no means claiming that government intervention is necessarily a bad thing -- but it is still easier for all concerned if people can better their lives on their own,
* In related news, SCIENTIFIC AMERICAN reported on another aspect of cellphone use in undeveloped countries, something that folks in developed countries are likely to forget: a lot of people in undeveloped countries are illiterate and can't read menus or the like. Cellphone giant Motorola conducted field research on the problem and has now introduced the low-cost "Motofone", which is being sold in South Asia. The Motofone has a reduced number of keys and no wording. It has a numeric keypad and individual buttons, labeled with icons, for call, end, menu, and phone book. The menu button brings up a display of icons, with the user cycling through them using a dial. Trying to figure out the icons took some work -- an alarm clock for the alarm function worked fine, a stack of coins to give the amount of prepaid call time left was often mistaken for a pile of food.
Motorola's rival Nokia conducted similar studies, but found that illiterate newbie users were usually helped by friends or relatives to get up to speed. Nokia simply added some audio menus. Some researchers think that as phones get smarter and more capable, they might be able to actually teach users how to read, logging mistakes and changing the "coursework" to correct for them.
BACK_TO_TOP* DNA BY MAIL: The development of 21st-century genetic analysis technology now allows anyone to analyze their genetic makeup, and according to an article in US NEWS & WORLD REPORT ("Unraveling Your DNA's Secrets" by Nancy Shute, 8 January 2007), many people are now making use of DNA testing services. Such companies ship out a little brush with a prepaid return mailer; the customer rubs the brush on the inside of the cheek for 30 seconds and then mails the brush back. A report is sent back a few weeks later.
Tests are now offered for a range of possible diseases, but critics are concerned that the DNA-by-mail companies are often on shaky scientific ground. Says Francis Collins, head of the National Human Genome Research Institute: "Many of the claims that are being made are quite fanciful. But the fact that many of these tests have not yet reached the point of rigorous scientific validation has not slowed down the interests of consumers and of entrepreneurs."
More traditionally, the usual subjects of genetic testing are prospective parents who want to know if they are carriers of genes for cystic fibrosis, sickle-cell anemia, or Tay-Sachs disease. The tests and their practical implications have relatively few ambiguities. The DNA-by-mail companies, in contrast, test for a much wider range of conditions. The DNA companies are flatly not under scrutiny of the US Food & Drug Administration (FDA), and the Federal Trade Commission has recommended that consumers be skeptical of claims made by these companies. US NEWS did an investigation of the DNA companies and got back a very mixed bag of results:
We've come a long way with DNA analysis, and we are still only scratching the surface of its possibilities. However, by that same card we're not yet at the point where we can use DNA testing as a very reliable crystal ball to tell our futures.
BACK_TO_TOP* IMMOBOTS EMERGING: In the modern era of automation everybody finds the notion of a robot as a mechanical man a bit quaint, but as reported some time back in MIT TECHNOLOGY REVIEW ("Immobots Take Control" by Wade Roush, December 2002 / January 2003), it's still a bit surprising at the forms a robot might take.
In 1999, the US National Aeronautics & Space Administration (NASA) Mars Polar Lander ran into a glitch as it tried to set down on Mars. As best anyone knows now from Earthbound tests, it got a fake signal when it extended its landing legs that it had touched down, and so the lander's control system shut down its landing rocket. The probe fell to the Martian surface and was destroyed. The control system didn't have the brains to figure out from other inputs, such as a radar altimeter, that the spacecraft was still high in the sky.
In contrast, that same year the NASA Deep Space One (DS1) space probe had a glitch of its own, with a critical maneuver disrupted by a stuck power switch. The results could have been completely wrecked an asteroid flyby, but DS1 had an autonomous control system that recognized something was wrong and came up with an alternate plan to save the flyby. Actually, the "crisis" was a drill, the stuck switch being simulated; the probe passed with flying colors.
DS1's autonomous control software had a detailed picture of the probe's subsystems and their interactions, with the ability to assess the state of the subsystems and figure out how to keep the system working if one failed. Such systems are increasingly common back here on Earth; since they are generally associated with machines that don't move around, they are often called "immobile robots" or "immobots" -- though the term is a bit inexact for DS1's system, since a space probe moves around quite a bit.
A traditional computer control system executes a preplanned sequence of events. An immobot is different: it has a model of the system that it is trying to control and can "reason" to respond to failures, even sometimes failures that the designers of the system didn't foresee. Immobots are now found in cars and office copiers, and should be encountered soon in heating and ventilation system, computer networks, electrical grids -- or in general any complicated system that can be controlled by a computer.
The problem with traditional software is that every contingency has to be considered in advance and coded for. That means a lot of work and cost to get things right. Advocates claim that the "model-based reasoning" of immobots does a better job, achieving excellent reliability at low cost. In the DS1 control system, each subsystem was defined as a "model" that defined its possible states: a valve, for example, could be OPEN or CLOSED, or in failure modes STUCK OPEN or STUCK CLOSED. Possible transitions between states and their probabilities were determined -- for example, it was unlikely, though not impossible, that the valve would go from STUCK CLOSED to STUCK OPEN. The control system was then provided with a "blueprint" of the subsystems that showed their interactions and relationships. In operation, the system monitored the subsystems and then used its reasoning engine to determine the workaround when a subsystem failed.
Xerox's high-end copying machines, which are truck-sized and used for heavy-duty publishing jobs, now incorporate immobot systems to ensure proper scheduling of printing jobs and working around problems. The copiers operate much more efficiently than their predecessors, with the control systems suffering little confusion when the copier configuration changes with an upgrade.
Modern automobiles have complicated electronic systems, sometimes with dozens of processors, and Toyota researchers have conducted work on an immobot system that can assess the inputs from automotive systems to provide intelligent diagnostics for drivers and mechanics. No production cars have immobot systems yet, but advocates believe that the car will be the first to introduce immobot systems to a mass market.
BACK_TO_TOP* JACK CHICK LIVES! When I was stationed with the Army down in Texas in the mid-1970s, every now and then I would find itty-bitty comic books scattered around the barracks. Some of them were fundamentalist Christian tracts, strong on preaching if short on artistic talent. I never saw them again after I left Texas and more or less forgot about them.
My recent work on evolutionary science has led me down some unexpected paths, and when I ran across a blog posting about somebody named Jack T. Chick who made fundamentalist Christian comics, that rang a big bell from my Army days. I did some poking around and found out that the little comics I had read were "Chick tracts", all the work of Jack Chick. These days he just writes the scripts, he has better artists do the artwork. The Chick tracts are fundamentalist Christian parables, heavy on the "fire brimstone & damnation", with warnings about witchcraft, demonic possession, and of course, ahem, "Darwinism". They portray a sort of universe along the lines of that of, say, BUFFY THE VAMPIRE SLAYER, but taken seriously and then some. However, on the principle that I can respect the people around me better if I don't concern myself too much with the things they believe, I will say no more about Chick's religious doctrines as such.
Chick does, however, have some notions that spill into the realm of conspiracy theories that make maintaining a benign blind eye and a straight face a bit more difficult. In particular, he claims that the Catholic Church is a satanic plot, going so far as to assert that both Islam and Mormonism were created by the Vatican as a means of furthering the church's evil plan, and the church also keeps a computer database of all Protestants so they can be nailed the day the counter-Reformation comes. Apparently, Chick got some of this silliness from a Spanish crony named Alberto Rivera, who claimed he had been a Jesuit priest working as a "secret agent" to subvert Protestant churches, and had even been covertly promoted to a bishop.
An anti-Catholic activist who had once been a priest talked with Rivera to find that he didn't know details of ordinary Catholic practice or doctrines, and concluded there was no way Rivera had ever been trained as a priest, much less become a bishop. One reviewer on Amazon.com of a book Chick had written said that he had got religion because of Chick -- the reviewer found the claims the book made so hard to believe that he had to go check up on them, and ended up becoming a Catholic. There is indeed a Zen to the Universe.
Not much is known about Chick other than he was born in 1924, was a sailor during World War II, and got religion after the war. He keeps a low profile and there are some who suggest the name is a pseudonym for an individual or a group, though legal records show he exists. It is said that about a half billion Chick tracts have been printed in all. [ED: He finally passed away in 2016.]
* I have to admit that I get some amusement out of the lunatic fringe. Incidentally, scouting the web led me to the website OBJECTIVEMINISTRIES.com that seems at first like a extremist fundamentalist Christian website, but on closer examination turns out to be a very sly parody, so dry that it takes some time to figure out that it's a gag. The site features a biographical page for the staff that looks straightforward, but the listing for the organization's accountant states he will give a 5% discount to Christian clients -- and a 10% discount to nonbelievers who convert on the spot. The website even supports legitimate Christian banner ads, lending it a further air of authenticity -- as well as providing a bit of revenue. It is so deadpan that it has been linked to by fundamentalist Christian websites and, of course, denounced by atheist websites.
Anyway, as far as the Chick tracts go, I'm still a bit puzzled as to how the tracts ended up being scattered around the barracks. I can only speculate that it was fishing by missionaries. What is more puzzling that I'd also find pornographic comics in the same format -- they called them "eight-pagers". They were junk even by the standards of porn, so I don't know who would want to buy them. Since I can't recall they carried any advertising -- not like the little fliers a colleague of mine once brought back from the Shinjuku red-light district in Tokyo, with insanely oversexed manga cartoon babes advertising vice palaces -- I can't think of a reason someone would want to just give them away.
The eight-pagers were generally parodies of comic strips, for example DENNIS THE MENACE: "They said I was too little to do this!" No, I won't elaborate. I don't run that kind of a website.
BACK_TO_TOP