Wednesday, August 08, 2007

Bridge Collapses and Failed States

Stephen Wolfram looks at the Minnesota bridge collapse -- and bridge failures in general -- in the abstract. Wolfram makes the point that certain mathematically optimal structures don't look simple at all. The strongest bridges, he argues, may not look it all.

But it's been known for a while that the best networks (shown at the bottom) don't have that kind of simple structure. In fact, they almost seem in some ways quite random. Well, what about bridges? I strongly suspect that there are much better truss structures for bridges than the classic ones from the 1800s--but they won't look so simple. ...

So what should the bridges of the future look like? Probably a lot less regular than today. Because I suspect the most robust structures will end up being ones with quite a lot of apparent randomness.

Those new kinds of bridges being built may be a bit shocking at first. After all, classic regular bridge structures--and things like the Eiffel Tower--are icons of our modern engineering-based civilization. And in fact, even biology--with its iterative process of natural selection--probably can't find structures as good--and irregular--as the ones I expect are out there.



It may be that history will show that one of the reasons the US did so poorly early in Iraq was because it labored under a mental structure with rigidities that facilitated failure. That contrary to conventional wisdom the fundamental problem with the Administration was that it had a plan -- but the plan was the wrong one. While I am not arguing against the utility of timelines and exit strategies in general, I have always wondered how applicable they were in complex situations. Anybody who expects events to develop according to a scheduled timeline and plan will be very lucky indeed. Every time I hear a politician citing some newspaper clipping to prove "the failure" to achieve a predetermined outcome I sometimes think that the citation itself is proof of intellectual failure of a different but more basic kind, one of which the politician is blissfully unaware.

What is really useful in complex situations is a loose kind of algorithm that will enable the decision maker to understand what is important, within a given period, and allow him to choose paths to an improved situation. Stephen Wolfram is of course, investigating physical and mathematical structures and I do not mean to translate his arguments directly into the messier human world. But being able to discern see the real information structure beneath the apparent randomness strikes me as a key skill. Otherwise we see, but don't really see at all; we plan but only to achieve our prejudices.

38 Comments:

Blogger ForNow said...

A structure with unexpected strengths could also have unexpected weaknesses. If it's complex and seemingly random, wouldn't it be harder to understand? -- which could mean costs in some sense somewhere along the line. I guess that at the very least it means testing the dickens out of it.

8/08/2007 07:25:00 PM  
Blogger ForNow said...

Note: I was talking just about bridges.

8/08/2007 07:30:00 PM  
Blogger Doug said...

Agree with fornow:
Simpler and more Robust (Brooklyn Bridge at over 100 years still carries ever greater loads) means more redundancy, and less chance that a flaw will go undetected.
(Latest word is that may have been the case here.)

Potential Flaw Found in Design of Fallen Bridge

Biggest fault with this Bridge was a design where a single failure brought down all three spans.

In just the picture below, we see one of the new Jumbo-Sized Concrete Trucks, what appears to be a load of Asphalt or gravel, a Concrete Pumping Truck, a pile of sacks of cement, pile of gravel, freshly poured concrete, and etc!
Well DUH!

Turns out there were 100 tons of gravel piled on the span, plus concrete trucks and associated, add some passing fully loaded Semis, all on a corroded, fatigued bridge carrying Nafta traffic beyond the designers dreams, and the recipe for disaster was complete.

Meanwhile, the bullet-proof "overdesigned" Brooklyn Bridge carries on.

8/08/2007 08:18:00 PM  
Blogger Doug said...

"If it's complex and seemingly random, wouldn't it be harder to understand?"
---
Also harder to inspect and maintain:
From the pictures it's obvious this bridge did not receive the continual application of corrosion resistant "paint" that is bestowed on the Golden Gate Bridge, year after year.
...and the main load bearing elements are not spread over and hidden throughout the span (with more area/areas exposed to the elements) as they were in Minnesota.

8/08/2007 08:26:00 PM  
Blogger Whiskey said...

Wolfram is the perfect example of the educated idiot. Dashiell Hammett covered that species in "This King Business" where the Continental Op has to help unseat a distinguished microbiologist who being educated and expert in his field was a disaster in a Balkan nation's Presidency.

Wolfram is extrapolating mathematical algorithms for sorting to the very mundane task of bridges: transfer loads. It's fairly straightforward. The classic way to accomplish that is the arch. Followed by the suspension bridge. Followed by the truss bridge. Many arched bridges from the 1200's are still in use today in Italy. Properly constructed they are immensely strong.

The Minneapolis bridge collapse happened due to building the bridge fast and cheap. Instead of picking two of three: fast, cheap, good.

There is no "randomness" to the physical (as opposed to the abstract world). Structures designed to transfer load WILL conform to certain physical constraints.

I *DO* understand what you are getting at W -- but I suggest that tried-and-true methods of systems innovation as applied to humans rather than a loose algorithm or even seeing underlying patterns. What made Bill Walsh great as the 49'ers coach was not the West Coast Offense which had it's roots in Air Coryell, but his ability to see the pattern (coaches/players relying on "toughness" and simple gameplans) AND the ability to get others to sign on to his plan to exploit that weakness. Not merely seeing, but adapting. Greg Boyington and Claire Chennault in China with the Flying Tigers, heck Col. Boyd and his OODA loop, back to Machiavelli, Sun Tzu, Musashi, and Clausewitz have all spoken to this. To see and lead others in unison to act.

The old recipe of greatness in human leadership. The most magic ingredient of them all.

8/08/2007 08:36:00 PM  
Blogger El Baboso said...

Networks evolve and are dynamic and bridges are designed and are static.

I think that it is almost axiomatic that the more hierarchical (planned) a network is, the more slowly it will evolve and adapt.

A well-planned operation against a nascent network can be very effective in disrupting its operations before it can begin exchanging intelligence, tactics, and other knowledge (let alone building its unifying narrative).

A well-established network can often be very effective against a planning, hierarchical organization since it can gather and disseminate intelligence faster, adapt its tactics more quickly, shift leadership responsibilities more quickly and has a unifying narrative that can to some extent synchronize and mass fighters against the planning organization.

Both the Gestapo and NKVD, following behind their respective armies, always had detailed lists of people to detain, deport and shoot. Why? To prevent opposition networks from forming. Putting aside the romance of the "resistance," the Germans were very successful outside of Russia and Yugoslavia (they were culminating by that point, anyway) and the Soviets were universally successful (yes, in the short term... in the long term we are all dead as they say).

I asked before how the Europeans were able to destroy religion and nobody took the bait. The answer of course to to create a hierarchy, or even better a bureaucracy. State subsidies and clerics who are little more than paid bureaucrats sucked the life out of European religion. They replaced a network of the faithful with a soulless organ of the state.

The good news is that the Iranians have already fallen into the trap. The Wahabbis aren't far behind with the muttawa, clergy in state pay and growing hierarchy.

So what are the selective pressures that we can bring to bear to increase the bureaucratization of Islam?

8/08/2007 08:43:00 PM  
Blogger Doug said...

Didn't hurt to have the greatest "Pattern Discerner" of all time at QB, either!

8/08/2007 08:44:00 PM  
Blogger Doug said...

Inspections, reports raise questions about bridge's safety

"Replacement of this bridge, and the associated very high cost, may be deferred," the engineers wrote, according to a copy of the report on the Web site of the Minnesota Department of Transportation.

Still, engineers said the fatigue cracking was a serious issue due to the lack of redundancy in the main truss system. Only two planes supported eight lanes of traffic, they wrote.

"The truss is determinate and the joints are theoretically pinned," the report states.

"Therefore, if one member were severed by a fatigue crack, the plane of the main truss would, theoretically, collapse."

8/08/2007 08:50:00 PM  
Blogger Doug said...

MNDOT was warned that the volume of NAFTA trucks threatened the bridge.

The MNDOT document "Bridge Facts" points out Minneapolis span, despite the deficiencies noted in years of official reports, was not under any truck restrictions.
At the time of collapse, it was considered safe, even for overweight truck loads of up to 136,000 pounds.

Nightmare scenario on NAFTA superhighway

8/08/2007 09:01:00 PM  
Blogger Whiskey said...

Doug --

More of the black magic of leadership. Despite his record at Notre Dame, Montana was undrafted because of a perceived lack of athleticism, height, strength, and ability to throw 80 plus yards.

No one believed in Walsh when he brought in Montana. A year later they went to the Superbowl. With additions to defense.

That's another part of leadership, talent recognition.

8/08/2007 09:38:00 PM  
Blogger Deuce ☂ said...

I beg everyone to watch this clip at Observanda. . You will hate yourself if you do not. And you will deserve to be hated for it.

8/08/2007 10:09:00 PM  
Blogger Unknown said...

Probably the classic example of wretchard's idea would be the *beaver dam*. They are built with whatever local materials are at hand, with plenty of room for individual quirks in design and construction. They are sturdy, flexible, and easy to maintain and modify; and some have been known to last a lifetime or more!

8/09/2007 12:30:00 AM  
Blogger Unknown said...

Structure, form, dynamic, static. You could go on, and on.

The world trade center buildings came down, stone bridges have come down, steel and concrete structures have come down. Lots of stuff goes up; then at some point, comes down.

Life begins, then as we know it, it ends.

Reasons abound. Change exists. Introducing change creates something new; sometimes good, sometimes bad.

I saw a guy last night on Fox news representing CAIR, looking much like a Saudi; sporting the royal family's special mini triangle goatee (just below the lip).

He was talking like a Harvard Law graduate; spending a great deal of effort on the notion of "reasonable" to put forth his point that American businesses must provide prayer time for muslims at dusk.

When I see the tail wag the dog, I know that nature has begun to play tricks. At some point, that dog will fall down; just like the bridge, and those twin towers.

I may not know what, exactly, will bring these things down, but I do know one can not violate the laws of nature...for long.

The strong survive, the weak perish. The fun is in deciding the meaning of "strong" or "weak" and for that I prefer not to go to a lawyer. Just a special preference of mine.

I can't wait for the lawyers to begin their personal education of the American public on this bridge.

I've already seen what they have tried on the war on terror, the twin towers, and especially on the notion of what "is" is!

Where's my faith?

8/09/2007 05:10:00 AM  
Blogger Robert said...

"There is no "randomness" to the physical (as opposed to the abstract world)."

The writer of this sentence knows nothing of the physical world. He should read A New Kind of Science and absorb the manifold real-world expressions of cellular automata. Order is far more subtle than it appears.

8/09/2007 05:11:00 AM  
Blogger Doug said...

Whiskey:
"Was ready to quit on the spot in the 1977 Purdue game when head coach Dan Devine sent quarterback Rusty Lisch in to replace injured senior Gary Forystek. With the Irish still trailing 24-14 late in the 3rd, Devine replaced Lisch with Montana, the third-string quarterback. He engineered a 31-24 victory, and remained the starting QB for the rest of his Notre Dame career.

Earned the nickname "Joe Cool" because of his ability to remain calm in the biggest of game situations. The best example of this was in Super Bowl XXIII against the Cincinnati Bengals. With the 49ers on their own 8 yeard-line, down by three points with 3:20 left to go in one of the biggest games of his career, Montana pointed toward the stands and said to tackle Harris Barton,

"Hey look over there. Isn't that John Candy?"

He then marched the 49ers on a 92-yard drive for the game-winning touchdown.
"

;-)

8/09/2007 05:55:00 AM  
Blogger Doug said...

(Although John Candy would stand out a bit in the stands, I guess.)

8/09/2007 05:57:00 AM  
Blogger LarryD said...

... That contrary to conventional wisdom the fundamental problem with the Administration was that it had a plan -- but the plan was the wrong one. While I am not arguing against the utility of timelines and exit strategies in general, I have always wondered how applicable they were in complex situations.

Murphy's Laws of War

Planning

1. The important things are simple.
2. The simple things are very hard.
3. No plan survives the first contact intact.
4. Prefect plans aren't.

8/09/2007 06:11:00 AM  
Blogger Doug said...

(another)
OT:
The Eternal City of Wafa Sultan

8/09/2007 07:05:00 AM  
Blogger cathyf said...

Doug, those statistics about the heavy construction equipment and the gravel pile are only significant if you ignore the normal level of load on the bridge.

The gravel pile sounds impressive -- but only if you don't consider the fact that one-half of the surface of the bridge had been scrapped off and ferryed off to the dump.

The heavy construction trucks sound impressive -- but only if you compare to an empty bridge, as opposed to the no-construction norm. The bridge is four lanes in each direction, two lanes on each side were closed off. So the weight of the jumbo-sized cement trucks and pumping truck sounds like a lot -- but it's not a lot compared to four more lanes of bumper-to-bumper traffic.

This was the MSP main artery, and tens of thousands of times over the last 40 years it has had 8 lanes worth of traffic -- including mega trucks loaded with heavy cargo -- crawling bumper-to-bumper over it.

The engineers who built the European cathedrals followed a simple trial-and-error method. If the building fell down, they used more structure when they rebuilt. If it didn't fall down, then it had enough structure. The problem with this method is, of course, how to detect when you have just barely enough structure.

I'm far more willing to believe that components failed gradually over time, until there was some "straw that broke the camel's back" moment and the whole thing came down. It may have even had something to do with the construction -- for example, the concrete on the bridge deck may have been bearing just enough load to stave off failure of some other piece, which was failing because other things had already failed and left too much load on the remaining non-failed components. Then they came and dug up some key piece of concrete and the whole thing collapsed.

(Have you ever been in a run-down house where as you walk around the floors feel "spongey"? I always refer to this jokingly as "the wall-to-wall carpeting is structural in that house." All joking aside, when an entire system has been brought right to the edge of collapse by a thousand different causes, the thing that sends it over into catastrophic failure can be as trivial as a gust of wind.)

8/09/2007 08:00:00 AM  
Blogger Yashmak said...

"Wolfram is the perfect example of the educated idiot."

Whiskey's comment was dead-on. The man may understand networks, but his belief that this gives him some insight into better bridge construction is similar to a seamstress believeing her knowledge of stitching gives her special insight into ramjet construction. They're totally unrelated fields.

Since his whole article is based on the premise that they're not unrelated, I tend to discount most of the rest of his observations as well.

8/09/2007 08:15:00 AM  
Blogger Doug said...

You're Right Cathy F,

I figured it was a few passing 136,000lb trucks, (which did not exist at design time) on top of the construction, but it boils down to what you say in the end, compounded by traffic beyond design specs, corrosion, possible design flaw in the gussets, and the fatal flaw in Basic Design whereby the whole structure fails due to a single failure.
---
Solution:
Don't try to design to a minimum plus a little fudge factor, OVERDESIGN it! (like "primitive" Bridges, still standing)
("In Praise of Waste" -Like Nature. - Hardin)

...and design something that will withstand the elements at least as well as much older bridges.

8/09/2007 08:20:00 AM  
Blogger Kevin said...

Wolfram clearly does not understand which factors are important in determining the design of bridges. While structural loading is certainly the most important of many criteria but it is not necessarily explicitly expressed in the final design of a bridge. With more and more architects getting involved with bridge design, image and aesthetic qualities are the driving factors in the appearance of a bridge and the only way that randomness and complexity will enter into bridge design is if the architect wants them to. But whatever the desired image, the structural, economic, and construction time criteria (among many others) must all be met for a successful project. Norman Foster’s fiasco over the Thames with the Millennium Bridge is a prime example of what happens when the architect’s desire for image trumps the critical criteria of structural stability, cost, and schedule.

Now if the various criteria (especially cost) are getting out of control there is always a temptation for the architect – who wants to build his dream no matter what -- to lie (or downplay) these defeatist factors so that the client will at least start the process of construction and then once underway the client will be less likely to pull the plug when the real costs and time schedule become known. A wise client takes steps to avoid being placed in this situation by having independent consultants verifying the costs and designs presented by the architect.

Which of course brings us to Iraq. Just like in building a bridge, it was not the actual military criteria (how to defeat an insurgency) that determined the design of the operation but however there was most definitely a plan for Iraq. The desired image for the project was of a decisive, dick-swinging American President who was going to create the required tableau rasa of all utopian dreams by sweeping away the house of cards that was Saddam’s Iraq and from its ashes to construct s a monument to globalisation, free trade, and multi-national corporate profit and good old-fashioned American hegemony. And this edifice was going to built whether the indigenous folks who found themselves living in Iraq liked it or not. And the UN, EU, and other wimpy organizations were going to eat dust in the process. (Please refer to Naomi Klein’s “Baghdad Year Zero” for details).

Now a bridge to Hawaii might make a great design statement, it would however very likely pose some serious practical problems. In the same vein the dream of a corporate paradise in Iraq did present many serious military, budgetary, and time challenges. For instance since WW2 a great power occupying a foreign country has never defeated a nationalist inspired insurgency (only the British defeated a Marxist insurgency in Malaya). And even if the counter insurgency work was possible, history shows it takes an exurbanite amount of time and money.

But the architects of OIF were not stupid. They knew from past experience that although the demolition of the Saddam regime would be rather easy, the construction of their dream state would however take time and money. But they also knew their client would be loath to accept so expensive and long-term a project. And while there is no doubt the US military is good at demolition, how good really are their construction skills? So instead of telling the truth, the architects instead played the client for a fool and promised it would all be cheap and easy – a cakewalk in fact. And the client -- the American people -- bought it hook line and sinker. And what about those consultants the client should have checking up on his architect? Well the media and opposition parties got either sucked in by the architect’s enthusiasm for his project or were scared by threats to their pampered status and in any case they totally failed to do their jobs of controlling the architects.

And guess what? Building the new Iraq is not so easy. It’s going to take more time. It is going to cost more money. And it is not going to be the neocon corporate paradise once imagined. And in fact it might never get built at all. But who’s to blame for all these delays and cost overruns? Well if you ask the architect and his supporter it’s all the American people’s fault of course. And maybe they are correct about this, after all in 2004 the American people had a chance to fire the architect and perhaps bring a much reduced project to a close but they didn’t. They believed the lies that the building of the new Iraq, as originally planned, would be completed soon.

The architects also blame with exaggerated venom the opposition party and media -- some of who have ever so slightly started to do their job in protecting their client’s interests. Most, however, are still failing – perhaps in order to cover up for their lack of any diligence in 2002-3. And perhaps the architects are right about this too; if these two groups of people had called out the lies in 2002-3 then we wouldn’t be in this mess right now.

And so what is a client to do now, almost four and a half years after construction has started with nothing to show for all the blood and tears except for more blood and tears? Give his architect one more chance? Perhaps we really are “turning the corner” and the project will soon be finished. Or is the project so important that it must be built no matter what the cost? The architect and his supporters sure think so but that is usually a decision left up to the client, the American people. And no one should be surprised that the many Americans are totally disgusted with all the lies and just want to cut their losses by firing the architect and walking away from the whole project. In the real world that’s how things work.

8/09/2007 08:30:00 AM  
Blogger Ash said...

great writing Kevin!

8/09/2007 10:22:00 AM  
Blogger Ugmoe said...

Here is a link to an article about John Doyle who has spoken and researched the connection between robustness and fragility.

http://www.nasw.org/users/dsteele/Stories/Robust.html

The key thing to understand about robustness is that "it's a trade-off," says Dr. Doyle. "Well-designed systems are all robust and fragile" at the same time. "Strictly speaking, it's not right to say one thing is more robust than another. You can think about how efficiently something achieves robustness."

Dr. Doyle says these types of trade-offs are mathematically inevitable. "All those complexities introduce new trade-offs," he says. "If you understand those trade-offs, you can design them in such a way that it all works out. If you don't, then you could be in real trouble, and the worst thing is, if you really forget something ? if there's some substantial vulnerability that you create that you're not aware of."

Humans, for example, are much more robust to temperature extremes like a Texas summer and a Chicago winter. But simpler organisms, like bacteria, tend to be adapted to very narrow niches. Some have to live at human body temperature, and will die if exposed to boiling water. Others thrive at the extremely high temperatures and pressures found at deep-sea vents, only to die when brought to the ocean surface.

But in the event of a huge meteor striking the Earth, as is thought to have happened 65 million years ago, more complex life forms, like humans (or dinosaurs), would be more vulnerable as a group than bacteria.

"You can get in this spiral of increasing complexity. And that?s what biological systems do. They tend to spiral in complexity, because you sort of keep adding new features that give you new robustness, and that creates new vulnerabilities," Dr. Doyle says. "And then a big meteor hits and you start all over."

8/09/2007 10:24:00 AM  
Blogger dla said...

Start with a plan, measure your success, adapt. Americans sometimes think in terms of one-hour TV show segments - hardly reality.

I think that the Iraq effort is interating remarkably fast on the "measure and adapt" cycle. I believe the speed has been enhanced by Bush allowing Generals to work mostly unencumbered by politicians. Bush has borne the brunt of the political fallout, demonstrating his greatness as a President.

The Bush team had a plan. It sucked. Their generals modified it based on results. New General, new approach, better results. I don't think people appreciate the scale of the operation and the speed of the changes.

8/09/2007 11:50:00 AM  
Blogger Mike H. said...

Maximum load on a big rig is 80,000 lbs. More can be carried with proper permits and special equipment called jeeps. Loading on a big rig is determined by a simple formula, each tire has to have less than 500 lbs. per inch of linear width. Most truck tires are 11 in. wide and the laws were passed when the standard was a 10 in. tire.

My last truck grossed at 106,000 with 26 tires on the pavement with a loading of 370 lbs per in. A car has a loading of 1500/7x4=53.5 lbs. per linear in. The roads are designed for a maximum loading of 600 lbs. per linear in. of width.

There is another requirement for the big rigs and that is the Federal Bridge formula W=500(LN/N-1+12N+36) where W=the max weight in lbs that can be carried on a group of two or more axles to the nearest 500 lbs. L= spacing in feet between the outer axles of any two or more consecutive axles. N= number of axles being considered.

8/09/2007 12:10:00 PM  
Blogger jj mollo said...

Wolfram is a lot more than an educated idiot. I suspect he's wrong about the bridge idea, but I also think that engineers should definitely open their minds a little to what he has to say, just because he's saying it. Bridges are all overengineered for whatever it is we know about and plan for. But the two weaknesses are that design elements are all analyzed with failure probabilities, and there are always Unk-Unks.

Since you are going to overengineer a bridge, there are lots of choices about how to overengineer it. My own feeling is that engineers have an exaggerated attraction for elegance and simplicity that may actually be unwarranted for dealing with unanticipated challenges.

From an economics point of view, by the way, we can afford a certain number of bridge failures as long as they come from a probability distribution and not from a pattern of failure. It is sometimes more cost-effective to build more bridges of lower quality. Put your money elsewhere and manage the disasters. You always have to ask, how much security are we willing to pay for.

The big problem today is that this failure might be the tip of the iceberg since it may be the result of an explainable trend rather than a statistically predicted failure rate. Inadequate maintenance and replacement is a natural outcome of a political ethic that regards visible investments more highly than sound ones.

8/09/2007 12:41:00 PM  
Blogger Doug said...

Design
• The average life span of a deck steel truss a 75-100 year life span of a typical bridge.


Bridge capacity• The capacity of the bridge was rated safe for legal truck loads (up to 80,000 pounds per truck), which is the standard use for bridge design today. The bridge was rated to be safe to carry permitted (overweight) loads of up to 136,000 pounds.

Bridge Factsheet (PDF)

8/09/2007 02:54:00 PM  
Blogger Dublin Saab said...

I had an engineering class that involved four months of math but concluded with a bridge building project. I was stunned by how poor many of the designs were. After all, the bridges were made from drinking straws and pins, and since the school supplied the materials cost was not a factor in design. Yet looking at some of the bridges I had to wonder why, in the name of all that good, some of my fellow students were pursuing a degree in Engineering.

Needless to say the top performing designs were all arch bridges with a big drop off to the nearest truss design. I did of course win with the only arch suspension design (it supported a 46:1 weight ratio before failing).

My schooling instilled in me a dislike of truss designs like the I-35W bridge. Trusses, compared to arches or suspensions just have too many possible failure points.

This is why, even before the I-35W failure, I say the quick Our Father every time I cross the Tappan Zee but roll over the Henry Hudson (arch) or Bear Mountain, Whitestone and George Washington (all suspension) without a care.

If anything in this world should be over engineered it should be bridges.

8/09/2007 02:56:00 PM  
Blogger Dublin Saab said...

And nuclear power plants. Those should be over engineered too.

8/09/2007 02:59:00 PM  
Blogger Doug said...

Poor Trade off in this case:

"• The average life span of a deck steel truss bridge is about 50 years, compared to a a 75-100 year life span of a typical bridge."
---
Weathering is a known unknown, and this design was especially vulnerable.

8/09/2007 03:01:00 PM  
Blogger Doug said...

"My last truck grossed at 106,000 with 26 tires on the pavement with a loading of 370 lbs per in. A car has a loading of 1500/7x4=53.5 lbs. per linear in."
---
A professor of mine used to rail at the wildly disproportionate amount of fatigue produced by those two numbers.

8/09/2007 03:11:00 PM  
Blogger Doug said...

A House That’s Just Unreal

Second Life is a Web site that gives thousands a chance to build online houses that are Architectural Digest-style fantasies.

8/09/2007 04:58:00 PM  
Blogger seanderson7 said...

What is really useful in complex situations is a loose kind of algorithm that will enable the decision maker to understand what is important, within a given period, and allow him to choose paths to an improved situation.

OODA -- Wretchard, this could truly be a teaching moment. All hail John Boyd! Find out why the Marines revere him as a modern Sun Tzu and how he changed modern warfare.

I always wonder how much of the supposed chaos in the White House is simply diversionary. War is hell, and tricking your enemies into rushing over a cliff is no mean feat. First you need them to get on their feet, then rush, then arrange for a cliff, then -- it's mean.

OODA

In Boyd by Robert Coram. See The Ghetto Colonel and the SecDef at p. 425. Dick Cheney was there at the beginning and is now in the belly of the beast.

OODA

Winds of Change links to Boyd
Boyd and Military Strategy
Boyd in Wikipedia -- S7

8/09/2007 07:50:00 PM  
Blogger Larry said...

I liked the "carrying NAFTA traffic" comment. Can anyone actually believe that NAFTA traffic, I suppose from Winnepeg to Laredo? was the straw that broke the camel's bridge?

8/09/2007 10:54:00 PM  
Blogger Kathryn Cramer said...

The subsequent post on the Wolfram blog, by programmer Wolfram Research programmer Yifan Hu, has an animation of how the bridge could hypothetically have collapsed with the failure of only three pieces.

Presumably, this calculation could be run with any number of bridge shapes. This does not appear to be the actual series of events by which the bridge failed in this particular case, but does show the fragility of its design.

Wolfram Research as a company is pushing toward making real-world on-the-fly calculations of this type (and indeed of the kind that might have helped in planning for Iraq) easier. Version 6 of the program has extensive built-in data libraries, such as CountryData, which should help bridge the gap between abstraction and accurate simulation. (I worked as a consultant on compiling these libraries.)

8/10/2007 04:32:00 AM  
Blogger Marcus Aurelius said...

Whiskey,

Conduct a little experiment. Put a pan on your stove, place some oil into the pan and then place 20 kernels of popcorn into the pan. Before you turn the heat on, predict when each kernel of popcorn will pop. I was originally going to ask you to predict the decay of radioactive atoms but popcorn is so much more easy to do in one's home.

I recall a lesson (pages 69-77 of Solutions of Partial Differential Equations by Dean G. Duffy TAB Books Inc. 1986) in a text on partial differntial equations I have in my library. It starts out with a situation where engineers were trying to develop the quietest possible treaded tire. They set up the equations and work through everything the result: If we were to extend this technique so that the treads occurred at completely random positions then the treads would produce very little noise at many frequencies and the total noise would be comparable to that generated by other sources within the car. Next lesson - The use of Fourier Series in the Solution of Ordinary Differential Equations: The Theory of Suspension Bridges.

A bridge is little more than a network of support structures.

8/10/2007 08:33:00 AM  
Blogger jj mollo said...

There is a post on Wired about Wolfram's idea using a conch shell as an exemplar. As I understand it, the fractal-like cleavage planes essentially make the whole shell absorb a blow, rather than providing a single point of failure -- the conch form of kevlar I suppose.

The conch survived because its adaptive engineering was sufficiently streamlined to respond quickly to predator innovations, apparently using nothing but simple rules, but the rules resulted in non-simple structures.

8/14/2007 08:46:00 PM  

Post a Comment

<< Home


Powered by Blogger