January 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
An Interesting Introduction to the Man Who Changed the World
January 1, 2010
We, like many parts of the Midwest, got hit with a snowstorm last week. It was quite a workout shoveling the driveway.
The snowplow, of course, made it more difficult, dumping wave upon wave of snow at the end of the driveway.
While resting one moment, I looked at the street being plowed, and sympathized with the snow-plow driver. They really have a tough job, when one thinks about it. Get too far away from the curb, and they're criticized for not clearing the streets well enough. Get too close, and they'll likely scrape a front yard, or worse yet, cause damage to the curb - or their plow.
What a no-win situation.
Isn't there a way to mark front yards to make the job easier?
For example, what if neighbors put flags in their front yards, along the street, indicating "edge of yard"? Would that work?
It works fine along the east side of our street, but what would happen along the west side?
Obviously, this wouldn't work well. And a thought came to mind. One could create markers along the way indicating a "change of direction".
What's not so obvious is how this simple idea might impact industry. It did - and the world.
Pierre Bezier did something like this in the automotive industry, and the applications have since "changed the world".
How does one capture the essence of the shape?
Straight lines? Maybe. What about curved lines? Is it possible to capture curved shapes using points?
That's the essence of the Bezier method.
Let's suppose I wanted to draw the following curve, from Point P0 to Point P2:
What Bezier envisioned was a series of control points describing a polygon, and an algorithm creating the curve:
(image from Wikipedia)
Now: what's the algorithm? It can be described algebraically, but also geometrically, and that's where things really get interesting. You may recognize the following:
The next issue of =EQUALS= will be dedicated to the method of Bezier, including applications, the algebra, the addition of more control points, meshes, etc.
A website of interest: THE MATHEMATICS OF STRING ART: with the subject line: How String Art Changed the World.
I invite you to play around with Bezier-related applets at the phenomenal Wolfram Demonstration Project website ...
The Death Spiral of Systemic Stagnation
January 2, 2010
An interesting phenomena takes place in educational conferences. I don't think it's restricted to education, but this is where my experience lies.
But I want to start from a different perspective.
Let's suppose there was a way to "beat the house" in Black Jack - 21. It's beyond counting cards. It's beyond basic strategy.
It's easy to learn.
The inventors of the system advertise to millions who annually go to Las Vegas and Atlantic City to play Black Jack - most of whom lose their money.
Nobody tries their system.
For years. For decades.
Would this strike you as odd?
It is odd - and unexpected. What might explain such a phenomena? It might be the case "the customer is not always right". The customer bought Microsoft products over Mac by a 95 to 5 ratio for years. We, as customers, don't always "buy the best".
So let's make an assumption.
Let's assume a product is the best - and we know it. Why might we still not use it?
It's complex. It's great, but it's hard to use. It takes too much time to learn how to work. It ... you name it, there's an objection.
Let's try to explain this - in education - regarding great educational movements. There's a ton of great stuff out there, yet much of it is ignored.
Let's grant teachers want an exciting and challenging classroom. If that's the case, they, clearly, would want to integrate any new and wonderful technologies and opportunities available out there. Movements. Materials. Internet. Simulations. You name it. This being the case, they'd consider the new materials.
But not so fast.
We also know the teachers are given curriculums at the district level. They've got a classroom of 25 kids of different abilities, and the task of getting through a prerequisite amount of material in 50 minutes. Not only that, but they've got NCLB, AYP, and other testing mandates hanging over their heads. Integrate new material? Are you kidding? Who has time for that?
A conflict / dilemma seems the result: consider the new materials / stick to the existing materials.
But it's not really a conflict, as you can see below. Because of the need to focus on tests, the existing classroom, time constraints, curriculums from above, etc., there really is no conflict because there is no choice. This requirement dwarfs all others:
This seems to explain why, despite "pie in the sky promises" and the appearance of grand things, "the good stuff doesn't make it past the front door".
But let's take this a step further. With the unintended consequence of that which is possibly simple to understand seems instead more and more complex to the novice:
... which ties into the phenomenon of annual educational conferences, frequented by the same practitioners, presenting new and exciting things to colleagues, with less and less application in the educational environment actually in need of it!
Seems pretty hopeless, doesn't it? That is, of great educational movements making inroads into the "system"? If actual historical performance is indicative of future promise, it is hopeless.
Unless something changes.
Or someone / something decides to address either the constraint or the core problem - freeing up the cloud above.
Trying to Understand the Nature of Air Molecules
January 3, 2010
The weatherman said the temperature was 10 degrees, but it felt like 5 degrees below zero. That's cold.
But how do they measure it?
What is heat? What is cold?
Pausing for a few moments, I realize this simple thing I've actually taken for granted all my life. Let's spend a few more moments trying to figure out what is going on.
For example, I’ve heard it said so many times “hot air rises” it’s become a cliché with me. The hot air balloon rises, of course, which doesn't prove much, though it does give me some anecdotal evidence to help figure out the game.
But why does hot air rise? What’s going on here? Let’s suppose we try to create a model that tells us what’s going on.
My Ambitious Target
Before I Get Started …
Let’s try to gain some intuition about what’s going on with these “air molecules”. For example, merely creating the graphic above, lots of questions come to mind. What is it, for example, that is “expanding”? Is it the molecule itself – in size, or is the heat causing the molecule to “bounce around” more, taking up more space but itself not changing in actual size – or is it something else? And what is it that’s outside the space taken up by the air molecules - between the molecules? What is this “empty” space? To acquire some additional intuition about this, lets graph a couple of items that seem relevant here: temperature and air pressure.
My Logic and My Data
Not only do I have an idea of what’s going on, but I can also infer some things. For example, if heated air molecules “spread out”, the space occupied by these air molecules should become less dense than the surrounding space. Therefore, this warm air rises. If it rises, it’s replaced by colder air molecules “sinking”.
What can I infer from all of this? There should be a relationship between temperature and air pressure. When it’s colder, the pressure is greater, and vice versa.
What does the data tell us?
Something’s wrong – and that’s a good thing! Now it’s time to get to work fixing the model, the logic, etc. The real work begins!
For example, might it be the case I'm reversing cause and effect here? I say the heat causes the air molecules to spread out. What if it was the case air molecules that are spread out cause the heat? Maybe that's the case - maybe not. Something more to investigate.
A Thought on Thinking "Outside the Box"
January 4, 2010
It's a cliché, so I won't go there. "Think outside the box". HOW? This is a better question. Where does the new idea come from?
Another issue. Let's suppose you are IN the box - not literally, of course, but figuratively - and intellectually. How does one "step outside" the current environment to see what must be done to fix / improve the environment?
One method I use frequently comes to mind, particularly after reading a new book by Eli Goldratt: Isn't It Obvious?
A chain of retail stores is the context of the story. One store encounters water problems, and accidentally lurches onto a strategy to improve profit dramatically - and rapidly.
They forge a unique working relationship with distribution - the warehouse.
Seeing after a while what they have stumbled upon has some grit to it, they try to figure out a way to spread the solution across the entire chain.
The distribution manager can't figure out how to do it.
Neither can the man running the store where it's now working!
He stumbles upon another idea - totally unrelated to his business - but conceptually identical to his problem.
He - and the warehouse foreman - visit the book distributor.
They get the "Isn't it Obvious" solution. It was obvious, of course, once they saw it.
What helped them "see it"?
The following structure, most people have thought, I use for math only. Actually, I use it for math about half the time. The other half I use it for just about anything you can think of. It might be thought of as an "operational metaphorical / conceptual generation machine".
THE OPERATIONAL METAPHORICAL / CONCEPTUAL
How to Think Outside the Box - by Getting "Something on the Table"
That is, in this instance, it might look like this:
Which isn't to say I would have used it like this! By that I mean the following, which brings us back to the initial question, "How does one 'think outside the box'?", or more generally, "How does one come up with new ideas?"
Above, this came to mind after reading the section in the book. "Isn't it obvious?" But a problem I have sometimes is merely telling myself, "Get something on the table?" This simple box at the top has many avenues to be filled - and that's the neat thing.
January 5, 2010
Major book goals for 2010:
This three-part railroad project:
A book on Constitutional Origins:
A book on the issues of math:
A book on something that really bothers me - prediction, uncertainty, the display of information, chaos, etc. Given all that we know today, and all of the great books on these subjects, there still exist the same problems I'd expect NOT to see! Might it be the case these book hinder - rather than help - in understanding the complex - and simple - world we live in? How might the essence of all these principles be brought together in a small introductory book? That's the goal of this book ... working title with no image yet:
No foresight and no back-up plan hardly constitute "wisdom"
January 6, 2010
The First Book of Kings
3:16 Then two women who were prostitutes came to the king, and stood before him.
3:17 The one woman said, “Oh, my lord, I and this woman dwell in one house. I delivered a child with her in the house.
3:18 It happened the third day after I delivered, that this woman delivered also. We were together. There was no stranger with us in the house, just us two in the house.
3:19 This woman’s child died in the night, because she lay on it.
3:20 She arose at midnight, and took my son from beside me, while your handmaid slept, and laid it in her bosom, and laid her dead child in my bosom.
3:21 When I rose in the morning to nurse my child, behold, it was dead; but when I had looked at it in the morning, behold, it was not my son, whom I bore.”
3:22 The other woman said, “No; but the living is my son, and the dead is your son.”
This said, “No; but the dead is your son, and the living is my son.” Thus they spoke before the king.
3:23 Then the king said, “The one says, ‘This is my son who lives, and your son is the dead;’ and the other says, ‘No; but your son is the dead one, and my son is the living one.’”
3:24 The king said, “Get me a sword.” They brought a sword before the king.
3:25 The king said, “Divide the living child in two, and give half to the one, and half to the other.”
3:26 Then the woman whose the living child was spoke to the king, for her heart yearned over her son, and she said, “Oh, my lord, give her the living child, and in no way kill it!”
But the other said, “It shall be neither mine nor yours. Divide it.”
3:27 Then the king answered, “Give her the living child, and in no way kill it. She is its mother.”
3:28 All Israel heard of the judgment which the king had judged; and they feared the king: for they saw that the wisdom of God was in him, to do justice.
The Lack of Wisdom of Solomon
I don't think the wisdom was in Solomon. I don't think he thought things through. I don't think he had a fallback plan.
Here's his plan: unable to figure out who's baby it is, he decides he will cut the baby in two with his sword. Why? The real mother, he reasons, will not let her baby be cut in two. Therefore, the real mother - and the imposter - will be revealed.
So the story played out.
Not so fast. The imposter-mother has already told us she's a quick thinker - and a devious thinker. Are you telling me she's not seeing through Solomon's plan? She will object as well - probably faster than the real mother?
Where does that leave Solomon? Back at square #1!
Though he's probably not back at Square 1 - that implies he's right back where he started. He's not, actually. Hasn't he lost some credibility with the towns-folk? With the mothers?
An interesting dilemma. An interesting problem. Who to believe? Doesn't this problem play itself out throughout history? In current events? In our own households? Who does one believe? By what evidence?
January 7, 2010
THE CASE OF THE LEVERAGED THEORY
When we last met Archimedes, he was being cross-examined by Mike Mason in "The Case of the Submerged Theory", where Mason demonstrated, via a simple example, the errors in the Archimedean Theory of Buoyancy and the "Eureka" moment".
The two meet again.
The scene: Archimedes, on the beach, drawing a series a strange figures, connected, in the sand. Mike Mason approaches.
You’re too much, my friend! Do you ever stop?
Do I ever stop what?
There’s too much to do! Too much I don’t understand – and of that I do, there’s always more understanding to reveal.
This reminds me of a problem I’ve encountered with one of your experiments I’ve read about.
I’ve read where you personally, through a series of pulleys, lifted a ship.
I don’t believe it.
What don’t you believe about it?
I don’t believe it, that’s all.
Could you be a little more specific about your lack of belief?
Sure. What I’ve read says you hooked up a series of pulleys – lots of pulleys – and were able to lift a ship.
What don’t you believe about it?
Well, suppose I’ve got the following two pulley systems:
Perfect! A good start!
And I consider ONLY the parts highlighted.
If I thought about putting these portions in a “black box”, it shouldn’t matter how many pulleys I have. There’s no change in the amount of work done!
Wait a minute! You …
Is what I say right or not?
If we’re just considering the boxed-in section, you’re probably more right than you know!
What do you mean?
Imagining all the friction going on as one adds pulleys, likely one will be unable to lift more and more. That’s considering, of course, only what goes on in the boxes.
So you didn’t lift the ship?
Of course I did!
But you just said more pulleys makes things harder. How could you have lifted the ship?
You restricted me to your boxes and your assumptions. What makes you think that has anything to do with what really happened in reality?
I don’t understand.
What did Archimedes do? Was he right? Is he right? And what powerful lesson does he teach Mason, in the case of Archimedes' Revenge!
And the power of the simple notion: actions have consequences
January 8, 2010
Yellowstone Park is in the news, again, as the
Government and Park Rangers continue to try to manage the system - this
time, with rare Bison.
It's not been working very well.
And this practice has a long history.
A word on Yellowstone, first. When I say "Yellowstone", the image I have is a northwest under siege from over commercialization, rescued by well-thinking government officials.
The northwest, of course, wasn't well-traveled or explored in the nineteenth century. Little was known about "Yellowstone", in particular, until the 1860s. Immediately after this initial exploration, President Ulysses S. Grant set aside the approximately 2,000,000 acres of land as "off-limits" - in 1872.
The government works fast.
Before I go on about my main story, the irony of this land take-over is striking. Many land treaties with Native Americans were ignored and violated, and yet, here's US Grant saying this massive amount of land is to be set aside from the uses of man?
But I digress.
Michael Crichton gave a presentation at the Washington Center for Complexity and Public Policy titled "Complexity Theory and Environmental Management" in 2005.
A video of it is here:
The goal here is to try and model some of the dialogue from that presentation:
It wasn't long after the land was set-a-side Park officials believed elk were becoming extinct. How they knew this I do not know, because all natural populations, of course, are subject to variations. Nevertheless, because of this belief, they naturally encouraged the growth of the population. Therefore, there was an explosion in the elk population.
Of course, actions have consequences. It's all well-and-good to have many elk, but elk have to eat. What do they eat? They graze. And if there's lots - and lots - and lots - of elk grazing, what happens to those grazing foods after a short period? They diminish, of course. This is bad enough if you're one of the flourishing elk, but what if you're a deer? An antelope? Your food is being depleted, too!
The consequences - unanticipated - yet predictable. The antelope and deer populations declined.
Now, what should the Park Rangers do - if anything? Seeing declining and antelope and deer populations, they sought to save the animals. How might one do that? They killed the predators feeding on these animals!
Hardly addressing the system by dealing with the core problem!
To what end? Remember, the reason these animals were dying was because of a lack of available food. Now, there's being created an artificially high number of animals fighting for this limited resource.
In this little episode, we see a tremendous iterative cycle of destruction.
In this little episode, we see a tremendous iterative cycle of destruction.
But it didn't end there, of course.
Part of what these elk, deer, and antelope grazed on were aspens.
And beaver use aspen to build dams. Dams are an essential part of the Park ecosystem. Without aspens, the beavers - and consequently, dams - disappeared from Yellowstone. The result: meadows drying very hard in the summer.
Another assault on the available food for the elk, deer, and antelope.
Another viscous cycle.
More to come on the (mis) management of Yellowstone in Part 2 of 3 of this series, including the nature of "complex" systems, prediction, variability, feedback loops, etc., as well as general lessons on learning from the past - learning from mistakes - learning from successes.
Actions have consequences. Oftentimes, these consequences are unintended. Just as often, they're predictable.
A haiku on that subject from last year:
Charging Blindly to the Predicted Future
problem / solution?
think before you act!
The Quadrant Solution
In considering above consequences that are both predictable yet unintended, the "quadrant solution" is a handy device for examining the range of possibilities. For example, how would we describe the thinking of a person who undertakes an action with predictable consequences - yet they are unintended? This is what we see above, isn't it? How about "ignorance"! Is there a description for consequences that are intended - but not predictable?
This isn't to imply, parenthetically, all actions are predictable, but rather to say the mere knowledge actions have consequence - some predictable - some not - is necessary for rational decision analysis.
January 9, 2010
The Preamble - Narratively
We the People of the United States, in Order to form a more perfect Union, establish Justice, insure domestic Tranquility, provide for the common defence, promote the general Welfare, and secure the Blessings of Liberty to ourselves and our Posterity, do ordain and establish this Constitution for the United States of America.
We the People of the United States, in Order to form a more perfect Union,
- establish Justice,
- insure domestic Tranquility,
- provide for the common defence,
- promote the general Welfare, and
- secure the Blessings of Liberty to ourselves and our Posterity,
do ordain and establish this Constitution for the United States of America.
In separating each element, we're confronted with these terms used - and misused - often. The "General Welfare" clause has been used to promote most all social programs, for example.
Why was there a "Preamble" in the first place? A preamble "to what"? We had the Articles of Confederation. Were the men of 1787 granted power to write a Preamble? A new constitution?
Some thoughts to investigate with this new book in the works:
Making Progress on the Work of Pierre Bezier
January 10, 2010
A Final Thought (for today)
One of the things Bezier realized seems to have been this: in the manufacturing process, specs are passed along from person to person, department to department. Mere copies are altered slightly, small deviations made, etc., and the next person suddenly does not have the same spec.
How do you capture the essence of the spec? The drawing? Above, this is done via control points and an underlying algorithm.
Plug in the points, apply the algorithm, and out comes the shape.
Of course, above this is done strictly geometrically. There's an algebraic method to this I'm still trying to understand. For now, I'm content just playing around geometrically!
1 of 4 introductory articles regarding an upcoming 3-part educational series
January 11, 2010
This is the first of four introductory / brief articles on an upcoming 3-part educational book series tentatively titled "Two Rails To A Point":
The four parts have as their goal the brief introduction of an industry with a tremendous history, a fascinating present, and a wonderous future. Few people know this. Right here in Kansas City, we reside right in the middle of the history of inner-workings of this country!
1. Kansas City / Union Station and the USA rail-passenger center;
2. Coal / commerce and the intersecting of power and
goods right here!
4. Classification Yards, with Kansas City having one of the biggest in the USA.
Fast-Forwarding to the Present
A while ago, I watched a documentary of Hadrian's Wall, the Roman wall constructed across now-England in the 2nd century:
The wall, mostly built by Roman soldiers in the 2nd century, was approximately 73 miles long, 10 feet wide and 15-20 feet high. After a few years, 14-17 full-sized forts were added, each holding 500-1000 soldiers.
These, as I remember, were really fortified cities.
All within a short distance - 73 miles.
You would think something like this would last forever. Indeed, there are many parts of the wall still around.
And even if the wall were abandoned, you would think the wall would last forever - mentally.
A shocking part of the documentary has stayed with me: after a few centuries of it's construction, there were places along the line where residents did not even know a wall once stood there.
How could that be?
What happens to history?
FAST-FORWARD TO THE PRESENT
The Kansas City Union Station
The Grand Union Station of Kansas City, Missouri. It was rescued from demolition in the 90s. Inside this majestic building are many businesses, Science City, a Railroad museum, a restaurant, etc. Also inside is the Amtrak station. There are several arrivals and departures each day. This is a far cry from yester-year, when passenger and train data was staggering:
How can this be? 80,000 trains in one year is approximately 220 EACH DAY! That's almost 10 every hour - for each hour of the day - every day of the year.
Logistically, how did this happen?
It's tough to tell, looking at the "lay of the land" now. There are a couple rails outside Union Station right now, but nothing to suggest a national hub of transportation - except for the Union Station itself.
What is going on?
And then I came upon the following picture, giving me an idea of the logistics of this mammoth operation. You can see Union Station in the upper left, but what's the rest of that? Train sheds? A series of tracks diverging from the main lines? None of that is there today! Also, look at the massive infrastructure at the bottom of the picture - the servicing of many cars, the roundhouse.
But none of this is here today.
Here is "Union Station", viewed from the other side. There are at least 23 sets of tracks at this site! How can this be? None of this is here today? How did this all work? What happened to it?
This was just 1/2 of the picture from the book - the "before" picture. Here is the before - and after.
This is "Hadrian's Wall" deja-vu to me. You see, I once worked in the building in the foreground, and parked in this parking lot, where once the transportation hub of the United States rolled through!
(Pictures taken from Jeffrey Spivak's "Union Station").
The Story of Triumph and Tragedy - and an Engineering Phoenix, Rising From the Ashes
January 12, 2010
Four-score and one years ago today, at this moment (6:00 PM Pacific Time), "the last truly untamed section of the American railway system was to become a thing of the past," as "the weakest link in our transportation chain has been replaced by one of the strongest, and we can now regard our railroad as complete." (Gary Krist - The White Cascade)
So spoke the Vice President of the Great Northern, L.C. Gilman.
What was tamed?
The Cascade Mountains.
How were they tamed? With the opening of the New Cascade Tunnel, a 7.79 straight-line railway hole through the heart of the mountains.
The Cascade Range, west of the Rockies. Size-wise, one appears as a parent to a child.
But size can be deceiving. The comparison of the Rockies and the Cascades, travel-wise, is striking, if we are to take the words of those who actually traveled over them. From one who who crossed the Cascades along the Oregon Trail: "The crossing of the Rocky Mountains ... was insignificant in comparison to the Cascades."
The goal of the New Cascade Tunnel was to connect the already-thriving Seattle directly with the rest of the country.
It was an engineering marvel, this 8-mile tunnel through solid rock. To expedite the tunneling, teams not only dug from both ends of the tunnel, but from two points in the middle! Yes - a shaft at Mill Creek was bore, and workers lowered into the shaft to dig in both directions.
Four points of progress.
Stop and think, for a moment, what it would take to align oneself so the tunnels actually met!
The New Cascade Tunnel. More to be written about this engineering marvel and the people who made it happen.
You see, here I want to talk about what was before the New Cascade Tunnel. Though it was New, it was not First.
The First Cascade Tunnel was itself an engineering marvel, replacing the eight switchbacks which zigzagged up the mountain-slope face, sometimes overcoming the mountain in 75 minutes, others up to 36 hours.
It all depended on the weather.
To solve this problem, the First Cascade Tunnel was completed in 1900. At 2.6 miles long, it erased the problems of switchbacks.
But it created problems of it's own.
High in the mountains, of course, snow falls - not in inches like most people understand "snowfall" - but in feet.
To protect the rail along the slope not in the tunnel, massive snowsheds were built. 40,000 feet of snowsheds.
But this was only one line of defense against the snow. The most powerful army to attack the snow were the rotary snowplows.
More powerful than the single rotary snowplows were the doubles, allowing the train to clear the tracks both in front of - and behind - them, without having to turn around.
In extreme weather conditions, these massive parabolic-shooting snow machines would escort trains through the mountains.
And in late February, 1910, the weather was extreme in Wellington, Washington, on the west side of the Cascades. An enormous blizzard.
It snowed - on one day as much as 11 feet!
And two trains, both headed to Seattle, needed escort across the mountains. One was a passenger train, the Seattle Express No. 25, running from St. Paul, Minnesota to Seattle, the other a mail train, also from St. Paul: The Fast Mail No. 27.
The high-priority 25 sat at Leavenworth, awaiting an escort, while dispatch weighed a tough decision:
The Fast Mail, you see, had top priority on the tracks, as it was delivering mail across country. The Great Northern was hoping to secure a large government mail contract, and hoped on-time delivery would provide the evidence of good service.
Never-the-less, the Fast Mail was hours behind the waiting 25, and the order was given to send the 25 up the mountain with a rotary escort, the reasoning being the track left behind should be clear enough for the 27 to make it up the mountains.
This, it turned out, was an insignificant decision.
Though rotary snowplows are magnificent in the job of removing snow, they too have limitations. If side-swiped by an avalanche, they can become stranded, as any other train. Additionally, with an avalanche comes not just snow but likely surrounding tree and shrub debris which, when sucked into the rotaries, clog the machine.
Both of these happened atop the Cascades with the rotary escorts, and both the 25 and the 27 became trapped at the Wellington, Washington depot.
Single rotors were dispatched to dig the trains out, with no success. Double rotors. Hand diggers. With no success.
And it continued to snow.
With heavy, wet snow falling on light, on a steep slope, it was inevitable.
Both trains were swept from the tracks, and down the canyon.
96 people died. The greatest avalanche disaster in the history of our country. The second greatest rail disaster. The Wellington, Washington avalanche. March 1, 1910.
From this tragedy came the New Cascade Tunnel.
It's still in use today.
Follow-Up Research Items
As with any story, retold like this above, there is a ton of work done to be able to rewrite what one think actually happened. When trying to rewrite it, the holes in the understanding are easily exposed. So too are holes missing from the presentations of others in their write-up!
Additionally, the story serves as an excellent entry-point to explore a number of things otherwise uninteresting. Here are a few from this story:
Snow in the mountains - why? All mountain ranges? Both sides? Why so much?
Mountain ranges themselves.
Avalanche theory / chaos theory / tipping points / etc.
Decision theory - not just on the part of the railroad (send the passenger train, hold it), but on the part of the passengers. Several left the train during the extended period prior to the avalanche, marched to a hilltop overlooking the Scenic hotel, and slid down a 1000 foot hillside. Can you imagine that? Not sledding down a hill, mind you, but simply sliding down a hillside, hoping you don't hit anything! It was, of course, a good bet. Here - but always? Under what conditions?
Aligning tunnels. How did they do this?
The rotaries became clogged with mountain debris. Isn't debris expected? How might one design a rotary so it filters out the crap, and only sucks in the snow.
Additionally, the steam-driven rotary locomotives needed water, which, when heated, drove the pistons. Where does one get water? They got it from snow. However, as with the rotary debris issue, so too was there an issue with crap being sent through the system, clogging it. How would one design a snow filter?
Another issue is one of losing focus. In the process of attempting to fix the rotary, the weather cleared slightly during the blizzard. The goal of the Great Northern, in attempting to fix the rotary, may have been misplaced, moved from "how do we rescue the passengers" to "how do we clear the tracks"? The former goal may have motivated them to evacuate the passengers down the hill. How does "focus" change? How does one ensure the goal is right and they're always pursuing it?
These are just a few, so far ...
Here's another one I just came onto ... looking at the passenger train video below, you can tell which direction the train is going. How? In "Burlington Northern Santa Fe Railway", Brian Solomon writes,
"To overcome the problems of diesel exhaust in teh Cascade Tunnel, GN installed enormous ventilating fans at the cost of $650,000 to flush fumes from the tunnel after each train. At the east portal in Berne, two 6-foot fans were powered by 800-horsepower electric motors, and a door was installed to maintain the airflow necessary to evacuate fumes from eastward trains climbing through the bore. Descending trains do not require ventilation, as they can drift down grade."
Down grade? How's this? Is there a tunnel length where ventilation is not an issue? Are there other considerations?
So much understanding to go! The research continues!
How does one put words - and meaning - to the world of data we live in?
January 13, 2010
The KC Star sub-headline read as follows:
"Homicide count for 2009 is down from 2008".
The numbers? 2008 homicides were 126; 2009 110. The title seems right.
What kind of question is that? Of course the count is down. The "numbers speak for themselves", right? And there they are!
Before investigating this odd question, let's read a bit of the article:
The front-page article is continued on page 12, with the following headline:
HOMICIDE: Nonfatal shootings up
Surely, there's a relationship between fatal and nonfatal shootings. One merely has to say, "Had a few people 'shot but not killed' been killed, the homicide count would have been up!"
This obvious sentiment is echoed in the article by Major Anthony Ell, who commands the East Patrol Division: "An inch here or an inch there and the homicide total could have been the same as 2008."
So is it accurate to say "the count went down" when there's this much variability in the system?
Let's look at the data:
There's a lot of variability in the annual numbers. What does it all mean? Did the Star report last year at this time homicides were up 32?
What are the implications of saying something like this? "We need more police?" "KC is a violent city?"
What are the implications of saying homicides are down? We're safer? Are we?
Are these sentiments correct?
Are things really changing?
They seem to be changing every year!
Or are they?
Suppose, for example, I flip 10 coins and get 4 heads. You'd expect 5, but know - intuitively - sometimes you'll get 5, sometimes not.
I got 4.
I repeat the experiment, this time getting 7 heads.
You're reporting the experiment. You've recorded the data. You report the following:
"HEADS UP DRAMATICALLY IN COIN-EXPERIMENT".
The headline is recording the count - accurately - but is it reporting the count - correctly? We know there's something wrong with this headline. Nothing really changed, and the data from experiment to experiment is merely subject to variation.
But how do we explain it? Report it?
Let's expand on this coin example to build a bit of intuition on the subject. Suppose I take, for example, 100 coins, and flip them. What would I get? 55? 45? 60? 42?
Here's a simulated result:
Now, we know these are really "all the same"; that is, everything going on is simple "the luck of the draw". Any report highlighting a specific number, we know, is wrong, and can lead to wrong results.
What about just reporting this graph? Is that enough?
We, hopefully, want to do something about the underlying system we're monitoring. Perhaps a secondary part of the "report" fixates on the system giving rise to "the same data", year after year.
That would be a start.
But there's more here. For example, suppose I hit 70 heads on one iteration. That's meaningful, isn't it? Not likely, but possible.
How can we modify our graph to catch some anomalies? Indicators. Control limits. Red flags. Markers yelling, "LOOK AT ME - I'M SIGNIFICANT! THERE'S A REASON I'M HERE!"
That's where the real reporting can take place.
(more to come on combining fatal & nonfatal shooting data to augment the rational reporting process)
Some thoughts on problem-solving
January 14, 2010
The I-70 corridor is a densely-packed multi-lane highway that, among other things, connects St. Louis to Kansas City.
In addition to the hundreds of thousands of cars that drive this system, 10,000 trucks are on the roads as well.
And the rail track is even more dense. Coal trains. Intermodal trains. Amtrak.
Amtrak just released this new video:
Increased on-time performance is promised due to track siding being built in California, Missouri. This will allow freight trains to "stand aside", passenger trains "passing on the left".
Where did this idea come from?
The following study:
While reading through this study researching my books above, I came upon the following (page 7):
I couldn't believe it! I read on, with interest (Page 9):
And, in the appendix (pages 35-36):
From this study came the California siding initiative.
A couple thoughts I've been wrestling with, regarding this analysis via TOC and the relationship between the constraint and the core problem (I know, for many of you, you're tired of hearing that lament!):
1. What is the relationship between the constraint and the core problem? (I know, for many of you, you're tired of hearing that lament!) Here, the researchers, in the analysis, start with core-problem identification via the Current Reality Tree, but then, in the appendix, define TOC according to constraint management.
2. The UP apparently has spent over 1/2 billion dollars over the past decade on track improvements along this corridor. The California siding costs less than $10 million. Is the 1/2 billion track maintenance, or have they done similar studies on how to "increase throughput", but to little effect?
3. California was chosen because the rails narrow from two tracks to one. The recommendation makes sense. However, if this is a bottleneck, I'd expect trains through here all the time. Likely, there's not, which means the bottleneck is not being properly managed.
4. Which leads to my last thought for now. The researchers claim the identification of the core problem led them to look for means to improve traffic flow. Maybe this is right. Is this different than starting with trying to identify the constraint and manage it?
More to come!
A Chronic Conflict Unresolved
January 15, 2010
The core-problem versus the constraint, Part II.
In Part I, I-70 rail corridor researchers identified a series of problems in the existing rail system from Kansas City to St. Louis. Like a doctor seeing a patient with many systems, who realizes one health issue is likely responsible for the myriad symptoms, so too the researchers sought to find that core problem responsible for the many system woes.
They found it in "Increased Train Load".
Their solution? There was a bottleneck at California, Missouri, where rail narrows from two tracks to one. Add a siding.
What's missing from this?
However, if I go to California, Missouri, right now, I'd bet dollars to doughnuts there is no train going through California, Missouri.
And I'm reminded of this scene from The Goal, where Jonah and the workers approach the NCX-10 machine, which has been identified as a system bottleneck:
Alex: "This is our NCX-10 n/c machine," I tell Jonah as we arrive at the big machine.
Jonah: "And this is your bottleneck, correct?" asks Jonah.
Alex: "One of them," I say.
Jonah: "Can you tell me why isn't it working right now?" asks Jonah.
Indeed, the NCS-10 is stopped at the moment.
I'm also reminded of a recent story from a friend in Michigan, who showed a video of one driver during rush hour acting in such a way the typical traffic jam was avoided. Phenomenal!
So there's something missing here. On the one hand, if there's more vehicles, in the rail system, there's increased traffic load. That's the core problem, and it makes intuitive sense.
On the other hand, if there's more vehicles, in the car system, there was not increased traffic load!
How can this be?
Moreover, something must be missing. What? Let's see what the driver did during rush hour as help. While driving along, he was tired of sitting in rush hour, bumper-to-bumper. A problem was the lane on the right, where cars would rush past, and then try to merge to the left.
We all know what I'm talking about. You also know how you feel when this happens. You dare not let that cheater in!
For the purposes of experiment, the driver thought, "Suppose I keep enough distance between me and the driver in front of me so anybody racing ahead can get in. Let them in early. What would happen?"
Traffic flowed smoothly, and he got to his destination quicker.
Non-intuitive, for sure, and I'm not even recommending one do this, thus rewarding the cheaters. However, it is very interesting to see how increased traffic volume did not equate with load issues. That's what was claimed in the rail example.
The car driver managed flow. He managed the system bottleneck.
Given what was said about California, Missouri, earlier regarding the (probably) absence of trains throughout the day, might it be the case train flow along the I-70 rail corridor is not managed according the bottleneck?
That is ... (minimizing the large image, now, but leaving it to remind the reader of the context of the rest of the article):
Problem solved! Let's ensure dispatch at St. Louis and Kansas City manage train flow according to the bottleneck.
Pretty obvious - so obvious, one wonders why they're not doing it already.
Why might they not be doing it?
One would wager, just as employees are rewarded for on-time performance, so too dispatch likely is rewarded for turn-a-round time, or some such metric.
To manage the flow - and bottleneck, of course, doesn't so much care about getting from or to a particular destination on time, but rather, to move everything through the system faster. RAIL THROUGHPUT!
But the conflict is interesting:
In order to have a well-run train system, we should reward dispatch for on-time performance. Most all businesses reward their employees this way. You meet your goals, you get a bonus. Of course, in this instance, we see it leads to dispatch not managing the flow of the system, or the bottleneck.
On the other hand ...
In order to have a well-run train system, we should seek to reduce point-to-point travel time, which requires us to manage flow - and the bottleneck.
Quite a conflict. Probably a chronic conflict! And a common one: Manage locally - or manage globally. And what are the implications? A California siding might improve throughput some, but what of the underlying conflict? Is it addressed much?
More important (to me) than whether it's addressed much is this: stopping at the core problem fails to verbalize - bring to light - a more powerful vision of the railroad and its operation.
Just Playing Around
January 16, 2010
Binary Symmetric Hieroglyphics: The First 120
Binary Symmetric Hieroglyphics:
Binary Symmetric Hieroglyphics:
Moving along ...
Binary Symmetric Hieroglyphics:
Moving Further Along ...
Binary Symmetric Hieroglyphics:
And more ...
Binary Symmetric Hieroglyphics:
Closing in on the end of the algorithm ...
Binary Symmetric Hieroglyphics:
Reflections on The Law of Unintended Consequences
January 17, 2010
Asked about the "Action-Quadrant" in the Yellowstone post above, it gives me a chance to not only revisit it, but also the initial article that gave rise to it.
The generative thought, actually, came from that oft-repeated plea, "But I didn't mean it to happen". Google "Law of unintended consequences" and you'll see example upon example of this phenomenon.
I'm tired of hearing "unintended consequences". Why? Most of the time, though the person may not have intended the consequences to happen, the consequences clearly are predictable!
What kind of thinking gives rise to this phenomenon?
A lack of it?
Here's the initial article ... on ethanol and the price of corn - and other issues:
The Simultaneous Nature of the Predicted Yet Unintended
The Quadrant Solution
In considering above consequences that are both predictable yet unintended, the "quadrant solution" is a handy device for examining the range of possibilities. For example, how would we describe the thinking of a person who undertakes an action with predictable consequences - yet they are unintended? This is what we see above, isn't it? How about "blinded thinking"!
Is there a description for consequences that are intended - but not predictable? How about "wishful thinking"?
Let's fill in the quadrant to see what we've got:
"This is John Galt Speaking"
(Ayn Rand: Atlas Shrugged)
"The law of causality is the law of identity applied to action. All actions are caused by entities. The nature of an action is caused and determined by the nature of the entities that act; a thing cannot act in contradiction to its nature."
"The law of identity does not permit you to have your cake and eat it, too. The law of causality does not permit you to eat your cake before you have it. But if you drown both laws in the blanks of your mind, if you pretend to yourself and to others that you don't see—then you can try to proclaim your right to eat your cake today and mine tomorrow, you can preach that the way to have a cake is to eat it first, before you bake it, that the way to produce is to start by consuming, that all wishers have an equal claim to all things, since nothing is caused by anything."
A Final Note
A final note on the quadrant solution above: notice that it does not evaluate the rightness of the action. For example, I can role a boulder down the hill, intending my neighbor's fence be destroyed. Predictable? Yes. Intended? Yes. This doesn't make it right.
It does, however, put this person on a different plain than the person operating in the diagonal quadrants above. At least the boulder-rolling individual is recognizing in accordance with the laws of reality!
Martin Luther King, Jr. Day
January 18, 2010
A Conflict I Still Wrestle With
What was the operational message of MLK, Jr.? Peaceful nonviolence? How does this work in reality? One sits at a restaurant counter where the sign says you're not allowed, and demand service. The police come, and you're arrested. You're struck. You're put in jail. This, done on a grand scale, must bring attention to the issue. It becomes an issue that cannot be avoided. It must be dealt with. Part of me agrees with this.
On the other hand, if you refuse me service while serving the man next to me, my first inclination is to throw a brick through your window. If you strike me, be assured you will be hit harder. But does this make the situation better, or worse?
Is there a context, I wonder, where both are relevant? Both are practical? Likely.
Watching - rewatching - Exodus, the quote below I believe wonderfully depicts the goal, whether the issue is Jews, Arabs, Blacks, Whites, or any group of people. And herein may lie a direction to a solution. We talk in terms of "groups" of people. We are all individuals. Wouldn't it be nice if discussions took place under the auspices of the nature of man, man's nature, man's relationship with reality, and a political system consistent with and recognizing that nature? Individual rights!
A QuickStart Publication
January 19, 2010
The Historical Background of Julius Caesar
A Look at Math
January 20, 2010
2009 airline revenue: Worst drop ever
By Ben Rooney, staff reporter
January 20, 2010: 1:05 PM ET
NEW YORK (CNNMoney.com) -- The airline industry suffered its largest drop ever in passenger revenue last year as a weak economy grounded many would-be travelers, an industry group said Wednesday.
The Air Transport Association of America said total passenger revenue for the major U.S. carriers fell 18% in 2009 versus the year before. It was the largest drop on record, exceeding the 14% decline in 2001.
The revenue decline was due to a 6% drop in passenger volume, and a 13% plunge in the average price paid to fly one mile, the ATA said.
I know I should be able to see, from these items, the fall in revenue is 18%, but it just doesn't come to me. It does no good to say, "Multiply (0.94)(0.87) and you get 0.82."
In my head, most of the time it doesn't compute.
And I'm glad - for a couple of reasons:
1. it makes me really think about what is happening.
For example, here's how I'm told the calculation works:
Probably right, though I'm at a loss, of course, if I don't see this immediately. Also, the 0.8178 shows up as the opposite end of 100%, so I must subtract this from 1.00 to get the correct answer.
Intuitive? Obvious? To many, sure. To many, including me, usually not.
What's the alternative? Let's add some units to the calculation, so I can, at a minimum, see how things are canceling themselves out:
This is a little better. What else? Here's one of my favorites: GET RID OF THE FRACTIONS AND PERCENTAGES! DEAL WITH ACTUAL NUMBERS!
What actual numbers? That's the great thing about it. They don't have to be real - just numbers. Say you've got 100 people flying at 100 dollars per ticket. That's $10,000 in ticket revenue. Now, I've got 6% fewer people flying at 13% less in ticket price. That means 94 people flying at $87 per person, with new revenue at $8,178. That means revenue is down about 18%.
Of course, words are just words. If you ever say the above to me, I will say, "Maybe - let me see". There are simply too many facts going on for me to keep track in my head. I've got to get something organized on paper:
You see how the shaded region becomes the last - and easy - calculation.
But I said there were a couple reasons. That was the first. It established confidence in me!
2. I'm doing the problem right. I'm a pretty math-literate guy, and I know I have trouble with problems like this. Imagine the many people math-illiterate whom I'm talking with? Do they understand the straight-forward approach, going from multiplication of fractions to the reciprocal of 100%? Maybe. Usually not. However, these are bright people who can see - and appreciate - the logic of the table above.
Therefore, there is a tremendous boost in communication and group confidence when the numbers are on the table.
And, it's been my experience, when then numbers are on the table like this, verbalized, visualized, and understood, at least one of them is wrong. And people only realize it's wrong when the see what's actually happening.
That is, the math is usually right, but there's something wrong with the problem being solved, be it a more appropriate number, assumption, or whatever.
As Gene Kranz said regarding the Apollo issue: LET'S WORK THE PROBLEM, PEOPLE. LET'S NOT MAKE THINGS WORSE BY GUESSING!
A QuickStart Publication
January 21, 2010
The Logic Chains of Julius Caesar
The Inter-Urban Railroad Connecting Olathe to Kansas City - and It's Origins
January 22, 2010
The Inter-Urban Railroad Connecting Olathe to Kansas City
December 17, 1903. Kittyhawk, North Carolina. The Wright Brothers. Flight.
Twenty-four years later, Lindbergh made the first trans-Atlantic flight, flying the The Spirit of St. Louis from New York to Paris.
From "off the ground" to "across the ocean" in 24 years:
Of course, in any paradigm shift like this, there are incremental steps. Look at the Wright Brothers picture, for example. Yes, they were off the ground - barely. When did real altitude first take place? 500 feet above ground? 1909. Charles K. Hamilton.
Lindbergh took us across the Atlantic. Who took us across the country? Calbraith Perry Rodgers: 1912.
Hamilton's flight took place at Aviation Park.
Rodgers' headquarters for his trans-continental flight was - also - Aviation Park.
Where is Aviation Park?
Yes, Overland Park, KANSAS!
Why Kansas? Why Overland Park?
Because "Mr. Strang believes in Overland, works for it, talks for it, spends his money for it, and knows from his thorough knowledge of the United States, in which he has built railroads in nearly every State, that Overland, for a healthful place, a beautiful place, a moral place, and a good place for investment and home building, can not be excelled and proves his faith in the people and town by living here, himself, with them and being one of them."
Who was "Mr. Strang", why was he here, and what has this to do with railroads?
It's an interesting story. And it starts with water. Lots of it. A confluence.
Confluence - a flowing together. That's what the Missouri and Kansas Rivers do in Kansas City. Two snakes, meandering from high ground to lower ground ...
In 1903, it rained - and rained - and rained. The two great rivers grew. The great flood of 1903. Kansas City became flooded. Over 100 people died. The extent of the disaster is shown by these "before and after" pictures:
However, a similar "confluence" of events was taking place at this time, and is often the case, from tragedy comes triumph.
You see, William Strang's mother lived in Kansas City.
Who was William Strang?
He was a man who was tired of the continual flooding of the town. Of residents being uprooted, of homes being destroyed, because of the low-lying downtown area.
But what to do?
He looked southwest and saw land on a bluff. Yes! Safety from low-lying terrain! Serenity from the downtown life. A new home!
But remember, this is 1903. There is no automobile yet. This is why people lived where they did.
How to get from here to there - and back again.
That was the given, to William Strang. You see, he was a railroad man, and he knew he could build a railroad - an interurban railroad - from Kansas City to his new town - his park overland - OVERLAND PARK!
But that's not the end of the story.
Many trolley and interurban systems in the country were electric at the time, and he didn't see the sense in investing in such infrastructure. But what's the alternative? A self-propelled gas-electric vehicle. A hybrid! The first IN THE WORLD! 100 YEARS AGO! RIGHT HERE IN OVERLAND PARK!
The Strang-Line Interurban: A Few Pictures
It doesn't matter that the vehicle didn't work as intended. There was too little power generated to make the inclines. The cars, made in Pennsylvania, were serviced by parts only made in Pennsylvania. The technology was ahead of its time.
But there it was!
And it was beautiful! Irene was divided into chair car and parlor sections. Inlay and glasswork put car right up into a class with some of the most opulent “private varnish” of the Age
Interurbans Without Wires
A modified railroad, the Strang-Line Interurban, eventually snaked its way to Olathe, operating until 1940.
And so I dedicate this article to William Strang, founder of Overland Park, visionary, who lived ... "right around the corner".
The Strang Statue - Downtown Overland Park
The Strang Mausoleum
A Few Research Projects to Follow-Up On
How did the gas-electric work?
What was the route?
How did Kansas develop from 1900 - today?
What is the relationship between this and light-rail today?
What are the dynamics of air, rail, and automobile traffic?
An anomaly: steam powered trains half a century before the automobile. Why did the automobile not develop sooner?
An anomaly: diesel-electric continues to power rail traffic today, much more efficiently than automotive technology? Why the gap?
A thought: one wonders how the interurban would have faired, had the car been invented 10-20 years later? The two (here) were nearly simultaneous.
How might things be different if an infrastructure were established? (On the other hand, would the interurban even have existed had the car been invented 10-20 years earlier?)
Mine is a story of trust and betrayal - and learning to trust again. I remember everything.
January 23, 2010
This is a story about me – the design. He drew me hundreds of times when he was young. But then he quit. Maybe he outgrew me. I don’t know.
But he got interested again, with computers, with spreadsheets, and started re-creating me – hundreds of me!
But I wasn’t the same.
He compromised. And then he quit. I thought I was not to be.
What prompted him to take another look at me? I don’t know. Maybe he knew he had quit. Maybe he knew he had cheated. Maybe he knew there was unfinished business. Maybe he knew – if he tried – he could bring me back to life.
This is his story, too. But for now, this is …
I’m very disappointed. I’m like an acorn, eagerly anticipating my growth into an oak tree. If I were this acorn, I’d just need the right circumstances, rain, weather, soil, animals to leave me alone, and I’d make it. A beautiful oak tree.
But I’m not an acorn. I’m really just a design – the idea of a design, actually. I’ve been drawn by hand, pencil, and ruler, hundreds of times in my life, and I’d hoped to be drawn by computer as well.
I almost was.
And this “almost” makes me sad.
Let me tell you my story.
He once was excited about automating the process. To create me, you see, means following a simple rule many times. That’s it. I’m a perfect candidate to be automated!
He wrote a simple set of instructions for kids to re-create me – many versions of me. How I looked forward to looking so different – yet the same! Many versions of me!
The simple set of Steps looked as follows ...
HOWEVER ... I immediately noticed something was wrong. My story continues next week ...
Mine is a story of trust and betrayal - and learning to trust again. I remember everything.
January 24, 2010
I immediately noticed something was wrong. I never had this “grid” in my middle, for example. My lines always stopped while descending downward, right-to-left, when they hit an intersecting line, and then restarted at a later point, completing the segment.
Why had he changed the program?
I knew why. He didn’t know how to write the program to stop at specific points – and then restart.
So he compromised.
I didn’t like my new look.
I also noticed my “top” and “bottom” were set as “inputs” by the user. I could have a very tall top, a short one, anything the user wanted. That sounds fine until you understand I was usually drawn so my top and bottom lines were parallel to the rest of the diagram.
I looked ugly.
Why had he done this – to me?
Again, my thought crept to compromise. He didn’t know how to calculate “parallelness”, and in the haste of getting the program written, left it to the user. He quit on me – and he quit on himself.
I wasn’t happy with my modifications.
He wasn’t either. But he didn’t know how to fix the problems. Unsatisfied, he left me alone.
And years crept by. He had shoved me to some obscure directory, continued to back me up, but the “recent date used” remained unchanged.
He had forgotten about me.
I was like a wonderful toy, forgotten in the bottom of a toy chest.
Sadness overwhelmed me.
But I started to notice a change in him.
He started documenting his work. To write short books. There’s no escape when you’re writing short documentaries on how to do something. It doesn’t mean the writer reveals he knows everything. On the contrary, the honest writer admits, when they’ve encountered a problem they don’t know how to solve, “I don’t know how to solve this.”
But this was programming and a spreadsheet.
And I knew he could program anything in a spreadsheet – if he only tried. If he “got something on paper”.
And he was.
“Just draw the image” became my mantra. “Draw the image and it will come to you.” He couldn’t hear me, of course, but the mantra provided me hope.
And he did drew it.
And it came to him, like I knew it would!
“Let’s get this done”, I heard him say. “For goodness sakes, it’s just lines, and I know everything about lines!”
“I want to draw the top section of the design. It must be parallel with the body of the design. What do I need to ensure ‘parallelness’? The slope, of course!”
“Get things on paper!”, I heard him, and the answer came easily. The calculation of the slope of line q was immediate:
and, applying this slope to the Point D coordinates (which he knew), allowed him to find the Point C coordinates (which he did not know):
Of course, there was a bit of work to actually find xC and yC. For example, xC was merely the midpoint of the design, but it still had to be found. This was easy. So was finding yC, which was found with the following process:
He had all the information. Programming it into the spreadsheet was the next part. Most people think programming is hard – what they don’t realize is it’s the easiest part of the process! Figuring out what to program is where all the work takes place!
He also had four parts to deal with: The top right, top left, bottom right, and bottom left. Of course, once he perfected one quadrant, I knew the other three would come quickly.
He hit “execute”.
And there I was. Multiple versions of “me”, like a person putting on different sets of clothes. The same person, looking different. I had my “top and bottom” back, and I, again, liked myself!
He was only half-way done, however. I still had that ugly “grid” in my middle. Would he fix that, too?
I don’t have to worry about him stopping now. Once he’s got momentum, it’s hard to stop him.
He won’t dare let this middle grid stand.
But this is a different problem. The downward line, moving from right to left, stops, then restarts, and concludes at the given point.
What will he do?
The first thing I’m certain of is this: he won’t make the same mistake twice. Get something on paper! He does!
Stay tuned for Part 3!
A QuickStart Publication
January 25, 2010
January 26, 2010
Watching a documentary on meteorology yesterday, I was struck by the straight-forwardness of the following statement:
60 knots is about 69 miles per hour, because:
1 knot = 1.151 miles / hour
Where did this come from?
Let's make a quick investigation.
A "knot" is equal to one nautical mile per hour.
OK - what's a "nautical mile"?
A "nautical mile" is "a unit of length corresponding approximately to one minute of arc length of latitude along any meridian."
Now wait just a minute. Where did this come from? Let's do a bit of work to see. I know the earth, nearly a sphere, has 360˚, and within each degree there are 60 minutes of arc (and within each minute of arc, incidentally, are 60 seconds of arc). That is:
Where does this leave me? I know, by definition, a "nautical mile" is approximately one of these. And I have 21,600 of them! Therefore, the earth's circumference is approximately 21,600 nautical miles. If I apply the 1.151 factor to this, I get 24,862 miles as the circumference of the earth. That is:
But you probably see the holes in all of this, right? The biggest one is this: What came first: the circumference or the factor? And why do "nautical" miles differ from "regular" miles?
What is the meaning of all of this?
To be continued - with an extraordinary conclusion!
Mine is a story of trust and betrayal - and learning to trust again. I remember everything.
(Part 3 of 3)
January 27, 2010
(continued from Parts 1 and 2 above) ...
The first thing I’m certain of is this: he won’t make the same mistake twice. Get something on paper! He does!
“That’s it --- keep plugging away.” I silently urge him on. I listen to him talk.
“OK – I want to draw a line from Point C to D, but it’s got to stop at Point G, and then restart at Point H. Let’s concentrate on Point G first.”
“G is the intersection of two lines, p and r. How do I find a point of intersection between two lines? I know this? The x and y coordinates of Point G are the same for both lines p and r. Therefore, if I had the equations for lines p and r, I could set them equal to each other, and find the coordinates!”
Of course, there was more work to do. But the essential work was done. Again, “What To” program is the majority of the game – “How To” program is the easier part!
The programming done, my own excitement growing, he executes the macro:
January 28, 2010
General Thoughts to Conclude My
KC / StL Rail Corridor Presentation
There are a number of things I'd like to comment on regarding the study, my project, and TOC in general. Many of these thoughts deal with hypotheticals I've raised in all of this. Since it's to be a project included in one of the books, it's OK to speculate, hypothesize, predict, etc.
The "increased train load" as the core problem was really an artificial stopping point. One can seek - and can almost always find - a more fundamental cause for the identified core problem.
What is the criteria for an "effective" stopping point in the identification of the core problem?
The goal of the analysis was "on-time" improvement. This partially moved us towards 100% on-time Amtrak performance, defined as getting 280 miles in about 6 hours. How would the analysis have changed if, instead, the goal was "Increase Mo River Runner ridership from 100,000 / year to 500,000 / year"? That is, increase throughput?
How could (should) the goal have been revised?
What if someone at MoDot had said, "Wait a minute - even if we get ontime performance up, we're not going to increase ridership much. Most people still drive."
Issues (1) and (2) raise a concern about "So What"? "What's the harm?" "There was a problem, and we've made the situation better.
To me, there might be great harm.
If the goal was on-time performance, and the California Siding issue addresses the problem, then the corridor-rail-issue has been solved, for most people! Do you really think this issue will arise anywhere in the next decade? A possible great-opportunity lost?
Consider the following likely dialogue:
MoDOT: "On-time performance is poor. We've hired you to do something about it."
Report: "After extensive analysis, interviews, and studies, we've concluded the core problem is there is increased train volume on the tracks."
MoDOT: "No kidding."
Report: "We've identified several system bottlenecks, and recognized one as having huge potential for solving the problems: add a siding in California.
MoDOT: You mean this study really says, "On-time performance stinks. It's because there are more trains. To address the issue, go where there is only one set of tracks instead of two, and add a siding."
Report: "That's right."
MoDOT employees looking at each other, all saying the same thing: I TOLD YOU!
Is there a "Presentation Cloud" here somewhere?
Something further, the more I think about this: If a solution is really this easy, might this be an indicator it's not really this easy. That is, we're missing something (like an underlying chronic conflict)? Maybe a 'red flag' signal we should obey, pause, and investigate further.
The authors start with TOC as seeking to identify the core problem responsible for the undesirable effects in the system. Once finding "train volume" as the issue, they immediately jump to searching for the bottleneck.
Why was this leap made? For example, if # of trains is an issue, but there are numerous trains carrying similar loads, one solution (good or bad) is to combine the trains. This wasn't suggested, but it seems to address the problem. Another, for example: do all trains need to come through Kansas City and cross the state? Can they be off-loaded / re-routed?
I bring this up because the leap was made to automatically look for bottlenecks, which suggests ...
The authors, in the appendix, define TOC by way of the focusing steps. Constraint management. Idenfitication / Exploitation / Subordination / Elevation / Go Back. But that's not what they did.
Which leads to a thought I tried to address with an interesting cloud in my "Architects" cloud regarding "what to change" --- there seem two paradigms in place: constraint management / core-problem analysis.
Under what circumstances are each applicable?
To be continued ...
(Part 4 of 4)
A QuickStart Publication
January 29, 2010
A Short Story (Part 1)
January 30, 2010
I was alone at the bar when the doors to the tavern opened to a flood of arriving patrons. I recognized them all. They were my co-workers at Cytronic Washing Machines, Inc. They were celebrating.
I knew why.
My friend, Jack, was leading the charge, and he was buying. “Come on, Charles” he bellowed. “What’ll you have?” It was clear this was not the first establishment they’d been to.
“Nothing for me, thanks” I said, and headed for the door.
Sam – our supervisor – grabbed me as he was walking in, and demanded I stay for a drink to celebrate Jack’s award as “Employee of the Month”.
I reached out to shake his hand, and accidentally dropped a paper on the ground. He retrieved it for me and, noticing writing on it, stole a glance. “Here you go! I didn’t know you were interested in fractals. My son is studying these in school right now! See you Monday at work, Charles!” He was in a festive mood. I wasn’t.
“No you won’t. I turned in my resignation to Human Resources this afternoon.”
Sam looked at me, shocked. We had always been friends. He was a good guy. There had been no acrimony in the office. This took him by surprise. “What’s going on? Something at home?”
“No – but it has everything to do with what’s going on right here – and on this piece of paper.”
Sam didn’t understand.
The crowd was hollering from inside the bar for Sam to join in a toast. He looked at me. “Promise me this. I’ll call HR and tell them to put that resignation letter on my desk. Come and see me first thing Monday morning and we’ll talk. Promise me that, Charles.”
“For you, I promise. Have a good time celebrating tonight, and I’ll see you Monday.”
It looked like a fractal, I guess, though that’s not what was printed on the piece of paper.
Let me explain, first off, what we do. We make washing machines. It that sounds boring, it’s only because most people think of a washing machine merely as something you cram clothes into, drop in some soap, and push a button.
There’s much more to it – as there is much more to any technology, particularly if you’re the one making things.
And there’s immense precision to the manufacturing of a washing machine – as there is in most technology. We deal in the domain of thousandths of a millimeter.
Jack’s team creates the machine itself, while my team is charged with door assembly.
Our door attaches on the left side of the machine via a steel plate secured with three screws, and swings open on a hinge. The door stays closes – and opens – because of a clasp on the right side. There is a certain tension necessary to keep the door closed, but not too much tension, else the door will not open.
As you can imagine, with the door secured completely on the left, but only partially on the right, there is a great deal of precision in the drilling of the holes. If any of the holes are off by the slightest, the door won’t open correctly – or it won’t close securely, and water will likely spill. The clasp will break. Lots of things can go wrong.
We’d been working as a finely-tuned process for quite a while – until recently.
Recently, my doors haven’t been fitting exactly as they had. I’ve had to re-drill holes and fix clasps breaking in the testing process. Inventory is stacking up at my work-station. It’s been frustrating.
I asked Jack one day if he was doing anything differently, explaining the issues I was having. Jack assured me he was not only doing nothing differently, he was working extra hard to make sure the washing machines were being built exactly to specifications.
I watched his team work one day, during my break. They were measuring. They were recalibrating. They were busy. On every machine coming down the line. And they seemed to know exactly what they were doing.
Things seemed OK. In fact, the flurry of activity suggested they were going overboard to do things right – and better. However, I knew they weren’t fine, because now my station was having troubles!
I checked all of my equipment. My processes. Nothing had changed – with me. It had to be his station.
I decided to measure his work. I quickly took measurements of the machine center, and noted the coordinates on a separate sheet of paper. I tracked the first 100 machines coming down the conveyor belt, and repeated this process.
Later that night, I entered them into a spreadsheet, and graphed them. The result was the earlier image you saw.
What did it mean?
And why was it happening now? I had a good guess. I thought this “extra work” on Jack’s part was responsible for more work on my part. To make matters worse, he was being rewarded for his performance!
A Weekend Visit
Still wondering about the issue, and now given a reprieve by Sam about my job, I decided to visit the plant over the weekend. The parking lot was empty as I pulled in. I walked through our inventory storeroom. Shelves and shelves of stock, unsold. I remembered working overtime to meet production quotas to get these built “to budget”. I knew how much I had paid my team in overtime and bonuses. And here the machines sat.
Oh well. It’s not my call.
However, looking at the machines, I got to thinking: these were built when everything was working fine. I wonder how these measurements look?
It was just me and the floor and the bloated inventory, so I decided to answer my own question. I measured 100 washing machines, capturing the machine center on a tablet. I went home, input this data, and graphed it:
What is going on here, I thought? I knew our tolerances were small, and in any industrial process, variation is inevitable. But this variation was minimal!
At least it used to be!
Here’s Jack with huge variation now, and he claims he’s doing things better! Worse yet, everyone thinks he is doing better, too! Jack’s getting awards for “Employee of the Month” for this!
I thought back to recent images of Jack, anxiously adjusting his machine as a new body part came down the line. It looked frantic, yet it was obviously intentional.
And it was obviously wrong.
I’ve got enough information for my meeting, which I now eagerly await Monday morning, with Sam.
A Short Story (Part 2)
January 31, 2010
I was paged first-thing Monday morning. If Sam was anything, he was prompt and diligent.
I came to the office and took a chair.
Sam started in. “Look – if it’s about Jack getting ‘Employee of the Month’, don’t worry about it. His station put in more time, and he particularly took it upon himself to improve the quality of his machines.”
“Is that what you think this is about – jealousy?”
“What else could it be? You were a happy employee last month. We implement a policy where Jack wins, and this month you’re quitting. You tell me. If it’s not that, what is it?”
“It’s this.” I tossed the “fractal-image” in front of him.
“What does this have to do with anything? You mentioned that as you left the bar Friday night?”
I opened my brief case and brought forth a small funnel, and a bag of marbles.
“What’s all this?”, he asked.
“A simple experiment. Do me a favor, would you?” We moved to the center of his office, where a beautiful rug covered the floor.
“Drop these marbles through the funnel, slowly, if you would.”
I held the funnel still, and the marbles progressed down the funnel, onto the rug. Often times, they rolled right, others left, and as more marbles collected on the floor, the falling marble would hit one and bounce away.
Nevertheless, after a short bit of time, 30 marbles were on the rug, in the following pattern:
“Most people think of industrial processes as being exacting sciences, which they are, but even with our plant, there’s a bit of variation everywhere. Right?”
“Of course”, Sam replied.
“There’s simple variation in the marbles just dropped, in fact. Right?’
“Right. Obviously.” Sam was growing tired of my demonstrating.
“OK. Let’s suppose I come along and offer you $1,000 to reduce the variation – to ‘make a better product’. What would you do?”
“I’d do whatever I had to do to remove the variation,” said Sam, happily.
“Like what?”, I said gaily. “Show me.”
“OK”, Sam said confidently. “But I’ll drop the marbles myself.. You just make sure you’re holding the funnel tightly so you don’t mess me up. In fact, I’ll hold the funnel and you drop the marbles.”
“Fair enough. You say ‘When’”.
“Wait a minute,” he said. “If I’m in charge of this process, I need a way to figure out which direction to move the funnel.” He went and got a roll of masking tape and created an ‘x’ and ‘y’ axis across the rug.”
“There. Now we have a bulls-eye, and a way to figure out how to move the funnel.”
I yawned. “Does that mean you’re ready?”
“I’m ready!” He held the funnel above the bulls-eye, and I dropped the first marble.
It hit the bulls-eye, but bounded off to the right.
“Shoot.” He eye-balled the distance the marble had rolled from the bulls-eye, and manually moved the funnel the opposite direction from the marble.”
I dropped another one, and this one moved closer to the target.
“Yes,” he exclaimed, and again, moved the target, adjusting the distance the amount opposite the bulls-eye from where the marble landed.
I dropped a third. This one careened away from the bulls-eye, making the distance farther than the previous.
Sam again corrected for the error, moving the funnel farther from the previous spot, but in the opposite direction.
Suddenly the game wasn’t such a fun game. I made him finish all 30 marbles.
We looked at the floor together, silently.
“What happened?” It was Sam, despondent. How could this be?” Why did the first 30 marbles make such a nice, tight pattern, while these are all over the rug?”
I brought out the “fractal picture” he had been making fun of.
He looked at it. “What does it all mean?”
“It means,” I said, “we acknowledged there is variation in everything we do. Extremely small variation, but variation none-the-less. We can control it. But when we react – or overreact – to it, we magnify it. It becomes worse. Everything becomes worse.”
“And are you saying we do this to ourselves?”
“It’s worse than that. Jack was a happy worker doing what he loved to do when you came along with this new policy. Suddenly, he was changing everything. He was figuratively ‘moving the funnel’ at his station, and in doing so, he was increasing the variation! He was, inadvertently, making my job harder – which is why you saw inventory backing up at my station. I suspect our customer-service agents are going to start fielding more calls because of the lower quality machines we’ve just made. And after all that, of your own doing, what did you do? You rewarded him!”
Sam sat on the carpet, not saying a word.
“Now you know why I said I was quitting. We didn’t do this to ourselves. You did it to us! And once a policy like this ‘Employee of the Month’ nonsense creeps into the system, likely more such policies will follow. No sir. Life is too short to live under a death sentence like that!”
“I’m putting a stop to it immediately. Will you stay?”
“Of course!” All he had to do was ask.
We shook hands and I started to leave the room. “Hold it a minute. Something doesn’t make any sense.” Sam was standing back behind his desk.
“You said changing the position of the funnel brought everything out of whack.”
“Do we ever get to move the funnel? How do we ever lower the variation?”
“I thought you’d ask that, so I printed out some other graphics.”
“Let’s suppose we try to graph this data another way, so we can track the quality of the machine, one-by-one. I’ve got ‘x’ and ‘y’ coordinates so one way might be to graph the total distance we’re “off center”.
“Here’s a graph of the first 30 machines above, done prior to the performance incentives.”
“OK”, said Sam.
“Notice the natural variation we saw in the first round of dropping marbles. Let’s suppose we agree ‘that’s just the way things are’, for now. That’s ‘the variation of the system’. We’re trying to find a way, maybe not to reduce the variability, but to see if we can find a way to detect when ‘something’s gone wrong.”
“What do you mean?”
“I knew something was wrong with Jack’s team right away. I could see it in the work coming down the line, and the re-drilling we had to do.”
“So why don’t we create a process to objectively capture this information, rather than me going down and watching Jack.”
“For example …”
“These lines are ‘control limits’, capturing the ‘natural variation’ of the system. As you can see, everything falls within these control limits. Everything is fine.”
“But do we have to be satisfied with this amount of variation?”, said Sam.
“A great question – no. Of course not. This data is not a law of nature. It’s just recognition of the capability of our system right now. But if we want to change it, we have to do more than offer $1,000 rewards! For example, suppose we found a way to clamp down the funnel when we’re dropping marbles. That would probably reduce the variation – rather than simply ‘holding it tight’”.
“OK. So what does the graph look like when you include the data from the next 100 machines? I know you’re itching to show me!”
“Here it is. As you can see, I’ve had to modify the y-axis scale to accommodate the expanded variation!”
“As you can see, had we had a performance scorecard like this in place, even if you didn’t believe me, you’d be asking yourself – and the team – right away: What’s going on, folks!”
“Let’s get Jack and the rest of the team together to talk about all of this,” said Sam, but then he continued. “Hold it. First things first. I’ve got to plan to rework those machines we’ve just built to ‘fix’ them. Can you work overtime tonight, you and your team?”
“Do you have someone ready to buy the washing machines?”, I asked.
“No, but we’ve got to get back on schedule, don’t we?”
“At any price?”, I asked, and knew there was much more work to do.
The Follow-Up Meeting
“Would you really have quit?”, asked Sam over a drink at the local bar, long after hours were over.
“I was ready to, wasn’t I?”
“But why? Didn’t you have confidence I would have done something about this?”
“Here’s the thing with programs like this,” I said. “We were going along fine, making a reasonable profit, producing a quality product. Were we doing as good as we could be doing? Of course, not. Look at all that inventory we’re paying time-and-a-half to produce, just to sit on a shelf.” I couldn’t resist throwing that in.
“And along you came with an ‘Employee of the Month’ program. What happened? The system went to hell pretty quickly. Customer service complaints were likely coming. Product quality was falling. Employee morale – at least mine – went into the tank.”
“And whose job is it to fix all those things?”
Sam responded, hesitantly, in the tone of a prolonged question: “Mine?”
“Indeed – and what would you have come up with? More programs? Sending the system further into a spiral!”
That is, this is the death spiral I saw coming, rapidly:
THE DEATH SPIRAL OF POOR MANAGEMENT