======================================================================== Cybernetics in the 3rd Millennium (C3M) -- Volume 3 Number 8, Sep. 2004 Alan B. Scrivener --- http://www.well.com/~abs --- mailto:abs@well.com ======================================================================== "All I Know About Operations Research I Learned from the Walt Disney Company" "I've always maintained that you can't just coast, if you do, you go backwards. It's just a slow way of liquidating... Let's do anything to get some action, you see?" -- Walt Disney, quoted in an email from SaveDisney.com ( http://savedisney.com ) "Stafford Beer has stressed this essentially cybernetic concept; that industry is an organism... It either evolves or decays." -- Gordon Pask, "Industrial Cybernetics" quoted in "Cybernetics of Cybernetics" (1974) edited by Heinz von Foerster ( http://www.amazon.com/exec/obidos/ASIN/0964704412/hip-20 ) I literally grew up with Disney. Disneyland opened when I was not quite 2 years old. At the time I was living with my family in Florida, which seemed a long way from the Disney empire (!). But I do remember when we got our first TV. (Time may have added some magic to this memory, I confess.) Our dad brought it in and fussed with the antenna for a while, and then managed to tune in the Mickey Mouse Club. Yay! We watched the roll-call, some skits, and then the Mees-ka, Moss-ka, Mouse-cartoon. It was a Donald Duck cartoon about Donald bringing home a TV for his three nephews. After fussing with the antenna for a while he managed to tune in the Mickey Mouse Club. "Wow!" I thought. "This gizmo is really something!" You see, I thought a television showed cartoon characters doing the same things you did. In 1959 my family moved to the San Diego area. Though we would be leaving behind all of our friends and family, my sister and I were sold on the idea because we would be close to Disneyland. We knew about the "Magic Kingdom" from watching the TV show "Disneyland" every Sunday night on ABC. Six months after the move we were finally able to make our first pilgrimage to the "Happiest Place On Earth" (TM). Our grandparents came to town and came along on an overnight trip to Disneyland and Knotts Berry Farm. (Knotts was pretty undeveloped at the time -- the high point in my recollection was a chicken that played the piano when you put a nickel in a slot. Our family never went back to Knotts, but we made annual visits to Disneyland after that.) The "Disneyland" show later became "Walt Disney Presents" and then "Walt Disney's Wonderful World of Color" on NBC, the network with a peacock to show off its new color broadcasts. I still remember the day our dad brought home our first color set. One of the first shows we watched was Walt's. A few years ago the Disney Channel showed a bunch of the old Disney TV shows in the dead of night under the title "Vault Disney." I taped many of them, and ended up posting my tape listings on the web as an off-site backup. Google found them, and I have received more email requests for copies of those tapes than all others in my collection combined. If the Walt Disney Company would ever re-release them I'm sure they'd find an eager audience. ( http://www.well.com/~abs/MediaDoc/VHS/disney_tv.html ) An excellent guide to these shows is in in the book "The Wonderful World of Disney Television: A Complete History" (1997) by Bill Cotter. ( http://www.amazon.com/exec/obidos/ASIN/0786863595/hip-20 ) ( http://www.billcotter.com/tvbook ) He has made the appendix listing all the shows and their air dates available on-line (bless him). ( http://www.billcotter.com/tvbook/appendix-b.htm ) * * * * * * * * I don't remember when I first heard the term "operations research." It was probably when I was in college, and reading all the Whole Earth Catalogs I could find for Stewart Brand's insights on cybernetics, and he made mention of the work of Stafford Beer (more on that later). I first became aware of the existence of "efficiency experts" from watching television. They were usually portrayed as villains wielding stopwatches and forcing people to work fast, or replacing them with robots. For example, the character played by Richard Deacon (later "Mel" on the "Dick Van Dyke Show") in the "Twilight Zone" episode "The Brain Center at Whipple's" (1959). ( http://www.spookytoms.com/Richard_Deacon_2.html ) This episode is now available on DVD in the collection "The Twilight Zone Vol. 40." ( http://www.amazon.com/exec/obidos/ASIN/B000055ZCV/hip-20 ) A more positive portrayal of the "efficiency expert" was in the character of Frank Gilbreth in the movie "Cheaper by the Dozen" (1950) starring Clifton Webb and Myrna Loy, ( http://www.amazon.com/exec/obidos/ASIN/B00013RCAM/hip-20 ) based on the book "Cheaper by the Dozen" (1948) by Frank B. Gilbreth and Ernestine Gilbreth Carey. ( http://www.amazon.com/exec/obidos/ASIN/006008460X/hip-20 ) I never actually read the book, or saw the whole movie, but I gathered that Frank Gilbreth was an efficiency expert who applied industrial engineering principles to the task of raising 12 kids. In researching this e-Zine, I found that Frank's wife Lillian was also an industrial engineer, frequently collaborating with her husband, and she finished the child-raising task alone after he died. There is a biography of her, "Making Time: Lillian Moller Gilbreth, A Life Beyond Cheaper by the Dozen" by Jane Lancaster. ( http://www.amazon.com/exec/obidos/ASIN/1555536123/hip-20 ) There is also a web site, greatwomen.org, with biographical and bibliographical information. ( http://www.greatwomen.org/women.php?action=viewone&id=65 ) Her books include: * "As I Remember: An Autobiography" by Lillian Gilbreth. Institute for Industrial Engineers, 1941. * With Frank Gilbreth, Lillian's husband. "Motion Study" Easton, Pennsylvania: Hive Pub. Co., 1911. * With Frank Gilbreth her husband. "Fatigue Study" Easton, Pennsylvania: Hive Pub. Co., 1917. * With Frank Gilbreth, her husband. "Applied Motion Study" Easton, Pennsylvania: Hive Pub. Co., 1917 Notice the dates on the technical works: 1911 and 1917! This motion study stuff has some historical roots. In fact, it dates back to 1877 and the work of Eadweard Muybridge. A web site called "The 20th Century Museum" by Rick Jones summarizes the story. ( http://www.dover-web.co.uk/20thcentury/1911-hollywood.asp ) Under the entry for "The Birth of Hollywood" it says: Amazingly, it all began with an argument between two friends. They were arguing over whether all four of a horses feet left the ground whilst running. One claimed that the horse always had at least one foot on the ground; the other said that the horse becomes air-borne whilst running. To settle the argument, they each placed a bet and approached Eadweard Muybridge, a professional photographer who was born in England in 1830. In 1852, Muybridge went to the USA to develop his career, and in 1877 the two arguing friends arrived in his studio and commissioned an ingenious set of photographs. To solve the problem, Muybridge set up a line of twelve cameras, each one with a trip-wire attached to it. Then, a horse was made to run past the line of cameras. Each time it passed a camera, it activated the trip-wire and a photograph was taken. The result of this was a series of still photographs of the horse in motion, literally the world's first "motion pictures". Muybridge developed his work further and developed the "zoopraxiscope", a precursor of cinematography, in which he printed the images of the running horse onto a rotating glass disk, and he spent the rest of his career studying human and animal movement with his pioneering invention. Of course the "zoopraxiscope" was the forerunner to the modern movie projector. ( http://www.kingston.ac.uk/Muybridge/img0014.jpg ) (By the way, he proved horses do at times lift all four feet off the ground.) An entertaining and informative dramatization of the use of the zoopraxiscope is found in an episode of the old "Disneyland" TV show entitled "The Story of the Animated Drawing." It has been re-released on DVD in the collection "Walt Disney Treasures - Behind the Scenes at the Walt Disney Studio" (2002). ( http://www.amazon.com/exec/obidos/ASIN/B00006II6P/hip-20 ) The write-up on Amazon includes: "The Story of the Animated Drawing" traces the history of the medium, including re-creations of Emil Reynaud's Theatre Optique (1892-1900) and Windsor McCay's vaudeville routine with his landmark film Gertie the Dinosaur (1914). This is the first of many overlaps between the life's work of Walt Disney and the development of Operations Research. * * * * * * * * "Walt Disney -- inventor of clan totem symbols ... hardware architect of the new realities about to be neurologized." -- Timothy Leary, 1979 "The Game of Life (Future history series)" ( http://www.amazon.com/exec/obidos/ASIN/0915238306/hip-20 ) I was about nine years old when the magic of Disneyland began to stop working on me. I began to see how the illusions were done. This ushered in a new era for me, in which my fascination with Disneyland shifted from the mythological to the technological. About this time I was also devouring books on stage magic, histories of magicians like Robert-Houdin, and various optical illusions. By the time I was 12 I chuckled when people claimed there were holograms in the Haunted Mansion ride. I knew most of the illusions were done with mirrors (partially silvered) using a technique dating from the 18th century called "Pepper's Ghost." ( http://www.dafe.org/misc/peppers/peppers.htm ) I was fascinated by the technology of Audio-Animatronics -- even the name (and the name "Imagineer" for the folks who made it work). The first and most primitive achievement in this new field was "walt Disney's Enchanted Tiki Room tech" in which a room full of robotic birds moved with compressed air were controlled by signals on a 14-track magnetic tape one inch wide, borrowing technology from the aerospace industry. ( http://members.tripod.com/labyrnth/animatronics/history/animatronics101-102/aa.html ) Some of you will recall that in C3M Volume 2 Number 11, Nov. 2003, "War Games, Money Games, New Games and Meta Games" I wrote: In 1976 I worked for ten weeks as an employee at Walt Disney World; I like to say I took the job to get the employee training program, which is not available any other way. Also, on the day after the stockmarket crash of 1987 I bought one share of Disney stock, which has since split twice to become four shares. In this case I like to say I bought the stock to get a lifetime subscription to the annual report. The employee training program at WDW included a tour of the Digital Animation Control System (DACS), a single minicomputer that controlled all of the rides and animatrons at once. We reached it by passing through the "steam tunnels" underneath the Magic Kingdom. Later, when EPCOT opened, a new DACS was built in the "CommuniCore" complex and visitors were able to see a show which explained some of the computers' functions using the ubiquitous "Pepper's Ghost" illusion to make translucent characters appear to dance between the machinery. ( http://home.cfl.rr.com/omniluxe/cmcr.htm ) While working at "the world" (as employees called it) I had an epiphany: you can never get to be Walt by working your way up from busboy (my job) at Walt Disney World. I realized this after talking to one of the portrait artists, who showed me his secret portfolio of Disney character drawings (such as Donald and Snow White) after making me promise not to tell anyone what I'd seen. He explained that he could be fired and sued if he was caught making unauthorized character drawings, even for his own education. I created a mental shorthand for this realization: "Where Bambi goes nothing grows." (This is a line from an old comedy routine, "Ajax Liquor Store" by LA radio personalities Hudson and Landry, from their 1970s album "Hanging In There." It was rereleased in the 1996 collection "The Best of Hudson and Landry.") ( http://www.amazon.com/exec/obidos/ASIN/B0000004QD/hip-20 ) Long after I left the employ of the mouse I continued to research the company, absorbing every annual report they sent me and also buying and devouring virtually every book about Walt and his empire. One unauthorized book was "Window on Main Street: 35 Years of Creating Happiness at Disneyland Park" (1991) by Van Arsdale France, who ran Disney's employee training program for many years. ( http://www.amazon.com/exec/obidos/ASIN/0941613178/hip-20 ) He describes joining the Disneyland team while it was still under construction in 1954, working for a former aerospace manager from Consolidated Vultee Aircraft (later Convair -- where my dad worked!) by the name of C. V. Wood, who everyone called "Wood" or "Woody." Walt hired Woody away from the Stanford Research Institute (SRI), where he worked on the study that recommended that Disneyland be built in Anaheim near the unfinished I-5 freeway, where the center of population for the LA basin was projected to move. Van France describes another early Disneyland manager thusly: Despite the confusion, Wood took the time to introduce me to Fred Schumacher, another pioneer who would guide me through the mysteries of my new life... He was one of Wood's aircraft associates and an industrial engineer by profession... Fred always had an industrial engineer's most valued implement, a stopwatch, in his desk drawer. It reminded him he could always get a job if he quit or got fired. Inspired by this passage, I bought an old-style 2-button analog stopwatch on eBay. (As I recall it was an overstocked Ukrainian knock-off of an American design.) In my last dot-com job the manager of Quality Assurance (QA) was always borrowing it to time user interface components. Another amazing and absorbing book I read was "Designing Disney's Theme Parks: The Architecture of Reassurance" (1997) edited by Karal Ann Marling. ( http://www.amazon.com/exec/obidos/ASIN/2080136399/hip-20 ) It was based on an art show at the Canadian Centre for Architecture in Montreal, culled from the archives of Imagineering concept and design art in the Disney Art Library. The curators were given unrestricted access to the material to use however they saw fit, and the resulting show and book offered a deep and multifaceted look at "theme park art" and its context in Western Civilization. From this book I learned that the architecture at Disney theme parks, though it may look like castles, chateaus, log cabins, spaceports and tiki huts, is fundamentally based on aircraft hangars. Like the dozens of aerospace facilities that sprang up around the Los Angeles basin during and after World War II, Disneyland and its sequels are made of large, strong steel superstructures originally designed for building and fixing airplanes. Ironically, Disney's newest theme park in Anaheim -- California Adventure -- has a "land" called "Condor Flats" inspired by the high desert flight test facilities (such as Edwards Air Force Base near Mojave, California) with a restaurant called the "Taste Pilots' Grill" built to look like a hangar. ( http://www.dcacentral.com/dca/gallery/large/construction/sept99_const_003l.jpg ) But of course it really is a hangar, making it perhaps the most honest building in a Disney park. ( http://www.intercotwest.com/gallery/photos/large/09000/200/9237.jpg ) In 1993 a new land called "Toontown" opened at Disneyland. Shortly afterwards the Imagineers opened their facility in Glendale to the public for what was probably the first and last time ever, for a meeting of the local Los Angeles chapter of ACM SIGGRAPH (Special Interest Group on GRAPHics). ( http://la.siggraph.org ) I was there. I actually just went digging through boxes in my garage and managed to find my notes from that meeting. But the one piece of information I was after, the exact date, was missing from a corner chewed off by a rat. How symbolic. (This is why I switched from cardboard bankers boxes to plastic sealed crates a few years ago.) Still, the notes triggered some useful memories. One of the speakers told us that when Walt wanted engineers with experience with "man-rated" systems for ride design, he recruited them from aerospace, where the designers of military aircraft had just such experience. The Imagineers showed us slides of the construction of Toontown at various stages, which was very interesting. But the biggest impression made on me was a "before" picture of the empty lot just after it was graded by bulldozers and before any construction began. It looked like a plot of muddy dirt. To this day when I see a plot of muddy dirt I wonder what fantasic environment could spring up there. In addition to readings and meetings, I have also learned about the Disney company's ways through direct observation. The Imagineers have become so expert at crowd control that they make it look effortless. For example, the Hall of Presidents attraction at Walt Disney World's Magic Kingdom seats 700 people, and they all exit at once at the end of each 23 minute show. This could create a huge lump of people, but the wide exit corridors split into three areas: a gift shop (the "Heritage House"), a snack stand ("Sleepy Hollow") and an route which opens out to to Liberty Square, with several vendors and carts nearby, including the open-air Square Portrait Gallery, Silhouette Cart and the Umbrella Cart. I have sat and watched with fascination (well, with interest anyway) as the crowds melted into these three areas and dispersed quite naturally, with nobody getting separated from their party and nobody seeming crushed or rushed. In contrast, I have been present at two separate theme park "meltdowns" when crowd control mechanisms failed dangerously. Once it was at Universal Studios Hollywood and once it was at Knotts Berry Farm. Both parks suffer from a current layout that resulted from poorly-planned growth and ad hoc additions. In both cases people were pouring into an area with insufficient exits, and then unable to get back out even as more people poured in. In each case I felt lives were endangered by the situation; luckily the crowds stayed calm, but any kind of panic would've resulted in a stampede that could have trampled people. Certainly the most ambitious logistical undertaking of Walt Disney's was one that was never actually undertaken. A few weeks before his death he made a promotional film, never intended to be shown to the public (it was for politicians and real estate developers to see) called "the EPCOT film," in which he laid out his plans for an Experimental Prototype Community (or City) of Tomorrow. I'm not talking about the theme park -- sort of a permanent world's fair -- that was ultimately built, but his original dream of a city of the future. Fan sites now document the contents of this film, ( http://www.waltopia.com/florida_film.html ) ( http://perso.wanadoo.fr/sebastien.barthe/lastfilm.htm ) and it has been released in the DVD collection "Walt Disney Treasures - Tomorrowland: Disney in Space and Beyond" (2003). ( http://www.amazon.com/exec/obidos/ASIN/B0000BWVAI/hip-20 ) Though Walt's future city was never built (some argue that the Disney-made town of Celebration has taken its place) he -- and those who came after him -- did put into place the government infrastructure to accomplish that dream, in the form of the Reedy Creek Improvement District (RCID). If you don't know what that is, I suggest you rush out and read "Married to the Mouse: Walt Disney World and Orlando" (2001) by Richard E. Foglesong at your earliest convenience. ( http://www.amazon.com/exec/obidos/ASIN/0300098286/hip-20 ) If you're not that curious, you can read the review at Amazon.com,` (copyright 2001 by Cahners Business Information, Inc.) which includes this stunning revelation: Disney World, in its agreement with the city of Orlando and the state of Florida, actually negotiated the right to construct and use a nuclear power plant at the amusement park. True, it has never built one, but according to this well-researched, cogently argued and eye-opening account of the complicated relationship between the Disney Company and the city of Orlando, it's a sign of the high price that Orlando has paid to become the home of "the most popular tourist destination in the world." A privately held corporation, Disney has created what amounts to an independently governed country "a sort of Vatican with mouse ears" within Orlando, says Foglesong, professor of politics at Rollins College. And/or you can read the heavily hyperlinked article at thefreedictionary.com, ( http://encyclopedia.thefreedictionary.com/Reedy%20Creek%20Improvement%20District ) which includes this explanation: The Reedy Creek Improvement District is a public corporation of the state of Florida created in 1967 by an act of the Florida legislature. Its primary purpose is to allow The Walt Disney Company [to] exercise powers normally reserved for municipal governments (including taxing and zoning authority) over the land that constitutes Walt Disney World. The district has its seat in Lake Buena Vista and its executive organs (serving primarily the WDW parks and resorts) include a building inspectorate, a fire brigade with an ambulance service, a planning division, and a comprehensive utilities service covering trash collection, energy, and all in-ground systems except telephony and broadband. In his book, "Married to the Mouse: Walt Disney World and Orlando," Richard Foglesong argues that the Florida legislature created the RCID under the belief that Disney would create an actual community in the area as part of its EPCOT project. However, once the Disney Company was granted governmental powers, it did not follow through on its promise of permanent residents in order to maintain control over the RCID. Managing their own government is certainly a big logistical challenge for Disney, but nothing compared to what managing their own city would've been. * * * * * * * * ...quite paradoxically, "Cybernetics" has been tainted from both sides by seeming at once too "establishment" and too "counter-culture." Its roots in "Operations Research" reminds me of the old saying, "You can kill more men with a pencil in Whitehall than with a rifle in Flanders," or whatever it is. It was probably the application of the principles of Operations Research in World War Two that lead to the allied fire-bombing of the civilian city of Dresden, to draw German fighters away from military targets. Later, defense secretary Robert McNamara under Lyndon Johnson was a big fan of a systems approach to waging the Viet Nam War. His attempts to maximize enemy body count lead to soldiers digging up graves and shooting water buffalos and putting it all in body bags to count as dead Viet Cong. -- "What Ever Happened to Cybernetics?" Cybernetics in the 3rd Millennium (C3M) Volume 2 Number 4, Apr. 2003 Though "Industrial Engineering" began in the 19th century, the field of "Operations Research" (OR) was an outgrowth of World War II. Scientists in England and the US were asked to help in the war effort. One obvious example of this was the Manhattan Project which developed the atomic bomb. My mentor, Gregory Bateson, was an anthropologist, and so was put to work writing propaganda leaflets to drop on Japanese-occupied islands in the Pacific. But in addition to these ad hoc projects, there was a systematic attempt to apply scientific method and quantitative analysis to various logistical problems of war. ( http://www.mors.org/history/history.htm ) The story I have heard most often is that a group of scientists in England analyzed the problem of how closely clustered bomber groups should fly. If they were too far apart they suffered more damage from flack, but if they were too close together they collided too often. The scientists found that they should fly closer together. Pilots had been resisting this for psychological reasons -- they felt worse about casualties they'd caused themselves due to bad piloting -- but the tighter clusters reduced overall losses. There seems to be a paradox in this story. On the one hand the scientists accomplished their goal: reducing casualties. But on the other hand they also had the unintended side-effect of reducing morale. Hmmm. After the war a number of people took the principles of OR developed for military use and adapted them to businesses, especially large businesses with major logistical challenges. A major contributor to this effort was Stafford Beer, who made a career of writing books, speaking and consulting on OR, which later came to be called "Management Science." What kind of problems does OR solve? An on-line curriculum of OR ( http://www.brunel.ac.uk/depts/ma/research/jeb/or/contents.html ) lists topics which include: Deterministic topics -------------------- * Data envelopment analysis * Facility layout * Facility location * Graph theory * Integer programming o Heuristics o Integer programming formulation examples o LP relaxation o Structured problems * Inventory control * Just-in-time * Linear programming o formulation o solution o sensitivity analysis * Master production schedule * Materials requirements planning * Network analysis - activity on arc * Network analysis - activity on node * Network analysis - linear programming * Network flow * Nonlinear programming * Soft OR * Vehicle routing Stochastic topics ----------------- * Decision trees * Forecasting * Markov processes * Quality control * Queueing theory * Simulation * Stochastic programming When I actually worked in aerospace, at Rockwell International (which I discussed at length last month), I would go down to the well-stocked technical library and browse, and I noticed there was about 20 feet of shelf space just devoted to books on "linear programming." I had no idea what that was, but I was pretty sure it didn't have to do directly with computers. (I was right about that.) Finally, in spring of 2001, I decided it was time to learn what linear programming was. I got a few books from the library and educated myself. It's a mathematical technique, one of the first to be applied in OR, in the class of what are called "optimization problems." Here is an example problem, paraphrased from an on-line tutorial on linear programming. ( http://fisher.osu.edu/~croxton_4/tutorial ) The problem statement is as follows: We are producing three products, p1, p2, and p3, and they sell for $2, $3, and $8 respectively. Our goal is to maximize our revenues. Each product has a per unit production cost of $1, $2, and $5, respectively, and we have a budget of $300. We have demand for the products that requires that we produce a combination of at least 50 units of p1 and p2. Also we have exactly 400 hours of available production time and each unit requires 2, 4, and 5 hours of production time, respectively. Also, as common sense would dictate, the numbers produced of p1, p2 and p3 must all be greater than or equal to zero. The way this problem is formulated, all of the equations are linear, i.e., do not contain powers, trig functions, or anything more complicated that sums of products, such as ax + by + cz. In this simple case the optimal solution is: p1=100, p2=0, and p3=40 which has a value of $520 This problem could be solved with just a little algebraic manipulation, but in general these types of problems -- usually with very large numbers of variables -- are solved with an algorithm called the simplex method. This method uses a few interesting facts about linear programming problems: 1) The constraints limit the solution to a "region of feasibility." In the 2D case (having only two variables, oversimplified for example's sake) this region is either a rectangle, or a rectangle with the corners cut off, which leaves new corners that may also be cut off, and so on. Constraints like "p1 + p2 >= 50" in the example above are what cut the corners off. ( http://fisher.osu.edu/~croxton_4/tutorial/graphics/graph1-5.gif ) In the 3D case (with three variables) the region of feasibility is like a rectangular solid -- a hunk of cheese if you will -- with its corners cut off with a straight slice, which leaves new corners that may also be cut off, and so on. ( http://fisher.osu.edu/~croxton_4/tutorial/graphics/graph-3d-3.gif ) 2) The solution is always at one of the corners, not in the interior or along an edge or other boundary of the region of feasibility. (This may seem amazing at first thought, but since the problem is linear it just means that all the "knobs" must be set at their highest or lowest possible settings to achieve the optimum which maximizes profit.) 3) The problem is always jiggered so that the origin (all variables equal zero) is one of the corners, a valid (but probably not optimum) solution. Another seemingly amazing fact about these problems is that you can find the solution by starting on any corner (the origin is used since it's always a solution) and then move to the adjacent corner which increases profit the most, recursively doing this over and over until you reach a corner whose neighbors all yield less profit -- that's the solution! The simplex method is a "hill climbing" algorithm, which works like climbing a hill by always going up in the steepest direction until you can't go up any more -- you must be at the top! The simplex method is one of a family of algorithms called optimal solvers. Since its development solvers have been devised for non-linear problems, and a whole software industry has emerged selling (usually high-priced) solvers to large institutions. By an interesting coincidence, a month after I researched linear programming I got a consulting gig in which I was able to use the knowledge: I produced a video on the new machine learning technique knows as "Support Vector Machines" (SVM), ( http://www.well.com/~abs/Biowulf/biowulf_screenplay.html ) and it turned out the advanced algorithm was based on optimum solvers, including linear programming. The mathematicians I worked with usually used the built-in solver in the popular MatLab program for scientists and engineers, from MathWorks. ( http://www.mathworks.com ) You can find more information about OR, linear programming and optimum solvers at the web site "The Mathematical Atlas" under the listing for "Operations research, mathematical programming." ( http://www.math.niu.edu/~rusin/known-math/index/90-XX.html ) * * * * * * * * "Remember, it all started with a mouse" -- Walt Disney ( http://www.wordiq.com/definition/Walt_Disney ) "MOUSE MUNCHES MIRARMAX" -- headline in "Daily Variety" May 1993 ( http://www.variety.com ) Most of my life I've had dreams about Disneyland. But it hasn't been the actual theme park in Anaheim. My "dream Disneyland" is actually a blend of Disney's magic Kingdom and the neighborhood where I grew up in La Mesa, CA. The main street is a kind of "morph" of my old street, Meadow Crest Drive, ( http://www.well.com/~abs/Cyb/4.669211660910299067185320382047/MeadowCrest.jpg ) and the actual Main Street in Disneyland. ( http://www.well.com/~abs/Cyb/4.669211660910299067185320382047/MainStreet.jpg ) One of the features is a "Tower of Missed Rendezvous" where you go to meet your friends and they don't show up. I must tell you I had a spine-shivering "deja vu" a few years ago when I saw a picture of the old "Tower of the Four Winds" which Disney erected at the 1964 World's Fair at the entrance to the "Small World" ride. ( http://naid.sppsr.ucla.edu/ny64fair/map-docs/pepsicola.htm ) (The ride was later moved to Disneyland but the tower was not.) It was identical to my dream tower! I realized I must have seen it before as a kid on the old "Disneyland" TV show, in the episode "Disneyland Goes to the World's Fair" (air date 5/17/64). So this was NOT a coincidence, but a buried memory which showed up in my dreams. In the mid-1990s I was working for Steve Jobs' computer company "NeXT" and the Walt Disney Company was one of our customers. I got finally go "on the lot" and see the old "Dopey Drive" and "Mickey Avenue" from Walt's days, ( http://www.mouseplanet.com/sue/wdsstrsg.jpg ) as well as the new administration building designed by Michael Graves for Michael Eisner to work in. ( http://www.mouseplanet.com/sue/wdstdb.jpg ) (See Sue Kruse's Disneyland Diary at Mouse Planet for a tour of the studio.) ( http://www.mouseplanet.com/sue/nov2000.htm ) I found that these visits to the company's headquarters had an effect on my dreams. Instead of dreaming of Disneyland, I began instead to dream of a place I knew to be "meta-Disneyland," the place where Disneylands come from. It was an infinitely-high skyscraper, sort of like a downtown Marriott Hotel from the 1980s in the "brutalist" style just before post-modernism became vogue, and each floor was a theme park. A mathematician might call it "the set of all possible theme parks." In one dream I took my family to a Kenya-themed park on the zillionth floor. Outside of my dreams, visiting "the lot" had other perks. One day a customer took me to the employee store, and let me use his employee discount to buy a book I'd wanted for a long time, "The Illusion of Life: Disney Animation" (1981) by Ollie Johnston and Frank Thomas. ( http://www.amazon.com/exec/obidos/ASIN/0786860707/hip-20 ) I got this $60 treasure for about $40. This textbook for animators, by two of the "Nine Old Men" of classic Disney animation, had several interesting revelations: * Some fine artists became Disney animators just to use the color palette of projected animation, which has more different visible colors than any other visual medium. * Walt didn't pay his animators much, but he did let them shoot as much expensive film as they wanted of animation "pencil tests." The only catch was that everybody in the animation department got to watch the results, including Walt, in the un-air-conditioned screening room they called the "sweat box." He felt that only by seeing how their drawings looked moving on the big screen could his animators learn. * Typically the first assignment a new animator was given was to draw a bouncing ball. It sounded deceptively simple, but turned out to harder than they thought. To look "real" not only does the ball have to squish when it hits the floor, it has to anticipate the collision squishing before it hits! Animation turns out to be about character, and even a ball needs a personality to be successfully animated. Well, this was all fascinating to me, but before I found this gem I actually ended up buying the wrong book! In some little bookstore in my travels I found a slim paperback called "Illusion of Life: Essays on Animation" (1991) by Alan Cholodenko, et. al. ( http://www.amazon.com/exec/obidos/ASIN/0909952183/hip-20 ) I began reading it and it didn't take me long to realize it wasn't the book I'd been seeking (recommended by a computer animator at Pacific Data Images speaking at an LA SIGGRAPH meeting) but instead was a set of papers presented at a conference in Australia on the "scholarly" implications of animated cartoons. Except for a hilarious talk by Chuck Jones (of Bugs Bunny & Roadrunner fame) it was all very deconstructionist and convoluted, and closed with an incomprehensible essay on Borges which seemed to have nothing to do with the topic at hand. But it began with an essay that pointed out that the dictionary definitions of "animate" say it means both "to give motion to" (like a puppet) and "to give life to" (like a zombie). It went on to point out: In terms of scholarship, animation is the least theorized area of film. In neglecting animation, film theorists -- when they have thought about it at all --- have regarded animation as either the 'step-child' of cinema or as not belonging to cinema at all, belonging rather to the graphic arts... If one may think of animation as a form of film, its neglect would be both extraordinary and predictable. It would be extraordinary insofar as a claim can be made that animation film not only preceded the advent of cinema but engendered it; that the development of all those nineteenth century technologies -- optical toys, studies in persistence of vision, the projector, the celluloid strip, etc. --- BUT FOR PHOTOGRAPHY was to result in their combination/synthesizing in the animatic apparatus of Emile Reynaud's Theatre Optique of 1892; that, inverting the conventional wisdom, cinema might then be thought of as animation's 'step-child'. Aha! Of course! Animation is a SUPERSET of cinema! This was so in the very beginning, and it is becoming even more so in the world of computer effects, with George Lucas striving to make "Star Wars" episode III shot on high-def video and digitally projected, so that it is untouched by film. In the animated/live action hybrid movie "The Adventures of Rocky & Bullwinkle" (2000), ( http://www.amazon.com/exec/obidos/ASIN/B00003CXJA/hip-20 ) our heroes from Frostbite Falls are given a ray gun that only destroys cartoons. At one point Rocky uses it to disintegrate what appears to be a live-action helicopter, explaining afterwards that it was a "CGI element" (Computer Generated Image). The first biography of Walt Disney I ever read, "The Disney Version: The Life, Times, Art and Commerce of Walt Disney" (1968) by Richard Schickel, ( http://www.amazon.com/exec/obidos/ASIN/1566631580/hip-20 ) explained something I really already understood, that the entire economic engine of the Walt Disney empire was powered by the feature length cartoons. They originated the lovable characters that created demand for the TV shows, plush toys and theme parks. Without them, the whole enterprise fails. This is why it seems so ironic to me that Disney CEO Michael Eisner has announced the end of 2D hand-drawn cartoons at Disney. Perhaps this is a repetition of the cycle in which the last management team was booted out at Disney. Company yes-man Card Walker and Disney son-in-law Ron Miller announced the effective end of 2D hand-drawn animation in the early 1980s, and Roy Disney and Stanley Gold forced them out and brought in Eisner in 1984. This story is told in "Storming the Magic Kingdom: How Corporate Raiders Forced a Revolution at Disney" (1987) by John Taylor. ( http://www.amazon.com/exec/obidos/ASIN/5551880427/hip-20 ) Now Roy and Stan have a website devoted to throwing out Eisner. ( http://savedisney.com ) * * * * * * * * "There is a certain charm to seeing someone happily advocate a triangular wheel because it has one less bump per revolution than a square wheel does." -- Chuck Swiger In late spring of this year I again wanted to learn more about Operations Research. I discovered a couple of years ago that OR is taught locally (to me) at the University of California at San Diego (UCSD) as part of the Management Science major offered by the Economics department. According to the on-line catalog, ( http://www.econ.ucsd.edu/ugradprog/majdiff.htm ) a Bachelor of Science in Management Science has these characteristics: * applied economics with a micro/management focus (no macro) * covers some functional fields of business * highly quantitative and can be used in these careers: Business, finance, government, law, public policy, etc. The course listing ( http://www.ucsd.edu/catalog/EconCour.html ) describes the three-semester class thusly: 172A-B-C. Introduction to Operations Research (4-4-4) Linear, nonlinear, and integer programming. Elements of game theory. Deterministic and stochastic dynamic programming. Lacking the time or money to enroll at UCSD just to take this class, I made a visit to the campus bookstore and discovered the text used is "Introduction to Operations Research" (1967-2001) by Hillier & Lieberman. ( http://www.amazon.com/exec/obidos/ASIN/0071181636/hip-20 ) It cost $150. I looked on the web and found a copy for $75, which I bought. I've been slowly working my way through it. Some of the things I've learned: * The main emphasis is still on linear programming, only now there are lots more ways to reformulate a problem into linear form. Tales abound of eager young graduates saving big corporations millions by applying OR techniques to resource scheduling problems. (I couldn't help but think about the havoc this could cause if misused.) * Microsoft's ubiquitous Excel spreadsheet actually contains a linear programming solver, which I never knew. You can solve the example problem I gave above by formulating it correctly in Excel and then selecting "Solver" in the Tools menu. ( http://www.economics.ltsn.ac.uk/cheer/ch9_3/ch9_3p07.htm ) There are also more robust solvers that can be purchased and added to Excel to solve more complex problems. * In actual practice, to use linear programming on an ongoing basis in a corporate environment, the optimizing solver has to be integrated with the Management Information System (MIS) infrastructure, to avoid having to constantly re-key data. This turns out to be more of a headache than the math part. * There is an increased emphasis on network theory, especially as applied to transportation and communication routing problems. By an interesting coincidence, a month after I began this study I got a consulting gig in which I was able to use the knowledge: I did some research involving network theory as applied to finding terrorists; this story is told in C3M Volume 3 Number 5, Jun. 2004, "Six Degrees of Buddy Hackett" (see the archives below). As a mental exercise, I have thought about applying linear programming to theme park operations. It is quite easy to compute the costs of running various rides and attractions in terms of personnel and other resources. What is extremely difficult is to compute the effect on the bottom line of NOT running them. It is easy to fall into thinking that X number of people are going to show up every day and pay their $50 plus each whether or not a specific ride is running. But applying this thinking across the board leads to the following strategy: * close all the rides and send all the ride operators home * close all the shows and send the performers and support staff home * take out the drinking fountains and raise the prices of soft drinks This may sound absurd, and yet it is not far from the behavior of the managers of Disneyland, especially under the leadership of the now departed Paul Pressler and Cynthia Harriss at Team Disney Anaheim (TDA). ( http://www.mouseplanet.com/shame ) ( http://www.mouseplanet.com/david/dk031204.htm ) The SaveDisney.com web site has a before and after photo-essay on Disneyland's Tomorrowland that shows how badly that land has been allowed to atrophy. ( http://savedisney.com/news/disneyland/tl/tl_ss061104.1.asp ) It also makes the case that Tomorrowland is a metaphor for the future and long-term vision of the Walt Disney Company. There is now a book which talks about the application of OR techniques to theme parks, "Walt's Revolution!: By the Numbers" (2003) by Harrison "Buzz" Price. ( http://www.amazon.com/exec/obidos/ASIN/1893951065/hip-20 ) Like C. V. "Woody" Wood, "Buzz" worked at Stanford Research Institute (SRI) on the original siting study for Disneyland. (And as I write these words it occurs to me for the first time that the names of the two main characters in "Toy Story" may have come from these two men.) ( http://www.amazon.com/exec/obidos/ASIN/B000059XUT/hip-20 ) Buzz's book includes an analysis of the relationship between capital investment and gate receipts in theme parks, in which he finds no obvious correlation. A form of this article was on the web for a while, and I snagged it. ( http://www.well.com/~abs/Cyb/4.669211660910299067185320382047/Brave5.pdf ) It has occurred to me that it might be very interesting to add every known variable (instead of just the two used) and turn a Support Vector Machine loose on the data to find the relevant factors in boosting gate receipts. * * * * * * * * "If Disney loses Miramax all they'll have left is a muppet and a water park." -- Robin Williams at the 2004 Oscars It seems like every day brings new revelations on the downfall of Michael Eisner. ( http://www.savedisney.com/red_message.asp ) The situation has been brewing a long time, starting about the time of Frank Wells' death, and continuing through the Jeffrey Katzenberg and Michael Ovitz fiascos, all of which is documented in "Keys to the Kingdom: The Rise of Michael Eisner and the Fall of Everybody Else" (2001) by Kim Masters. ( http://www.amazon.com/exec/obidos/ASIN/0066621097/hip-20 ) My personal favorite choice for the job is Apple cofounder and Pixar CEO Steve Jobs. The problem with bean counters is they don't know where beans come from. Jobs knows. Unlike Michael Eisner, Walt Disney created new media: sound cartoons ("Steamboat Willy"), color cartoons ("Flowers and Trees"), feature-length cartoons ("Snow White"), abstract animation ("Fantasia"), theme parks ("Disneyland") and Audio- Animatronics ("Walt Disney's Enchanted Tiki Room"). True, Walt didn't do it alone; many of "his" innovations came from an unsung genius named Ub Iwerks: Mickey Mouse ("Plane Crazy"), the multiplane camera ("The Old Mill"), Circlevision 360 ("America the Beautiful"), the sodium prism matte ("The Parent Trap"), and the projection on sculpture technique "(The Haunted Mansion"). His story is told "The Hand Behind the Mouse: An Intimate Biography of Ub Iwerks" (2001) by John Kenworthy. ( http://www.amazon.com/exec/obidos/ASIN/0786853204/hip-20 ) But Walt gets credit for this too -- he could attract and retain creative talent, like Jobs, and unlike Eisner. Jobs and his teams have also created new media: the color PC ("Apple II"), the Windows/Icons/Mouse/Pointer (WIMP) computer interface ("Macintosh"), the designer computer ("iMac"), the portable digital music player ("iPod"), and the 3D computer-generated cartoon ("Red's Dream"). For those purists among you, yes, I know about Doug Engelbart at SRI inventing the mouse, and the Rio portable digital music player, and "Breaking the Ice" from Symbolics. Walt didn't invent Technicolor either. But Disney and Jobs each deserve credit for creating new markets for new media, and what is a medium without a market? For the Walt Disney Company to become vital again it must get back into the business of creating new markets for new media. * * * * * * * * "I don't make pictures just to make money. I make money to make more pictures." -- Walt Disney Since the early days of Stafford Beer and Gordon Pask, there has been a marriage of Cybernetics and Management. Managers are among the true generalists in our society, needing a generalists education, now more than ever in this high-tech millennium. Many of my readers have told me they are interested in this. One of my subscribers recommended "The E-Myth Revisited: Why Most Small Businesses Don't Work and What to Do About It" (1995) by Michael E. Gerber. ( http://www.amazon.com/exec/obidos/ASIN/0887307280/hip-20 ) (If this was you, I lost your contact info. Please email me again!) Another very good book I have found on systems theory for managers is "Dealing With Complexity: An Introduction to the Theory and Application of Systems Science" (1988) by Robert L. Flood & Ewart R. Carson. ( http://www.amazon.com/exec/obidos/ASIN/030644299X/hip-20 ) Working on this issue of the e-Zine has given me the idea to work up -- at some point -- "A New Curriculum for Operations Research." I'm not sure what all I'd put in it, but I do know it would have more anecdotes from real business problems, more chaos theory, and especially more examples from the Walt Disney Company. ======================================================================== newsletter archives: http://www.well.com/~abs/Cyb/4.669211660910299067185320382047 ======================================================================== Privacy Promise: Your email address will never be sold or given to others. You will receive only the e-Zine C3M unless you opt-in to receive occasional commercial offers directly from me, Alan Scrivener, by sending email to abs@well.com with the subject line "opt in" -- you can always opt out again with the subject line "opt out" -- by default you are opted out. To cancel the e-Zine entirely send the subject line "unsubscribe" to me. I receive a commission on everything you purchase during your session with Amazon.com after following one of my links, which helps to support my research. ======================================================================== Copyright 2004 by Alan B. Scrivener