James L Adams—People and Products

Big System Failures

I was writing a post yesterday about effects of the internet on Christmas, when the electrical power went off in my home. It came back in a couple of hours, which is better than usual. But my portable was uncharged, and I couldn’t get back to my post until today.  Our electrical service seems to drop out several times in a year, usually in the rainy season (tree limbs fall on the wires, equipment that doesn’t like water gets wet, construction in area, etc.).  But we are going through a long dry spell, and have not had rain or strong winds in weeks—business as usual.

 I have now rebooted computers, changed all (most) of our clocks, charged batteries and am back in business.  But I spent yesterday with an old friend who is a user of large computer systems, and we got into a discussion of system reliability, so lets think about that, rather than the internet and Christmas, since you all have just experienced that interaction.

 Things tend to fail when unprecedented and complicated, just after being manufactured (sometimes called “infant mortality”), and after long usage (sometimes called “wearing out”,)  The system supplying our house with electricity is certainly complicated.  But I would not call its components unprecedented or even that exotic, and I trust that they are manufactured with care, and perhaps if highly critical even “run in” a bit before installation, because repairing failures results in real costs to the utility and annoys its customers.  But yet it fails.

 Big systems have failures. There are many reasons for this, not all predictable. There are seeming exceptions to this, such as the two Voyager Spacecraft launched in the 1970’s and still operating now over 15 billion kilometers from the sun. But the amount of care in design, manufacturing, and repetitive testing involved in the Voyager project, combined with a design that focused on simplicity and ruggedness, plus the predictability and benign nature of space (once you get past earth orbit) are not typical in the production of terrestrial devices or the environment in which they operate. And the voyagers too will eventually fail.

 Computers fail. Yours probably have. The internet is full of stories of large and costly failures.  Put “Large Computer Failures” into your browser, and you will get a large number of examples.  Approximately a month ago, the Bay Area Rapid Transit sytem (BART) suffered a major problem with its computer system (here) which caused a large number of people to miss appointments, work schedules, and so on.

 Why is it that visionary individuals and companies never seem to consider system failures in their ambitious plans.  I cannot help but wonder how much thought about system reliability is going into planning such complex systems as highways containing computer controlled automobiles, a sky full of UMV’s (drones), and an ever increasing number of robotic machines in our lives.  The government is not likely to release the reliability statistics of military drones.  I have not heard statistics on the reliability of industrial robots, or the few exquisitely cared-for automatic automobiles running around at present.  But I am happy to see that concern about such things is increasing and think there should be more.

 When our power goes off, we suffer some annoyance, and some worry about such things as long outages which can cause inconveniences such as thawing food, non-working cordless phones,computers, TV’s, etc. But failures in complex automatic control systems can cause deaths. Consider automobiles.

 Automobiles are typically driven by someone until failures become annoyingly frequent before they are scrapped.  What does that say about Highway 5 or Los Angeles being full of computer driven cars?  We hear of airplane accidents due to pilot error when taking over when the autopilot fails.  A common explanation is that pilots  spend so much time manually controlling the plane that they lose their instincts on what to do in an emergency.  Shouldn’t we worry more about computer driven airplanes with out-of-practice “standby” human pilots —or no pilots at all?.  Many leaders of the digital world believe that computers are more reliable at controlling complex equipment in a complicated environment than humans.  I have  not seen any data that convinces me that this is true over the total lifetime of such equipment.

  I have a reputation as a booster of creativity, innovation, and entrepreneurship, and in fact I am a believer.  But just because something is technically feasible on a small scale does not mean that it should too rapidly be introduced on a large one with attendant social complexity.  If your computer driven car hits another computer driven car, who is liable for the damages?  The car manufacturer?  The digital equipment manufacturer?  Are cars insured rather than drivers?  Are periodic maintenance checks mandatory?  How often does your car operating system require updating?  There are many factors involved other than the capability of hardware and software.

 

Christmas Greetings

I love Christmas because I can dork around with PhotoShop.  Example below. (forgive lack of shadows, sloppy scaling, etc.)

 

 

Two Views of the Internet

 As I told you in my post of November 3^rd, I was planning to order a number of the books reviewed that day in the New York Times that were concerned with the internet.  I have been reading through them, and the first two (listed under recommended books) held remarkably different views.  The first, Smarter Than You Think, was by Clive Thompson, a writer who has appeared in such publications as the New York Times, Wired, Fast Company, and Mother Jones.  He is a proponent of the internet, and I share many of his views.  As I have said before, I consider the internet a work in progress, with already great value, but great potential to improve —for instance in quality control (bad words for many internet folk) and increasingly smart, and perhaps specialized search engines. In particular, social networking, I believe, is in its infancy and it is difficult to guess what directions it will ultimately take.  I worry most about the effect of continually monetizing the internet.  I can’t help but think of the direction television, in particularly network television, has taken in order to squeeze more money from advertising.

 The New York Times Review of Thompson’s book is here, and portrays it well.   It is an honest report on the web and its potential starting with its service as improved memory for the individual, and ending with its function in connecting society.  It was easy to read and left me feeling good about the internet and the future. 

 The second book, Entitled Status Update – Celebrity, Publicity, and Branding in the Social Media Age, is by Alice Marwick, who after a research job at Microsoft, joined Fordham University as an assistant professor in communication and media studies.  The book reflects her experience working in and her research on the San Francisco computer scene during the post-bubble times of Web 2.0.  I loved the book for many reasons, even though as the review (here) said, she definitely verges into social-science-speak from time to time.  But as a professor, I have learned to enjoy academic dialects.  In particular the author loves to use the term neoliberal to describe the young entrepreneurs in her book, which she describes as “persons having a philosophy advocating deregulated free-market economic policies as a means to freedom and prosperity”.  I won’t quibble, except many of the people she interviewed seemed totally self-centered, which I didn’t think liberals were supposed to be.  And neo-cons believe in the free market. But not as much, I guess, in free expression, and more in spending time and effort converting others to their beliefs.

 I loved the book because I am a technical person living in Silicon Valley, working in a University whose new buildings seem to be named for entrepreneurs who have made a pile of money (Hewlett, Packard, Gates, Allen, Yang, etc.} and are filled with eager potential entrepreneurs.  I have a 50 yard line seat to watch the unfolding of the digital world.  Also because I have long been fascinated by social movements in San Francisco that promise a new, revolutionary, and more enlightened life (beatnik, hippie, now this).  Even though they do not result in radical change in everyone’s life style, they provoke thought and action, and always result in beneficial changes in viewpoint.

 In my view,  many of the younger digital advocates described by Ms. Marwick are a bit too much in their pursuit of becoming a brand and changing the world, especially those who she calls micro-celebrities. But they will help move our increasingly huge population to better itself whether they succeed or not, and the more of that we have, the better.

  Finally I enjoyed Ms. Marwick’s heroic effort to separate her scholarship from her personal feelings.  To the extent she failed, the book became more fun to read. I think I will send her a fan e-mail.

 

STEM (Science, Technology, Engineering Math) In Education

In the Opinion section of the Dec. 8 New York Times there are two pleas for more emphasis on STEM (science, technology, engineering, math) in education.  The first one, by their editorial board, is here.  The second, a column by Thomas Friedman is here.  You have probably run across similar pleas recently, because they are very much in the media.  Such pleas have occurred before—for instance after the successful Soviet launch of Sputnik—and probably are now occurring because of a combination of the importance of these topics in the economy and the availability of good jobs in such areas.

Most suggestions on how to remedy this situation focus on teachers and schools.  It isn’t that easy.  Granted that curricula and material should be updated and teachers be given more opportunity to stay current in these fields, but teachers are also carrying a huge load of socializing students, dealing with parents, and carrying out various administrative tasks—all for relatively low pay and lack of recognition.

Parents who do not work in these areas must try to contribute more by staying at least current enough to discuss them with their kids and help them with their homework.  These topics are relatively difficult to master, and often a good bit of encouragement is needed.  After the launch of Sputnik, an updated approach to handling math in the schools (New Math) was tried, but failed, and I was convinced at the time that part of the reason was that parents were not comfortable with the concepts.

People working in these fields, and the companies and laboratories they work for must become more involved in education.  This is difficult, because not everyone is a good teacher, and people who are good in their fields are busy at their jobs.  But if the U.S. is going to have more STEM people, industry and laboratories will have to play a part for motivational reasons as well educational.  Classes in school do not, and probably never will, give a good insight into the life of people in these activities. 

Unfortunately, STEM areas have changed very rapidly in recent history, due to things like digital electronics and computers, discoveries in areas such as genetics, space science and nanotechnology, and the rise of competence in them in many countries (China, Japan, South Korea, India, Etc.)  I was an undergraduate at Caltech, earned a Ph.D. from Stanford, have been involved in engineering in both industry and academia my entire career, and I can barely keep up.  As science and technology change, they become more complex, more esoteric, and less consistent with our intuition. I studied the theory of relativity as an undergraduate, but wasn’t sure I believed it.  A few years later we were putting corrections into orbits for planetary spacecraft to account for it.  Now we have string theory. As another example, think how much the discovery of the structure of DNA/RNA and the development of medical scanning have affected us in relatively few years.

As another problem, and one which has been around for all of my life, we need to do something about the negative attitudes that many people have about these areas. When I entered Caltech we were given a series of lectures about how to minimize our “Geek” reputation, which came with being a student at an engineering/science school.  This included advice such as “never talk math or science at a party”, and “drink and play a lot of sports”.  I later spent some time in art school at U.C.L.A. and didn’t seem to need suggestions on how to be cool.  Despite all of the money Bill Gates has made, the breakthroughs Microsoft has brought us, and the incredible good he is doing with his foundation, many people still are sure he is a nerd.

And we have to drop the “two culture” habit —dividing the population into what Stanford students used to call the “techies” and the “fuzzies”.  I’ve spent many years attacking it, and it is very deep in the U.S.  I suspect it is a legacy from the British.  It was the topic of C.P. Snow’s now famous 1959 Rede lecture, which is here.  After all, accredited engineering programs in U.S. universities require at least one full year of courses in the humanities and social sciences.  Why aren’t majors in these fields required to take at least one full year of engineering and natural sciences?  And why can’t I talk about math, science, and engineering at parties? 

 

Solder—A Good Product

Today I decided to work on my list of home maintenance tasks.  The first one I tackled was to find and repair a leak in our 110 year old iron yard-irrigation pipes. The leak had been going for a while, and had not caused a convenient surface pool, so I spent a few hours slipping around in gooey mud (clay soil) and scraping mud off of my shovel blade.  I found it, but the condition of the pipes was such that I decided we didn’t really need to irrigate the yard until April, so I shut off the yard water, and turned to fixing a nice old lamp that had belonged to my wife’s grandparents.

 The lamp’s wires led to the sockets through sharply bent tubes and various channels through the structure.  I found that wires and sockets alike were either broken or worrisomely used, so I changed all of them. making many connections as I went through it.  But although the connection locations inside the lamp structure were accessible through removable plates and such things, there was not enough room to use wire nuts.  So I twisted them together, soldered them, wrapped them with tape, and bingo—new lamp.  Much better than making ditches in the mud.

 But I realized how much I like to solder.  If the joints are clean and they are at the right temperature, the solder happily flows to cover all surfaces, bridges all gaps, and turns an ugly bunch of twisted wire into a shiny and beautifully conducting electrical joint. And you can do this without spending large amounts of money on equipment and learning esoteric theory. The photo at the left shows the content of my soldering equipment box, and as can be seen, it is old and dirty, but performs very well indeed.

  I was one time involved in the quality of soldered joints while at the Jet Propulsion Lab, and became interested in the history of soldering. I learned to solder from my grandfather, (clean everything and then clean it again— heat the joint, not the solder).   He was a part time blacksmith and sheet-metal worker, did larger scale jobs (tanks, funnels, buckets, ducts, etc.) and heated his irons with blowtorches. I was restricted to tin cans a single iron, and a tin snips, but that was enough to start.  In fact I scored in the second grade because we were studying dairies and had to build a model one.  I made my part out of soldered together tin can metal.

 As far as history, I figured that soldering had been around at least as long as my grandfather.  But I found out that the Romans had used solder to join the pipes in their empire, and that the solder had approximately the same mixture of lead and tin that was being used at JPL.  This was interesting.  Tin helps solder wet the surfaces it will join, which is good.  But beyond that,  a mixture of 63% lead and 37% tin is called a eutectic mixture.  It has the lowest melting point of any mixture of lead and tin, and will go from molten to solid without going through a stage when one solidifies while the other remains liquid —excellent characteristics for solder.  But I don’t think the Romans had metallurgy books that told them that, and through practice they figured it out. And the most commonly used solder these days for manual work is probably 60% lead, 40% tin—shades of the Romans.

 Going backwards, the Sumerians used solder to assemble their swords, the Egyptians for their metal work, and so on.  Soldering has always been used heavily in jewelry. Apparently soldering has been around for at least 7000 years—even before my grandfather’s time. As you might expect, Wikipedia contains a large amount of information on solder and soldering.  It is here.

 Of course there are many types of solder, since the label includes any metal or combination of metal that can be molten and when solidified will hold other metals together.  There are soft solders (what I used on the lamp), hard solders (silver solder, etc.) and brazing metals (copper based).  Increasingly, due to health and environmental concerns, solders used in electronics and for other purposes are no longer containing lead.

 Solder is produced in many forms —wire with or without a flux core, bars (plumbers), and sheets (jewelers).  It is also applied in many ways, ranging from manually (me) to mechanically.  Components with leads that penetrate circuit boards are commercially soldered to the circuitry with what is called wave soldering.  A wave soldering machine  contains a pool of molten solder in which a wave is formed, and the bottom of the boards pulled through the wave. Surface mounted components are attached by a different process called reflow, in which components are attached by a mixture of powdered solder and an adhesive flux, the solder then being melted by heat.

  But all of them produce the same magic—a quick attractive joint that is both electrical and mechanically sound.  Not as strong as a welded joint, in which the parts of the joint themselves are melted, but good enough for many applications.  And it can be produced quickly and at low cost—good for me, good for the electronic industry.