posted
Is there anyone else here who has discovered the writing of Nevil Shute, whose full name is Nevil Shute Norway?
I love his books, especially "Round the Bend", "Trustee from the Toolroom", "The Legacy" which may be more familiar to U.S. readers as "A Town Like Alice". Another favorite is "No Highway". His most well-known book is "On the Beach" but I'm not as fond of it and don't think it's very representative of his work.
He's a British writer who was also an aircraft engineer and wrote in the period between the wars, and during and after the second World War. He moved to Australia at some point and many of his books are set there.
I'm reading his autobiography now, called "Slide Rule" and it's got on the back a blurb that should go on that thread about blurbs you don't want to appear on your book, "...probably the best, most thrilling and most interesting story ever written about an engineer." But I think it actually is extremely interesting and even thrilling. He's a great writer, after all.
The thing that is striking me so much about this book so far, and I'm only about halfway through it, is his account of the disaster of the R101 airship built by the British government. He assesses very directly the causes of the trouble, or his view of them. I've read Richard P. Feynman's account of the Challenger disaster in the U.S. in the 80s, and the two accounts correspond point by point. They made almost exactly the same mistakes! I'm just floored by how similar are the two cases.
In both situations the pressures and realities of the government system, when they set about to design and build a complex technical machine which represents a substantial advance from the current state of the art, are simply causing technological disaster. Governments aren't suited to do such things. People in power in government will always choose the political decision, because all of their job security and advancement depends on that, and does not rest on real considerations of technical suitability, efficiency, affordability, and excellence of design. The engineers were very concerned about problems, but management simply put a brave face on it and overrode their objections, because of political considerations. The entire process was flawed and many terrible design decisions were made because it was too late to back out, they'd committed and it would look funny if they'd changed their minds. Instead of using proven parts, like engines that were already in use on other aircraft, they stuck to their own design which came in at double the expected weight and so on. Exactly. The. Same. Mistakes. It is somewhat shocking to me that this was all so clear already in 1931 or 1932.
We (the human species) had fifty years in which to absorb this knowledge and take it into account between these two disasters, and we failed to do so. We can't afford not to learn from the mistakes of our past. We must do better than this.
Quite aside from that, though, do check out Nevil Shute if you haven't already. He is a wonderful writer. Very understated but extremely smart, wise, and just a master storyteller.
posted
We did an experiment in Behavioral Science in College.
We separated into small groups and were given data about the failure of a key component in a race car engine at certain temperatures and asked whether we would race on a given day. The component had failed repeatedly before and this breakage *could* result in catastrophic engine failure, but never had.
Are group was unanimous: "go for it! if the motor dies, well... racing is a risky business". All of the other groups reached the same conclusion, though I don't know if they had more debate.
The data was *precisely* the data from the Shuttle O-ring Failures. Granted that catastrophic engine failure on a space shuttle is a bigger deal than on a race car, what is very similar is the presumption of risk and the aggressive attitude of those involved. I would not be surprised to find the astronauts, given the same data, saying "hey... we run risks here... that's our job" if it was just them, though I think the inclusion of civilian McAuliffe would have tempered their characteristic bravado and impatience and increased their equally characteristic professionalism and attention to detail.
All just supposition, of course, but an interesting group study that has stuck with me ever since.
Posts: 3846 | Registered: Apr 2004
| IP: Logged |
posted
"hey... we run risks here... that's our job"
I saw an interview with McAuliffe where she basically repeated these words. The fact that she was a civilian should have little bearing on it. She knew she was taking a huge risk.
That said, I agree with Bob. The viability depends on its visibility. So public opinion outweighs the opinion of the "can do culture."
Posts: 3735 | Registered: Mar 2002
| IP: Logged |
posted
well, a certain amount of can-do is necessary to get things done...
but yes, no disagreement from this quarter, just saying that large groups of college students almost unanimously decidded the same thing with the same data.
as for Ms. McAuliffe... I didn;t mean to imply that she didn't know the risks, but that her presence would temper the willingness of a professional flight crew to take risks... not her fault or theirs... or a bad thing. Just that it's there.
Posts: 3846 | Registered: Apr 2004
| IP: Logged |
posted
A catastrophic engine failure on a car doesn't mean that the driver would be harmed.
But then college students also don't think much about the risk of boozing until they puke. Or electrocuting people to teach them a lesson.
Posts: 8501 | Registered: Jul 2001
| IP: Logged |
posted
To be fair, that experiment has been duplicated with people of varying ages/nationalities/etc, and I don't think there was much of a difference between college students and the rest of the world.
Posts: 5656 | Registered: Oct 1999
| IP: Logged |
posted
I guess in addition to the loss of life in situations like this, I deplore the expenditure, too, which is far higher than necessary and produces a huge boondoggle instead of a smoothly working system. I deplore also that the main thrust of our space program (or Britian's airship program in the 1930s) was taken up with and dictated by such a badly flawed body of work. It sidetracks the whole industry.
For instance, in the case of the R101, even though there was a successful and proven design by industry, the R100, along the same lines, the decision was made to shelve the whole airship industry in Britain because of the completely avoidable disaster on the R101. In the case of the Shuttle, of course, the human species' entire manned thrust into space was brought to a halt.
Engineering 101, How to design and build a disaster:
1. Design from the top down. Instead, design should always proceed from the bottom up. For instance, engines should be designed and tested first, then the rest of the machine designed around the engines. Use off-the-shelf components with proven track records whenever possible. When suitable components don't exist, design and test the components first before the large scale design is firm.
2. Commit early to a specific design framework, with later changes made impossible for political reasons as problems appear. I've seen equally disasterous results by limiting the suppliers of components for political reasons, rather than technically choosing the best product with the most reliable manufacturing and distribution system. Instead, technical considerations should always be allowed to dictate the components and designs used, and changes during the design process are to be expected and accepted as normal.
3. "Spin" of results to appear better than actually seen; hiding things up when anything goes wrong. When you do pioneering development work, things always go wrong. Smooth design and testing with no major bugs is only a feature of MATURE industries and incremental designs. Pioneering, of necessity, requires a process of discovering problems and redesigning to take into account what you find out. People who don't understand this should not be overseeing pioneering design work. If it's necessary politically to gloss over problems and pretend our initial choices were always right, then we're in the wrong business altogether and we are predestined to build a badly flawed machine. There is huge pressure on engineers, more and more as time goes by, to say everything's fine, no problems here! The customer, your management, everyone wants you to tell them that. It's very important, it's your professional duty, to be entirely honest at every step about your true assessment of the technology from a physical standpoint. This may well cost you your job, but if it does, then it was a lousy job that you are better off without. Because Mother Nature cannot be fooled! She doesn't care how well you spin things to make them look good, she is inexorable in her assessments.
4. Rush at the last minute to complete tests, check-out, commissioning for political reasons. Skip over some tests and change criteria on others to insure that you pass. Instead, when you're checking out a pioneering design, expect glitches to pop up at every stage, and be sure to take time to put them right. Never cut corners for the sake of show.
5. Place the power of go/no-go decisions into the hands of people who aren't technical. Allowing nontechnical management to override the recommendations of the designers and testers on whether the machine is safe to operate is a recipe for disaster.
If we learn these lessons for all time, we can aviod almost all such disasters in the future. Government workers are usually rewarded for doing nothing, and punished for trying to be conscientious and exacting in their work. Goverments have no bottom line, so they put political considerations ahead of performance, economy, and all other concerns. Therefore they are unsuitable bodies for carrying out large scale pioneering design in technical fields. In government work, fulfilling the paperwork requirements and satisfying the letter of the law becomes paramount, to the exclusion of the actual enterprise at hand.
If we fail to learn these lessons we are bound to repeat, over and over again, disasters like R101, Challenger, Columbia, and countless others. The human species is running very close to the wire, now, of survival or extinction. We can't afford to squander our resources, not just money but technical expertise, sweat equity, and the confidence of the people in our ability to proceed with important advances in space and other technological fields.
Posts: 6246 | Registered: Aug 2004
| IP: Logged |
posted
I saw this thread on another forum... three posts down is a piece by Bill Whittle about the loss of the two shuttles. It's very long but well worth the read, IMO.
Printer-friendly view of this topic
![[ROFL]](graemlins/rofl.gif)
But I think it actually is extremely interesting and even thrilling. He's a great writer, after all. ![[Smile]](smile.gif)
