Arguing is something that people do when they are emotionally involved, or perhaps even as recreation, as so wonderfully captured by the classic, timeless Monty Python "Argument" sketch (click here for a summary of the sketch; it also has links to the script and a video clip.) Engineers even like to sometimes argue technical issues "after hours", while unwinding at the local snack shop, watering hole, or conference.
But what should engineers argue about: The "best" processor? The "best" operating system or language? Which debug tools to use in various circumstances?
Yes, those are viable topics, but I'd like to propose some bigger, broader topics for engineers to debate with their fellows. To help you all get started, I have included links to some previous columns on these topics:
What went right (or wrong) with Space Shuttle program, and why didn't it live up to its initial promise (see here)?
Is spread-spectrum clocking a clever engineering technique, or is it a down-and-dirty cheat (see here)?
Should we be switching to Daylight Savings Time, or is it an outmoded concept from an era now gone, and which now brings no real benefit (see here)?
Is circuit design—as distinct from IC design—still a widely needed skill (see here)?
Should we replace the standard car steering wheel with a joystick (see here)?
Will the outcome of Texas Instruments' planned acquisition of National Semiconductor be a net gain, loss, or neutral (see here)?
Is IBM's Watson an indication of how far computers have come in replicating the human brain, or of how little we actually know about the brain (see here)?
And the big one: should "climate science" and even be considered as science, in the classical, traditional meaning of the term "science" (see here)?
Remember, it's OK to argue based on your personal beliefs, but it is also good to try to argue both sides of an issue—it's an important mental exercise.
Are there other "big picture" topics you would suggest engineers have a spirited argument about? ◊
RE: Space Shuttle
What went wrong with the space shuttle? The inital concept I belive was incorrect.
Putting people on board a spacecraft makes the launcher much more expensive because the rocket system must have redundant systems and ways for the crew to escape. Before the Challenger accident the Shuttle used to fly with live boost stages in the payload compartment. These stages had saftey devices to prevent them igniting in the Shuttle bay (which could ruin your day). After Challenger, these cargoes were banned. As a means of getting people to and from space the shuttle became very expensive as it was designed to haul large mass cargo into orbit along with the astronauts. Once the idea of taking satellites up wnet south, the shuttle had limited missions to perform until the ISS came along.
I believe that had the Shuttle been designed as a people carrier only, not a cargo delivery vehicle, it would have been much more useful. The spacecraft would have been smaller, and could have been launched by a classic rocket (payload on the front end). This would have allowed the crew a chance to escape from a Challenger accident. This would also have prevented the Columbia accident as there would be no foam to damage the tiles.
Cargo could go up the way it goes now, on Atlas heavy, Delta heavy or possibly an updated Saturn system. This scheme would have been less expensive in the long run (30 years of flights) and should have been much safer to operate.
A model is just that: A model based on the information gathered at the time of its construction. Wise scientists and engineers are quite comfortable to discard a model (plum pudding) for a new construct as the science dictates. This is the very reason why the atomic model continues to go through many revisions. Even the Bohr model has its limitations. (It “works” for hydrogen atoms.) If we do accept the current (“working”) models for science and engineering, we must also be aware of, and accept their limitations. There would be no need for the super-collider if the nuclear model was complete. It is never wise or productive to get emotionally attached to a construct. (Newton’s comfortable model of the universe compared to Einstein’s “radical” model.) We must allow the model to evolve as the knowledge dictates and accept the outcome. This process is never-ending.
"Is spread-spectrum clocking a clever engineering technique, or is it a down-and-dirty cheat?"
It's best not to get hit with a bullet in the first place. But if you have to take one, would you rather take the force in the small area of the bullet or spread it out over a bullet proof vest?
Most everybody agrees CO2 warming exists AND that it is very minor. Alarmists claim the minor warming, however, increases water vapor levels which have a stronger greenhouse effect which further increases water vapor levels which... (positive feedback).
This worked in simulations, but weather balloon data disagreed. Recent satellite data has also shown no water vapor increase. The alarmist acknowledge this but claim as they don't understand (or refuse to) why the levels haven't increased, they may not be wrong. Continuing to put faith in failed models is not science. Using the plum-pudding model today, for example, wouldn't be very scientific.
My guess is that it will take less time for autonomous cars to show up and become pretty standard than it would to re-train all of the drivers on the road to use a joystick.
Once kids are thought to drive with a joy stick, it won't be such an issue. Still, I question the useability of a joystick as an appropriate steering mechanism for a modern car.
The airplane analog doesn't really work. Aerodynamic surfaces have a very different feedback response than does a tire of the road. The level of precision required to keep a 70mph car away from the vehicle in the next lane, in most cases, is not needed in flying. Military folks can fly with that kind of precision, but only after millions of dollars worth of training.
Good topics covered. I would be interested in Should we replace the standard car steering wheel with a joystick? Not sure how a steering action can be replaced by a joystick. Anyway it would be easier to use joystick then a steering of course.
Sure, let's take on climate science Bill. I re-read your stand on it. Now here's mine. I've been to see the glaciers in Canada and Alaska. They're melting. Big time. So is the polar icecap. The coral reefs are burning out. Change on a climactic scale is happening.
It's observable. I'd like to study and understand what's going on. That's called science. Developing theories about the collected data, that too is called science. Climate science.
According to your strict constuctionist views, as I read your opinion piece, it can't be called science until it has a tested and working theory. Seems to me that plenty of physicists had flawed ideas of what the atom was (the cubic model (1902), the plum-pudding model (1900), the Saturnian model (1904), and the Rutherford model (1911)) before the Bohr model put electrons in circular orbit around a positively charged nucleus. Erroneous models don't negate the science.
I'd have to endorse that. Although NASA's manned space program has suffered terribly at the hands of a fickle public, NASA has lead the way with error-detecting and -correcting codes, remotely-upgradeable systems, image post-processing, and eking out the last little bit of life from disposable systems.
Re "30yrs sitting on their hands" being unfair. I'd agree with that. While humans may still be stuck in LEO, we not only have robots on Mars, but we have other robots that have taken pictures of the robots on Mars, including one in descent with its parachutes deployed. We have robotic probes around Mercury, Saturn, on the way to Pluto, soon on the way to another asteroid. We've got space telescopes recording gamma rays, x-rays and we have all of the upgrades to Hubble.
One of these days, hopefully, we'll see some ground breaking human space exploration again, but until then, I stand in awe of all of the robotic exploration currently in progress.
Join our online Radio Show on Friday 11th July starting at 2:00pm Eastern, when EETimes editor of all things fun and interesting, Max Maxfield, and embedded systems expert, Jack Ganssle, will debate as to just what is, and is not, and embedded system.