 |
 advertisement
|
 |  | Letters: Faster? Better? Cheaper?
posted: 05:21 pm ET
21 April 2000
|
letters_000421
Historian Howard McCurdy's views on NASA's "faster, better, cheaper" approach generated the following exchange.
To the Editor:
Mr. McCurdy is correct in his views that, with proper organization and technical leadership, "better, faster, cheaper" does not mean compromising on quality. Many of us on the Pathfinder Flight Team expressed these views to NASA management. Unfortunately, NASA management does not comprehend these principles. As a result, the majority of JPL's (the Jet Propulsion Laboratory's) spacecraft work was contracted out and most of us have departed the lab. Dan Goldin recently stated that the high Silicon Valley salaries were drawing us away. This is not true because that condition existed for many years prior to our departures. Instead it is the lack of "hands-on" work. JPLers like to get their hands dirty building spacecraft, not pushing paper. As we depart the lab, NASA and JPL will lose their spacecraft-building capability. Unfortunately, another wave of my Pathfinder friends have just left the laboratory.
 |  | |  | More Stories |
| | | | |
|  | Related Links |
| | |
| |  |
Steven A. Stolper
The writer was a flight software engineer for the 1997 Mars Pathfinder mission.
To the Editor:
I would not call this a good critique. McCurdy's basic argument is that the computer industry manages to make things faster, better and cheaper, so why can't NASA? He basically seems to have little knowledge of the computer industry. Examine this passage:
"If you use the right management techniques and you do the correct engineering, and you do the testing that is necessary -- and you don’t under-fund the project! In concept, it is a workable system. It works in Microsoft, it works at Digital Domain, it works at Intel, it worked for Mars Pathfinder."
Basically, this is not an accurate analysis of the computer industry. First, Microsoft is not cheap on the scale of a Mars mission; a Microsoft product launch costs many times more than a NASA mission. Even then, though, Microsoft manages to reduce cost, in essence, by shipping buggy product and then using the customer dollars to fix the worst of the bugs. Raise your hands: how many of you have used the "control-alt-delete" feature on your Microsoft computer? How many of you have heard the phrase "blue screen of death"? OK, if you were entering the Mars atmosphere, you would be dead now.
Further, the computer industry looks like a successful implementation of "faster-better-cheaper" because it is built on the corpses of its numerous failures. The secret of the computer industry's success is that -- as an industry -- it tolerates failure (not necessarily on a company-by-company basis, but as an overall industry). Anybody remember Osborne? Coleco? Adam? Lisa? Sinclair? Wang? Atari? Every single one of these were market leaders in their day. Has anybody read the book Soul of a New Machine? A great study of technology innovation; and have any of you used a Data General machine in the last, say, decade?
So, if I personally were using the computer industry as a model to understand how to make "faster-better-cheaper" work, the conclusion I would draw is that you have to tolerate failure. Your "faster, cheaper" space probes will fail, and then when you find out how they fail, you would fix the problem. You would develop lots of different products -- lots of different companies (computer) lots of different spacecraft (NASA) -- and when some of them fail (as some computer designs fail), you dump those designs and go on.
Geoffrey A. Landis
Author, Mars Crossing (available from Tor Books December 2000)
http://www.sff.net/people/geoffrey.landis
Howard McCurdy responds:
Actually, the basic models for "faster, better, cheaper" can be found in the "skunk works" philosophy developed by Kelly Johnson during World War 2 and the old National Advisory Committee for Aeronautics (NACA) field centers out of which NASA was formed. In the NACA experience, bearers of the underlying culture emphasized in-house technical capacity, hired exceptional people, gave them hands-on work, let them do lots of testing and encouraged them to take risks on projects where they usually tried to do something new.
When the space race began, the NACA model was supplanted by the organizational necessities of large-scale project management. Officials directing Project Apollo, for example, contracted out the bulk of their work and instituted formal systems management as a means of controlling the coordination problems that inevitably arose. The results were highly successful and -- in today’s environment -- prohibitively expensive.
"Faster, better, cheaper" returns to a system in which a small group of people working closely together on a project try to control risk by creating a team with a high capacity for fault recognition and problem solving. This requires exceptional people and a great deal of testing, risk-taking and hands-on activity. Various "Information Age" companies have used this approach for product development.
Unfortunately, the teamwork techniques are fragile. Officials who insist upon cheap projects and then fail to institute the necessary teamwork principles invite failure. This happened with Mars Climate Orbiter. It will keep happening until the basic management lessons are learned.
Howard E. McCurdy
American University
SPACE.com welcomes Letters to the Editor. Letters intended for publication should be under 250 words, and may be edited for style and clarity.
|
|
|
