Dec05
Ten Years After Columbia
on December 5, 2012 at 12:31 pmFormer flight controller (and Shuttle program manager) Wayne Hale has been writing a series of blog posts about his recollections of the events leading up to the disaster. This week, he recalls the harbinger of the previous flight, that should have warned NASA about the problem, but didn’t.
There were simply too many issues to rehash everything, every launch.
I would call that a bad design. People naturally add complexity to things all the time. Simplicity is simply not valued as much as it should be. In cases like this, nobody has a big picture view of things meaning they will always fail in some unpredictable way. I refuse to accept that level of complexity is needed. People that assume it has to be are wrong.
A paraphrase of that Einstein quote seems to be tickling the edge of my brain… “as much as needed, but no more.” He understood simplicity.
Well, the complexity grew from the design constraints, schedule, and budget, which limited the number of iterations in the design cycle and didn’t give them the freedom to backtrack (schedule, budget) when a chain of earlier design decisions created extreme complexities during subequent phases.
It might be the result of NASA’s can-do attitude at the time (we can make anything work, not matter how complicated), and perhaps the subsequent decades of software engineering experience and hard-earned wisdom could provide some perspectives on the problem, directly addressing issues like complexity, robustness, supportability, maintainability, “brittleness”, feature creep, bloat, kluges, hacks, and patches on top of patches.
You could liken Wayne Hale’s situation to a manager for a major operating system release, with programming teams so buried under bug reports and chasing down customer issues that a critical error warning was eventually going to slip through unaddressed.
And yet there are people that idolize NASA and even resent the possibility of today’s Golden Spike announcement. I see this Google+ post by Amy Shira Teitel who works at Discovery News (http://news.discovery.com/contributors/amy-shira-teitel/)
Her post is here: (https://plus.google.com/107051665537162034944/posts/CEYDJUsCJQL)
It’s just like this post from yesterday (http://www.transterrestrial.com/?p=46314):
But sometimes the operational solutions are simple, as in the following:
Challenger – Don’t launch when it is too cold. (Yeager)
Columbia – When you see something large fall off the ET during launch, take some time to look at the damned orbiter before you bring it home.
Hindsight is the only exact science. Cheers -