No Way to Run a Space Program
This morning's papers bring a couple of articles about the ongoing investigation of the Columbia disaster. The Washington Post highlights the latest area of investigation, a possible problem with the explosive bolts connecting the Shuttle to the boosters. The New York Times opts for the confusing headline "Shuttle Investigator Calls Theory on Foam 'Hokum,'" which gives the impression that some member of the committee has discarded the theory that insulating foam breaking off the booster tank damaged the left wing. Of course, that's not what the story says-- what's being dismissed as "hokum" is the prevailing theory about why the foam fell off the tank. There's no dispute about the fact that foam did come off the tank, and it's still the leading candidate for having done the fatal damage. I have no idea what the headline writers at the Times are smoking, but clearly, it's all Howell Raines's fault...
There's a bunch of stuff worthy of comment here, including Doug Osheroff's upholding of the Feynman tradition of investigation by experimentation-- the "hokum" line draws support from experiments Osheroff conducted in his kitchen sink. But heartwarming as that is, that's not what moved me to post.
The disturbing thing in these articles is stated most explicitly in the Post:
NASA's understanding of its own technology -- or lack of it -- was a recurring theme yesterday in the board's discussion of its probe into the causes of the accident. "Over and over," said retired Adm. Harold W. Gehman Jr., the board chairman, investigators have turned up issues involving shuttle components for which NASA engineers had "analysis" but no data based on experiments.
In the same vein, regarding Osheroff's experiments:
What he learned was that the foam-shedding problem was poorly understood and did not work the way NASA engineers had predicted. He also learned that the foam temperature readings engineers had been showing him since March did not represent "real data" but were based on a simulation. The temperature swings affecting the foam during launches "had never been measured" in the 27-year history of the shuttle program, Osheroff said.
And, finally (emphasis mine):
After the accident, NASA and the board for the first time tested the bolt catcher's ability to absorb the impact of the exploding bolts. They found that the catcher fractures under less pressure than predicted, Barry said.
This is simply disgusting. What in hell are these people getting paid for? This isn't the way to do science. This isn't even the way to do engineering. This is fucking alchemy.
It's bad enough when they fail to test robot probes and the like. That's incredibly stupid, but at least in those cases, the only things lost are mechanical components. But not testing components of a manned vehicle? What could they possibly be thinking? Obviously, common sense is too much to expect from government contractors, but if nothing else, you'd think they'd have some sense of pattern recognition... They evidently learned nothing from the Challenger debacle.
Everyone involved in this travesty of an engineering project should be stripped of all their technical degrees, and made to start over at the high-school level. Hell, they should be busted down to elementary school-- that's where they first start talking about the scientific method, which NASA has apparently collectively failed to pick up along the way.
The most damning bit comes at the end of the Times piece, though:
Russell D. Turner, former president and chief executive of the United Space Alliance, the main shuttle operations contractor, said that contractors had put safety first and that his company had cut its potential bonus by adding to safety spending. Mr. Turner said United Space was well aware that foam debris was damaging the orbiter on every launching but that the foam was from the external tank, the province of another contractor.
Dr. Osheroff, with no evident humor, called that "very funny." United Space, he noted, is partly owned by Lockheed Martin, which produces the external tanks.
This whole thing leaves me with the sort of speechless rage usually reserved for John Ashcroft's policy decisions. There simply aren't the words to describe how deeply and profoundly wrong this is. This goes beyond "disgusting," all the way to "criminal." If this is the way these clowns conduct their business, then at the very least, they should lose their contracts. Personally, though, I think I would favor strapping the chief executives and program managers of the responsible contractors to the leading edge of the Shuttle wing for the next launch...
You Are Responsible for the Technology You Choose to Use
The Invisible Adjunct is back after a Movable Type meltdown that wiped out a few posts and one long and interesting comment thread. Brad DeLong suffered a similar problem a few days ago, though he appears to have taken down the post mentioning it.
Which brings up an issue about blogging tools that has always amused me: lots of people spend lots of time bitching about Blogger and its occasional posting outages, and the company has been verbed as a term for losing archive permalinks. I've seen dozens of posts of the form "Person X needs to get off Blogger and onto a Movable Type powered site" (if it's from Glenn Reynolds or one of his acolytes, add "Sekimori-designed" in there (another point of bafflement, as I'm generally underwhelmed by their designs-- they avoid the worst sins of too-flashy web design, but, really, an afternoon spent over at the Web Design Group would be cheaper and more useful in the long run.)).
And yet, I can't think of a Movable Type-powered blogger who hasn't suffered a debilitating crash of the type suffered by Brad and the Adjunct. I also can't recall Blogger frying anybody's site anywhere nearly that bad...
I No How to Spell Reel Good
I had my lab class this term submit their formal reports electronically, and I've been marking them up in Word (using the "Track Changes" feature). The lab in question is the karate board lab, in which students measure the deflection of a pine board as a large force is applied to it, and from that determine how fast a hand would have to move to break the board.
On two different papers now, I've had students describe the procedure as "measuring how far the board bended." Which struck me as a "you fail, for not even spell-checking your paper" sort of error.
To confirm my suspicion, I ran the first paper through the spell-check, and got nothing. Puzzled, I highlighted the paragraph containing the offending word, and tried again. Nothing. I created a new file, typed "bended" into it, and ran the spell-check.
"Bended" is apparently in the standard dictionary for Microsoft Word. That pretty much tells you everything you need to know about the company.
(For the record, SpellChecker.net, the service Blogger uses for spell-checking posts, does flag "bended" as an error. Sadly, it doesn't suggest "bent" as an alternative, but at least it flags the word...)
They Wrote a Book About It, Said It Was Like Ancient Rome
I'm thoroughly exhausted from this weekend's family visits, so substantive academic blogging will wait until tomorrow. In the meantime, though, it's nice that our elected officials care enough to do idiotic things to provide me with an adrenaline rush so I can get through the day:
Senator Larry E. Craig of Idaho is blocking the promotions of more than 850 Air Force officers, including young pilots who fought in Iraq and the general nominated to bail out the scandal-plagued United States Air Force Academy, in a rare clash between the Pentagon and a senior Republican lawmaker.
Mr. Craig's price to free the frozen promotions now awaiting final Senate approval? Four C-130 cargo planes for the Idaho Air National Guard.
Pentagon officials express outrage that for more than a month Mr. Craig has single-handedly delayed the careers of hundreds of officers and stymied important Air Force business for a handful of parochial planes. They are vowing not to give in to his pressure. Calling the move blackmail, one senior military official said, "If we say yes to this, Katie bar the door." The official, like others contacted for this article, spoke on the condition of anonymity, fearing retribution from the senator.
Is there any chance we could get a horse appointed to the Senate? It'd be cheaper in the long run, and it could only improve the functioning of the government...
(Story via Ted Barlow.)