Xref: utzoo sci.space:6453 sci.space.shuttle:992
Newsgroups: sci.space,sci.space.shuttle
Path: utzoo!henry
From: henry@utzoo.uucp (Henry Spencer)
Subject: Re: 95% vs. 99.9% reliability
Message-ID: <1988Aug9.205336.5810@utzoo.uucp>
Organization: U of Toronto Zoology
References: <3763@teklds.TEK.COM> <2087@silver.bacs.indiana.edu> <3659@drivax.UUCP> <1704@eneevax.UUCP> <20043@cornell.UUCP>
Date: Tue, 9 Aug 88 20:53:36 GMT

In article <20043@cornell.UUCP> dietz@gvax.cs.cornell.edu (Paul F. Dietz) writes:
>... The Rogers
>commission did NOT say that the shuttle would be safe to launch in warm
>weather.  They said that so many things were wrong with the joint design
>that it was impossible to determine what actually caused the leak...

True, as far as it goes.  From the data they supplied, though, it is
reasonable to infer that the combination of warmer temperatures and going
back to the old low-pressure leak-test procedure would greatly reduce the
risks.  Neither cold nor high-pressure leak tests had a perfect correlation
with joint problems, but both correlations were quite strong and the
combined correlation was even stronger.

Since there is no such thing as perfect safety, reduction of risk is all
one can realistically discuss.  Many people thought that a handful of
simple precautions, including those mentioned above, should reduce risk
enough to permit urgent missions to be flown by volunteer crews.  Sigh,
NASA didn't agree...
-- 
Intel CPUs are not defective,  |     Henry Spencer at U of Toronto Zoology
they just act that way.        | uunet!attcan!utzoo!henry henry@zoo.toronto.edu