---------- Forwarded message ----------
Date: Tue, 6 Mar 2001 16:47:07 -0600 (CST)
From: Gene Mechtly <[EMAIL PROTECTED]>
To: [EMAIL PROTECTED]
Cc: [EMAIL PROTECTED]
Subject: SI at NASA
Dear Mr. Weinstein,
Today, I downloaded and printed the Inspector General's Report
"Assessment of NASA's Use of the Metric System, G-00-021" 2001 Feb 20,
and would like to make the following comments.
My direct experience with NASA has been ten years as an employee
at the MSFC (including work at the ABMA, and the GMDD of the AOML in
Huntsville). While there, I wrote "The International System of Units"
which was first published in 1964 by the NASA Scientific and Technical
Information Division as NASA SP-7012.
In 1965 (Jan 18), Raymond L. Bisplinghoff, then NASA Associate
Administrator, directed NASA Research Centers to use SI preferentially;
His words were, in part, "...The International System of Units of 1960
(SI) should therefore be considered the preferred system of units, and
should be employed in the future by the Research Centers as the primary
system in all reports and publications of a technical nature, ..."
Unfortunately, for the past 35 years the manned program and the
aerospace industry have delayed if not overtly opposed full implementation
of SI in all NASA programs. I am pleased that the Inspector General's
Assessment makes recommendations for stronger movement in favor of SI.
However, there is one major distinction that is missing from
the IG Assessment; *software* versus *hardware* metrication.
Metrication of software for calculations and data analysis
is easily implemented with benefits which greatly exceed costs. Updating
mental habits to SI is a small price to pay for fewer errors in data
exchange (The disaster of the Mars Climate Orbiter is an example).
On the other hand, discarding hardware objects which were not
originally designed to millimeter specifications and replacing them
by objects designed originally to mm specifications would be, in reality,
extremely costly and objectionable. A reasonable objective is, instead,
designs of *new* objects, and the *specifications* of old objects in SI.
An old and proven valve design can be retained, but given new millimeter
specifications with appropriate tolerances for manufacturing and testing.
Necessary total hardware replacement is a false claim by opponents of SI.
Please ask the NASA Chief Engineer to make clear distinctions
between software conversions to SI (easy) and conversion of hardware
specifications to SI (more difficult) when writing new directives to
implement preferential use of SI throughout NASA.
Sincerely,
Eugene A. Mechtly, College of Engineering
University of Illinois at Urbana-Champaign (uiuc)
