Dear John,

Thank you for your insight into engineering practice at NASA. I am
fascinated to watch as NASA goes through its metrication process. It's the
process that I think is important � the actual metric theory stuff is fairly
simple and can be learned relatively quickly � it is the human and social
process of metrication that I find fascinating.

Again, I have intetrspersed some remarks.

on 2003-09-02 06.55, John S. Ward at [EMAIL PROTECTED] wrote:

> On Monday 01 September 2003 00:04, Pat Naughtin wrote:
>> Dear John and All,
>> 
>> I have interspersed some notes from an Australian perspective.
>> 
>> on 2003-09-01 03.13, John S. Ward at [EMAIL PROTECTED] wrote:
>>> Hi,
>>> 
>>> I'm a scientist/engineer working for NASA.  The metric system is used
>>> extensively at the Jet Propulsion Laboratory, but inch-pounds are still
>>> dominant.
>> 
>> What do you mean by the word, 'dominant'? Could you guess what proportion
>> of measures (and calculations) are done in metric measures and what
>> proportion are done in old (USA or UK) measures? Do you know of anyone who
>> calculates the chemistry of fuels (for example) in kilograms and then
>> changes their results to pounds when they have their results. I know thatin
>> the brewing trade, there is a tendency to do this because the calculations
>> are so easy in metric and so difficult in Imperial (or USA sustomary)
>> measures.
> 
> Hi Pat,
> 
> The short answer is, everything made out of metal is made in inches, and the
> large majority of most spacecrafts are metal.  The large majority of
> mechanical drawings are in inches, and most fasteners are "standard" (i.e.
> they don't make sense in any units, but are specified in inches.)  JPL is the
> most-metric NASA lab, and yet inches dominate here.  Other labs deal mainly
> with manned space flight and legacy programs (Space Shuttle, Space Station),
> which done almost completely in inch-pounds units.
> 
> There is a very strong correlation between education level and metric usage
> within NASA.  Most scientists and Ph.D. engineers strongly prefer metric.
> Thus, the really high-tech stuff like electronic device design and
> fabrication is done all metric.  Higher level requirements and interfaces are
> also often all-metric.  However, we eventually turn these designs and
> requirements into inches drawings, because as far as I can tell no NASA
> machinists think metric.
> 
> Machinists make a big point to not do anything in metric, and if you give them
> a metric drawing, they will pencil in inches everywhere, and make the part in
> inches.

Be fair to the machinists. Have they got things like metric lathes and
metric machining mills. It might be necessary for them to pencil in the
inches, fractions, and 'thous' simply because that is the only way they can
set-up their tools to make the parts specified in the drawings. If I only
had old lathes, I would be truly annoyed if the engineers kept specifying
things that I couldn't easily make � I might even get annoyed at the units
preferred by the engineers and try to avoid them.

[For about 10 years, I worked with the Australian Commonwealth Scientific
and Industrial Research Organisation (CSIRO)*, and from my laboratory to the
lunch room I walked through the machinist's workshop � they were less than
kind to scientists and engineers (*&#$ %^&$  (&#$% for instance) who
specified units that weren't available on their particular range of
machines. They were very proud of their workmanship and got annoyed when
their job was made more difficult. Also consider that of all the people who
change to metric, it is the machinists who have the most difficult task,
largely due to the settings on their older machinery.]

* CSIRO might be roughly equivalent to NASA in its composition by
professions and by crafts. CSIRO has a high proportion of scientists,
engineers, and architects.

> To me, this seems like a lot of extra work and creates opportunity
> for error.

True.

> Yet most machinist bear a grudge against the metric system, and
> make it a point to avoid it as much as possible.  They also make it a point
> to train new machinists inches-only,

New trainees have to be trained into the arcane ways of any craft � twas
ever thus. On a more practical note, they have to be trained to use the
available (old) machines, and if the culture of NASA is the same as every
workshop that I have ever been in, it is the newest machinists who get to
use the oldest equipment.

> indoctrinate the anti-metric grudge, and

This goes with the territory � literally the territory of the workshop floor
� everyone else, especially 'higher-ups' is to be treated with scorn. It's
an 'us vs them' thing. We all do it � for example, how's your opinion of
senior management to-day?

> avoid buying metric tooling.

If this is true, it is a serious problem. This issue needs to be
investigated and have resources (physical, financial, and human) thrown at
it at soon as possible. In my opinion, this is not an issue that should be
treated lightly, nor treated slowly.

> Therefore, most engineers draw inches drawings.

Is this a cop-out because the engineers and the machinists need to spend
some time talking to each other?

> The real reason the JPL is not metric is because there is no organized
> conversion effort.

This is a truly serious issue. Maybe senior management has not yet
discovered how to go about it; maybe senior management does not regard
metrication as a serious problem (they don't remember MCO); maybe senior
management doesn't have the resources to tackle such a large all-embracing
cultural change such as all of the the measures and all of the the measuring
systems used by NASA.

Whatever the problem, NASA needs to investigate and act on an 'organized
conversion effort' as a matter of urgency.

> Everyone is left to fend for himself.

This is bad. This is espedially bad in an environment that includes
scientific and engineering and craft components, as all of these are highly
motivated by developing jargon words, jargon phrases, and even jargon
equations to develop and define their careers within the professional group
to which they see themselves as belonging. [After 10 years at CSIRO, ask me
how I know!]

> There is no clear
> message what units or parts are preferred, or when to use which system.

This is solvable, but it will require leadership from NASA's engineers, but
leadership is what engineers are good at.

> There is no training.

If this is the case, it is appalling. Training is the cheapest component of
a metrication program. Develop some metrication aims, specify some
metrication goals, find some metrication training contractors (let me know
if you want some help) and get on with it. As I said, training is the
cheapest component of your metrication program, and if you use external
contractors, you do not need to be saddled with an ongoing financial
commitment.
 
> There appear to be no rules, and no deadlines.  Most
> people blindly continue to design and buy all inch-pounds hardware and
> equipment, apparently oblivious to two acts of Congress and an Executive
> Order requiring NASA go metric.

As I said, develop some metrication aims and specify some metrication goals.
Use the acts of Congress and the Executive Order requiring NASA go metric as
tools to convince anyone who will listen to help in the development of
NASA's measurement aims and goals.

> Yet at some point most hardware is truly
> mixed.  I'm sure its very common to have a 1/4" screw and an M6 in close
> proximity in the same component.

This sounds like engineering, so I'll duck. [We agricultural science
persons, with our one year program of agricultural engineering should bow
low and grovel before our seniors and betters.]

> The crazy thing is, practically everyone here has high exposure to the metric
> system and has had to learn to cope with dual systems.  A well-planned,
> organized conversion would be relatively painless and in the long run would
> make practically everything easier for everyone.  Unfortunately, the will to
> get it over with does not seem to exist.

Feel confident, John. You know and I know that in the end, NASA will be a
fully metric organisation. The only doubt about NASA's metrication is how
long the process will take. In this context it might be interesting for you
to ask your colleagues these two questions.
1   Will NASA eventually be an all metric organisation?
2   How long do you think it will take before NASA is all metric?

Sadly for you, when compared to other organisations that I have seen, NASA
seems, so far, to have chosen one of the slowest possible routes to
metrication. 

Excuse me if you have seen it, but I am reposting a draft document (with
slight modifications) on the relative speed of metrication that I posted
here recently (see below).

Cheers,

Pat Naughtin LCAMS
Geelong, Australia

Pat Naughtin is the editor of the free online newsletter, 'Metrication
matters'. You can subscribe by sending an email containing the words
subscribe Metrication matters to [EMAIL PROTECTED]

Ways to go � to metrication
Pat Naughtin

Your posting about metrication at NASA set me to thinking about the possible
routes to metrication. It seems to me that there are four main ways to go,
with many others falling somewhere between these four routes. As you read
this you might like to consider where NASA currently sits among these
pathways.

My four ways are:

1   The Worst Way
To use this method, you essentially aim to use 'hidden old units'. Let me
explain how to do this. Firstly, you convert all the old values in your work
directly to metric, eg. if one of your spare parts requires an 18 inch
component, you call this a 457.2�millimetre component, or a 45.72 centimetre
component, or better yet you list both 457.2�mm, and 45.72�cm, and
18�inches, and 1' 6" in your parts list. Secondly you provide all of your
staff with extensive conversion charts and booklets (that include the
457.2�mm value as well as some 69�432 others � 69�427 of which your staff
will never use!) designed to fit neatly in the bottoms of desk drawers and
to paste firmly on the insides of office cupboard doors � in these positions
these charts and booklets can delay your metrication program for many years.
These two simple techniques should delay metrication by upwards of 100 years
and they should cost NASA heaps as they are clearly based on the idea of
giving an illusion of metric progress while retaining our individual
mindsets. Regard 100 years as the absolute minimum for this path. Consider
the New York Stock Exchange as an example; by hiding the new (in 1793)
decimal currency behind the old (in 1792) 'pieces-of-eight' in quoting stock
prices, they were able to delay decimalisation of stock exchange quotes from
1793 until 2001 � a delay of 208 years � very impressive. Think also of the
oil industry's barrels (1860s) and you will begin to see the delays that are
possible.

2   The Nearly Worst Way
This technique places its trust in dual measurements. You supply all staff
with dual sided rulers and tapes, and copious quantities of conversion
charts. You insist that all documentation is done using dual measurements,
although it is rare for companies using this route to specify which takes
precedence presumably on the basis that the random generation of precedence
by each member of staff is the best policy for the company as it gives each
member of staff freedom of choice. You can probably plan for this technique
to take your company around 75 years to 100 years for its conversion. As an
example consider the Kodak company who made their decision to use 16 mm B&W
film in 1910 and 16 mm color movie film in 1929; they also decided to use
photographic paper based on ten inches by eight inches sizes with multiples
and sub-multiples of this inch size. This dual measurement has delayed
metrication at the Kodak company for 93 years so far, but in 2002 they began
yet another metrication effort � they might be able to make the transition
within the 100 years that I suggest for this technique.

3   The Second Best Way
Use centimetres. This way you can keep those lovely fractions in your
vocabulary. Remember that using millimetres exclusively, almost completely
removes the need for fractions (together with their attendant extra time and
costly errors). Do not form a company-wide measurement policy � let each
department form their own. In this way the manufacturing departments will
eventually use millimetres as others do in their crafts; the office staff
will eventually use inches and centimetres � as others who sew and knit use
these measures; the sales staff will change their measures according to how
they see their customers in terms of conservatism/progressiveness � in any
case the sales staff often don't really care to know about measures (other
than sales figures at the end of the month). It's hard to guess how long
this pathway takes. I have direct experience of the textile industry in
Australia and they are still struggling (while making snail-like progress)
with metrication after 30 years � my guess is that it will take them at
least another 20 years, so my estimate for the 'Second Best Way' is between
50 and 75 years. It looks like the Microsoft company are, in part, using
this technique � when I try to use one of their drawing applications, I have
to divide centimetres into halves, quarters, and eighths � very decimal!

4   The Best Way
Use other people's experience. Find out where metrication has been
successful anywhere in the world and simply copy their techniques. Form a
measurement policy � by saying things like 'Of all the length measures
available in the metric system, NASA has chosen to use millimetres, metres,
and kilometres' and 'The centimetre will not be used in calculations and
they will never be written down'. 'It is the policy of our company to use
rounded metric values of components wherever we can � we believe that this
will lead to a simplicity that will reduce errors in production,
administration, management, and sales'. These quotations are adapted from
Australian building literature of the 1970s. Using this technique the whole
of the Australian building industry had essentially metricated in under a
year, and if you count the laggards, definitely had completed the process in
under two years.

It seems to me that NASA has chosen a path somewhere between the 'Worst Way'
and the 'Nearly Worst Way' to metrication � I wish you well � but there's no
need to put the champagne on ice just yet, it's a long time till 2103, or
should that be 2153.

Cheers,

Pat Naughtin LCAMS
Geelong, Australia

Pat Naughtin is a speaker, writer, and editor. Pat has written several books
and he has edited and published many others. For example, Pat was the lead
writer for the 'Wool' chapter of the Kirk-Othmar Chemical Engineering
Encyclopedia, and he edited the measurement section for the Australian
Government "Style manual: for writers, editors and printers". Pat is
recognised by the United States Metric Association as a Lifetime Certified
Advanced Metrication Specialist (LCAMS).
--

Reply via email to