Re: [OT] Re: Question for designers...

"Wayne Lundberg" <Waynelund@xxxxxxxxxxxxxxxx> wrote:
"D Herring" <dherring@xxxxxxxxxxxxxxxxxxx> wrote in message
news:b9ydnWkwp6ga8E7bnZ2dnUVZ_gWdnZ2d@xxxxxxxxxxxxxx
Joe Strout wrote:
In article
<yuBAi.56386$ax1.51236@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx>,
"Wayne Lundberg" <Waynelund@xxxxxxxxxxxxxxxx> wrote:

I was brought up with the metric system in my school, drafting,
geometry and
all that stuff. I started serious design work in the metric system
and
ended
up pulling my hair out at the total confusion of that damned decimal
point
and where to put it.

This seems very odd to me. You find it easier to multiply and divide
by numbers like 12, 16, and 144 than numbers like 10, 100, and 1000?

A half inch is 1/2"; a quarter inch is 1/4"; a sixteenth of an inch is
1/16"; three times that is 3/16"... A half centimeter is 5mm;

Or 1/2 cm.

a
quarter centimeter is 2.5mm;

OR 1/4 cm.

a sixteenth of a centimeter is 0.625mm.

or 1/16 cm.

What's three times that?

3/16 cm.

Plus your metric set probably only
contains multiples of powers of 10... so most of these sizes aren't
available...

If the sizes you need aren't available, you bought the wrong tools. I
don't really get your point.

Unit conversion usually only happens in higher-level design tasks such
as "how long are 50 Xs" -- at which point the unit conversion only
incidentally hinders an otherwise complex operation. Hence
"scientists prefer" metric while "laymen prefer" English units.

I don't buy it.

The only laymen that prefer Imperial units are the ones that never learned
to work in metric units.

The fraction system used with inches is a base two system so you can only
divide easily by powers of two. Decimal metric units at least makes it
easy to divide by 5's as well. What's 1/5 of a cm? 2 mm. But what is 1/5
of an inch? Can't really do it easily.

The bottom line is that both systems have special cases that are easier
than the other systems, but that in real life, not much happens in those
special cases. You have something 6 3/16" or 15.7 cm and you want to
divide it in 4ths and it's not trivial in either system. But at lest, with
the metric system, you can pick up in $5 calculator and do the math
directly since all our calculators work in decimals. But in imperial
units, you have to play all games to figure it out like converting to
decimal, then doing the math, and then trying to convert back to fractions,
or dividing the inches and the fractions separately and adding the results.

Decimal is a horrible base for practical use.

The only thing worse is the mixed-base system of Imperial units. Base 12
for inches, base 2 for fractions of an inch, base 10 for feet.

Metric units are better because there is no conversions required since we
do all our math in the decimal system already. Or the only conversion is
learning where to put the decimal point instead of remembering what to
divide by (12, or 4, or 8, etc).

If all you have to do is measure something, any system would work fine.
You could put random marks on a tape and label them TOM, DICK, and JANE,
then make it the standard and it would work fine. You just tell people the
desk is 3 marks past JANE wide. But when you have to do math, like adding
two numbers, or subtracting, or dividing a length into equal parts, then
metric is a clear win because it's directly compatible with the decimal
system we use for doing math with.

In most shop work, we learn how to do the math problems without doing math.
Instead of measuring the width and subtracting another width, we just use
the work piece like a slide rule and measure the first dimension in one
direction, and the second dimension in the other direction so we don't have
to do the math. As long as all work can be done that way, then we could
just as easily use the TOM, DICK, and JANE ruler. But the minute you have
to do math, Imperial sucks, and metric wins big time.

The ancient Babylonians
had it right - base 60 contains all the small primes, making all
common fractions readily available.

Base 60 would suck big time. You would have 60 symbols to memorize and the
multiplication table and addition table you would have to memorize would
have 3600 locations in it. The kids might get it all memorized and learn
how to add by 6th grade if you were lucky.

Base 12 would probably be better than base 10, but again, the cost of
converting the world to a base 12 system would be a real bitch.

But pi is the optimal base from a theoretical perspective -- it
optimizes the tradeoff between how many symbols exist and how many
symbols are needed to express a given value -- but nobody uses it
since its difficult to represent integers... And binary is natural on
computers...

- Daniel

Awesome response Doctor Daniel! I must put this in my reference material
for sure. Makes sense to me!

I wasn't as impressed with the logic.

Wayne

--
Curt Welch http://CurtWelch.Com/
curt@xxxxxxxx http://NewsReader.Com/
.

Relevant Pages

  • RE: Just a brain teaser (for the math guys)
    ... You’ll notice that I divide all ... When I went back and did it manually by numbers alone, in other words, just ... math genius here might tell why the answers are different. ... know that this is more of a math question and has less to do with Excel. ...
    (microsoft.public.excel.newusers)
  • Re: Just a brain teaser (for the math guys)
    ... "greenusmarine53" wrote: ... to check the math, I counted all of the people who completed the requirement ... So 50% of the employees in each work section completed the requirement. ... You’ll notice that I divide all ...
    (microsoft.public.excel.newusers)
  • Just a brain teaser (for the math guys)
    ... Okay, so I've managed, with the help of those here, to make a pretty ... You’ll notice that I divide all ... math genius here might tell why the answers are different. ... know that this is more of a math question and has less to do with Excel. ...
    (microsoft.public.excel.newusers)
  • Re: Excel is unusable for finance or accounting.
    ... The problem isn't really in the floating-point representation of numbers, per se, or the ... Could Exel do floating-point ... The processor can't do decimal math natively, so it'd have to be done in software, ... handle simple fractions and decimals (see thread 'Glaring Excel calc ...
    (microsoft.public.excel)
  • Re: [PATCH 1/3]: Replace kernel/timeconst.pl with kernel/timeconst.sh
    ... test for 32 bit math in the script that cares, and this worked in Red Hat ... A larger $HZ results in a smaller $SHIFT by the number of bits needed to store ... The bottom $SHIFT many bits are all zero before the divide. ...
    (Linux-Kernel)