Re: [Phys-l] Significant figures -- again

[John Denker <jsd@av8n.com>]

On 03/13/2012 11:02 AM, Paul Nord wrote:
> The traditional method of writing a number to indicate the error or
> significant figures would be to include the trailing zeros.  This
> indicates the precision in a manner consistent with professional
> publications.  For scientists it communicates the precision clearly.

No, it doesn't.  Among the many many failings of sig figs, they
do not communicate the precision clearly.

Professionals do not communicate using sig figs.  Rather than
taking my word for it, I suggest you look at the way the
professional metrologists at NIST communicate their results.
They do not do it by means of trailing zeros.  A few minutes
ago I cited this example:
  http://physics.nist.gov/cgi-bin/cuu/Value?e

I also insist that 3 ± 0.1 is grammatical as well as unambiguous.
In accordance with the axioms of the Hindu-Arabic numeral system,
3.0 is by definition equal to 30/10 which is the same as 3.  It
is entirely grammatical (and more concise) to write 3 instead of
3.0.

The only reason I sometimes write 0.1 instead of simply .1 is for 
legibility.  Studies have shown that a leading decimal point is 
sometimes overlooked, so (depending on font size, context and other
details) the leading zero might help.

In any case, the fact remains:  When I write a number, the number of 
digits does not imply anything about the uncertainty, significance,
tolerance, or anything else.

Relying on the trailing zeros to communicate uncertainty etc. is
a really bad practice, and should not be condoned.