units

Eric Greisen egreisen at valen.cv.nrao.edu
Fri Apr 2 10:12:30 EST 1999


Tim Pearson writes:
 > On Wed, 31 Mar 1999, Francois Ochsenbein wrote:
 > 
 > >    - the Ohm sould be written with a capitalized O,
 > >          to be consistent with its origin (physicist's name)
 > 
 > I disagree with this (trivial) point. I think that it is the normal
 > practice of publishers in the english-speaking world *not* to
 > capitalize the names of units, even those derived from proper names,
 > such as "newton", "ohm", "joule", "jansky", "ampère", "ångström", etc.
 > This may be a recommendation of the various international unions - I
 > am not sure - but it is certainly advocated by, for example, the Royal
 > Society (London).
 > 
 > This follows normal english practice, in which proper names are
 > capitalized, but not words derived from proper names, such as
 > "sandwich", "boycott", etc.

     We are here not talking about spelled out names generally but
about abreviations for units.  The Ohm and Angstrom ones are special
cases since A is amperes and a is ato and O likes too much like 0, so
we are spelling them out as "abreviations".  I note that the IAU style
manual refers to "janskys" as "Jy" which is what all radio astronomers
use.  I really don't care whether Angstrom or Ohm are capitalized, and
perhaps since we are spelling these out we should leave them without
capitals as the IAU style manual does with spelled out words and as
both units references actually do.

 > 
 > However, tastes differ; and there are some circumstances in which it
 > is appropriate to capitalize the name of a unit, such as at the
 > beginning of a sentence.  So software should be tolerant of variant
 > capitalization (i.e., it should be case-insensitive).

     If we become case insensitive, then m milli and M mega and A
ampere and a ato become hard to tell apart.  I am no fan of case
sensitivity, but it is unfortunately necessary here.  It is of course
true that a clever parser could have more lenient rules than the
standard - that is normal and perhaps even required of FITS readers in
general.

Eric Greisen



More information about the fitsbits mailing list