Date: 2001-07-06 20:53:18
--- In boost_at_y..., "Paul A. Bristow" <pbristow_at_h...> wrote:
> > From: Deane_Yang_at_y... [mailto:Deane_Yang_at_y...]
> > 3) Transcendental functions
> > Any standard transcendental function, such as sin, cos, exp,
> > MUST take only unitless quantities and return unitless
> Have I missed something in that log and antilog (exp), sin &
sin-1 ... are
> so if you log(length) the units are somehow hidden (* unity
> seem quite right)
> But if you then antilog(log(length)) you really do get get back to
> that really does have unit of length.
> (Similarly for sin and its inverse).
> So the information about length _could_ usefully be retained
through the log
> (etc) process.
This is mathematically correct, but the question is whether it is of
any real use? As far as I know, it is never useful to take the exp
or log of a quantity that has units. You really can't do much with a
log(length) that you couldn't do much more easily with the
original length itself.
Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk