[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

the nature of \sin




A few days back:

Justin(?)> have `log' and `sin', etc typeset in many different ways.

Here I sort of agree.

Alan> The base font (the one used for \log etc) will be a text
Alan> font, and will be usable in horizontal mode for setting
Alan> text matter.

But this and related comments bother me.

Suppose that we are using Palatino for prose and cm for math.
(Palatino and thousands of other fine prose fonts lack their own
custom designed math system.) Then two candidates for \sin are
Palatino roman and cmr. To forbid the use of either would be
silly.  Experience makes me prefer cmr even though, in a Palatino
style you will not get away with using cmr in prose. 
But I am sure some of you would argue for \sin in Palatino roman.


It has been asserted that \sin is a text word.  That is a 
cardinal sin because we all know the translation of \sin into English
is sine.  

Then what on earth is \sin?  The essential nature of \sin (or at
least the one that interests me here) is that it is a
*mathematical word*.  By this I mean a sequence of mathematical
characters of analogous type (I call them letters) that
*together* represent one indivisible mathematical entity. (That
the entity is a TeX operator here is irrelevant).

Now there are some reasonable rules for mathematical words:
unlike prose words they cannot be hyphenated, but their
letters should kern much as prose letters kern.

The font choice of the letters of a mathematical word may cause 
some dispute, but in any event \sin must not be interpreted as
a sequence of three math symbols, say the product of
s and i and n.  Thus the use of cmmi10 is clearly ruled out.

My personal choice for \sin is usually cmr10 as \fam0.  Now even
if I were to adopt Palatino Roman for \sin etc., I might well find 
good reason to maintain  cmr10 as \fam0 and rather wheel in
Palatino roman as \fam10...

Let us not make unnecessary rules; fashions will and should
change. TeX will have to yield to fashion and so will need all the
flexibility we can provide.

L. Siebenmann <matups.matups.fr>