Chronology Current Month Current Thread Current Date
[Year List] [Month List (current year)] [Date Index] [Thread Index] [Thread Prev] [Thread Next] [Date Prev] [Date Next]

Re: typesetting math and physics

First, let me say that anyone who pays attention to these kinds
of details will do a magnificent typesetting job, whatever the
particular decisions on such matters.

I've written two textbooks and typeset both myself with Plain TeX
(not LaTeX). Don Knuth is also a detail-oriented person and you
may find some help and inspiration in the TeXbook, even if you're
using LaTeX. When I was in grad school I had the pleasure of
taking a course from Knuth and I can assure you that he would
love to answer your questions. Unfortunately for you (and me),
he has recognized that this weakness of his keeps him from getting
work done so he no longer has an email address and he generally
prevents himself from being sidetracked by people like us.

The minimal version of Plain TeX doesn't even come with a bold
italic font, although such a font is available in nearly all
implementations. I use it only for math symbols in section
headings. Same for bold math symbols. Plain TeX has only one
set of lower-case Greek letters, call them italic if you like.
I've never heard of putting e (2.718...) or d (differential
operator) in plain type, though I understand the logic of it.
Certainly Plain TeX provides no convenient means of doing so.
You'd have to say {\rm d} or something like that. I usually put
a full space between a number and the unit following (e.g. 42 km),
but I use the twiddle symbol so TeX will never break a line
at that space: 42~km. The TeXbook says to use a thinspace,
but I think that puts 'em a little too close together. I
suppose you could define your own amount of space that's more
than a thinspace and less than a full space. I've never bothered.
I do try to use roman subscripts when it's a full word or a name
or an abbreviation for a person's name, though I'm not always
consistent with one-letter abbreviations.

Another comment on vectors: While they're prettier in bold,
I've concluded that this gets in the way of students learning
to use vector notation correctly in their own work. So in my
second book I just used arrows over light italic: {\vec v}.
Nobody has complained so far.

Dan


Textbook authors:

Do you typeset your own math or does some junior editor at the publisher do
it?

When writing up a lab in LaTeX a while ago I looked in Serway to see how to
typeset vectors and then I wrote the lab the same way, i.e., with bold but
straight-up roman letters for vectors. Now I read in my new
_A_Guide_to_LaTeX_ that

-------begin quote from page 142-------
1. Simple variables are represented by italic letters.
2. Vectors are written in bold face italic.
3. Tensors of 2nd order and matrices may appear in a sans serif font.
4. The special numbers e, i, pi, as well as the differential operator d,
are to be _written_in_an_upright_font_ to emphasize that they are not
variables.
5. A measurement consisting of a number plus a dimension is an indivisible
unit, with a smaller than normal space between them. The dimension is in
an upright font.
-------end quote from page 142-------

These guidelines are confirmed by the NIST Special Publication 811 "Guide
for the Use of the International System of Units," which adds "these rules
imply that a subscript or superscript on a quantity symbol is in roman type
if it is descriptive (for example, if it is a number or represents the name
of a person or a particle); but it is in italic type if it represents a
quantity, or is a variable." Also, mu should be in roman type when it is
the SI prefix for micro and should be italic when it is the magnetic moment.

Items 2 and 4 are the ones that pertain to my question. Vectors should be
italic, but most of the first-year physics texts on my shelf don't have
italic vectors (Sears and Zemansky and Young and Freeman 10e does). But no
one seems to put the differential operator d in an upright font. Why not?

Another question is: should the cross product sign itself be bold? Sears
and Zemansky and Young and Freeman 10e do, but they also boldface some of
the equals signs and the integral signs in vector equations. Other
publishers have bold cross products but non-bold equals signs. Are there
official conventions for these? There seems to be little consistency
between texts and sometimes even within a text.

Thanks,
Larry