Replied: Fri, 24 Mar 1995 15:08:00 +0000
Replied: "Michael Downes <MJD@MATH.AMS.ORG> R A Bailey <R.A.Bailey@qmw.ac.uk>, baskerville@tex.ac.uk"
Delivery-Date: 
Date: Fri, 24 Mar 1995 08:30:35 -0500 (EST)
From: Michael Downes <MJD@MATH.AMS.ORG>
Subject: Re: amslatex
In-reply-to: <16618.9503201129@galois.maths.qmw.ac.uk>
To: R A Bailey <R.A.Bailey@qmw.ac.uk>
Cc: MJD@MATH.AMS.ORG, baskerville@tex.ac.uk
Message-id: <796051835.339546.MJD@MATH.AMS.ORG>
MIME-version: 1.0
Content-type: TEXT/PLAIN; CHARSET=US-ASCII
Content-transfer-encoding: 7BIT
Mail-System-Version: <MultiNet-MM(369)+TOPSLIB(158)+PMDF(4.3)@MATH.AMS.ORG>

> I am writing a series of tutorials on doing Maths in LaTeX which are
> appearing in Baskerville, the magazine of the UK TeX Users Group.
> In No. 2 I introduced \emptyset. This provoked a query from a reader,
> to which I replied in No. 3. My reply was not very polite about the
> AMS, so I will send you the article right after this so that, if you want
> to, you can respond with a letter to the editor or a short note.

I'm glad that you brought this question to my attention because it seems
to be a simple misunderstanding that can be easily cleared up. :-) The
question is, to be precise,

  Why does "The LaTeX Companion" show a version of the symbol \emptyset
  in Table 8.7 (p 219) that is a circle with a diagonal line through it,
  in contradiction to the actual symbol that most LaTeX users will get
  from the \emptyset command, i.e. a slashed zero?

and

  Why does "The LaTeX Companion" show a slashed zero for the command
  \varnothing in Table 8.20, which purports to show the symbols provided
  in the extra math fonts of the AMSFonts distribution?

Your conjecture was that the AMS had done something to swap around the
usual symbol for \emptyset with the other symbol that is provided in the
msbm font. As it turns out, however, that conjecture is false; a search
of all the files in the AMSFonts and AMSLaTeX distributions reveals that
there is not even so much as a mention of \emptyset anywhere. And if you
look at the AMSFonts User's Guide, you'll see that the symbol shown in
the symbol tables for \varnothing is not the slashed-zero but the other
one.

The crucial point seems to be that "The LaTeX Companion" used Lucida
math fonts (if I recall correctly) and therefore the math symbols shown
in *all* of the Tables 8.3--8.20 are actually Lucida math symbols, and
*not* symbols from the fonts (cmsy, cmex, msam, msbm) that most LaTeX
users will see.

I therefore offer two counter-conjectures:

1. One of the authors of "The LaTeX Companion" is of like mind with you
that the AMS \varnothing symbol (plump circle with diagonal line) is
actually the preferred symbol of mathematicians for `empty set' and
explicitly swapped the definitions of \emptyset and \varnothing in the
preamble of "The LaTeX Companion".

2. The makers of the Lucida symbol fonts intentionally placed a
right-thinking empty-set symbol in the font position that is occupied in
the cmsy font by the slashed-zero symbol. And they put a slashed-zero
symbol in the font position that corresponds to the font position
in the msam font of the plump-circle-with-diagonal symbol.

I'd wager on the second conjecture. It would probably be a good idea if
you pointed out to the authors of "The LaTeX Companion" that their use
of the Lucida math fonts has led to some confusion and they ought to do
something for future reprintings to counteract the natural tendencies of
the Lucida fonts with respect to the two symbols in question :-)

Regards, Michael