\pageno=22
\input texline
\centerline{\bf Another Book Review}
\medskip
\noindent
{\it Desktop Publishing Skills, James Felici
\& Ted Nace, 180pp, Addison-Wesley, {\sc isbn}:0 201 11537 9.}
\smallskip
\noindent
Perhaps the title may put you off. After all, `desktop publishing'
covers such a multitude of sins. But first look at the
sub-title: {\it A Primer for Typesetting with Computers
and Laser Printers}. This is a far more accurate description
(although it might sell fewer books). Felici \& Nace are really
talking about computerised typesetting, with low or
medium resolution output devices. 

The book itself was typeset on a Linotron 202, having
been prepared with MagnaType. This is a commercial
(and rather expensive) typesetting program, which will
run from {\sc at}-compatible micros. At the time
Felici \& Nace did their book MagnaType had no
preview facilities, although current versions of the software
do. Since many books on dtp seem to have been done
at home (possibly with potato cut-outs), this provides
a welcome alternative: it is typeset quality throughout.

The authors take some time before they delve into
software. They provide a clear and distinct account
of  the  vocabulary of traditional typesetting
(and later we realise they also have the vocabulary
of computing). At last the jargon that is so often thrown
at us: `signatures', `h\&j', `letterspacing', `tracking', 
`quad left'
and so on, is explained. Letterspacing and tracking are
rather interesting, since they are features not provided
by \TeX. It is interesting to speculate why. Perhaps
they were not used by Monotype Modern 8A, the model used
by Knuth for Computer Modern (that's {\it why} it is
called {\it Modern}: in fact, {\it Modern} doesn't
really imply anything recent --- the first {\it modern}
typeface was Bodoni's of 1775). 

Letterspacing refers
to a stretching out of the interletter spacing of a line
to make it fit better. As far as I can see, squeezing
the line does not count. This is commonly found in
newspapers (and found also in Felici \& Nace):
tracking is a more global concept. Again it involves the
interletter spacing, but it does so universally,
affecting all interletter spacing equally, everywhere.
Tracking is a typographic response to the way the eye
reacts to type in different sizes. As type grows in size,
the eye perceives the interletter spaces growing
faster than the letters themselves. To compensate,
the interletter spacing should be proportionately
tighter for larger sizes. I assume that the |tfm| files
associated with different design sizes help to compensate
for this, but of course you could introduce a more tightly
tracked font by giving it its own |tfm|. You might have
to alter the ligaturing too. A loosely tracked font
would probably not require ligaturing at all. Of course
you would not be able to alter the tracking in this way
for a font with only one design size (like the \PS\
fonts). 

Naturally there is some discussion on letter
forms. At last someone explains something about
`the rhythm of vertical strokes' and `stroke frequency'.
And the examples bring this out very well.
They also explain why there are 72.27\,points to the inch,
and not the apparently more rational 72. 
An America pica is
defined as equalling 0.166040\,inches. Presumably this
was to obtain an `exact' conversion which still gave
about 72 points to the inch. 72 points to the inch gives
an irrational conversion for picas to inches. Simple really.

Like Richard Southall, they too distinguish {\it font}
and {\it typeface}, describing a typeface as a set
of characters sharing a common design that distinguishes
that typeface from all other character sets, while a
font is the physical source from which the typeface is
generated ({\it cf\/} fountain --- a source or wellspring).
A daisywheel for example can correctly be described
as a font. The marks on the paper are not a font.

Besides explaining the typographic side they also
discuss the hardware --- especially the imaging
device (not just {\it laser} printers, but {\it led},
{\it lcs}, {\it ion-deposition} and the rest). And having
described the digital nature of the marking engine,
they  also discuss the digital nature of fonts.
They describe  outline and bit-mapped fonts, and
bring out their relative merits.

It should be clear that Felici \& Nace are not going
to favour one system over another, and this allows them
to look at a number of {\it page description languages}
reasonably objectively.

How does \TeX\ fare? Quite well. References to \TeX\ crop
up a few times, and an example occurs under `Professional
Composition Systems'. The example is actually from
\AmSTeX, but the principles are the same.

One of the achievements of Felici \& Nace is that
once they have explained something, it sounds rather
obvious. They make it all sound quite easy and straightforward.

In the last chapter they consider the assembly
of a dtp system, outlining several sets of criteria
which can be used to evaluate different systems,
in the light of perceived needs. This provides a set
of reasonably objective criteria. In the end though,
choice of an appropriate system depends to a large extent
on knowing what it is you want to do.

Right at the end of the book is a glossary of terms,
quite useful in its own right. The index is more useable
than some I have seen.

I have no hesitation in recommending this book to anyone
with any interest in getting more out of
computerised typesetting. It is easily the best book
I have yet seen on the subject, both in terms
of its treatment of the subject, and in the presentation
of the material. For once, form and content
enhance one another.

\rightline{\sl Malcolm W Clark}