\centerline{\bf Book Reviews}
\smallskip
\noindent
`Raster Imaging and Digital Typography' is the
proceedings of the workshop held in Lausanne in October
last year, with some supplementary material from another
workshop on font design systems held in 1987 in
Sophia-Antipolis. The proceedings has two major sources of
interest: the general subject matter of digital type
design; and the fact that it was set with a \LaTeX\ style
which emulates the Cambridge University Press house style.
Several \TeX\ notables were involved in this production,
Richard Southall, Rick Furuta, Philippe Louarn and Victor
Ostromoukhov.

The content is quite varied, with papers from Jaques
Andr\'e, Roger Hersch, Neenie Billawala, Debra Adams, the
Nanards, Peter Karow (of Ikarus fame), and Richard
Southall. Although the \MF\ papers may have some greater
interest, all of the papers  have something to
offer, although, as usual, some seem so esoteric that it is
difficult to see applicability, even in the long term.
Some are therefore `research' reports, rather than reports
of application. 

Among the papers I found most interesting was Adams \&
Southall's {\sl Problems of font quality assessment\/}
which helps by establishing some terminology, but also
establishes some criteria for quality assessment. Bruno
Maag's {\sl Shape investigations with bitmapped
characters\/} takes the four letters H, n, b and v and
demonstrates the problems of realising these
letters effectively at 300\,dpi. The historically minded
will be fascinated by Ittai Joseph Tamari's {\sl
Digitization of Hebrew fonts\/}, which looks at the
evolution of printed Hebrew characters from 1475. A taste
of the future may be present in Jacques Andr\'e and Bruno
Borghi's {\sl Dynamic fonts} which starts where Knuth
stops, taking the notions implicit in the Punk font and
make it truly dynamic so that each character changes
when it is used. No character is ever the same twice.
So, some good stuff, and well worth a read.

But: I was left rather unhappy about the production
quality. First, the book is under-inked, a fact that
Richard Southall volunteered when I talked to him about it.
Basically, the fonts looked spindlier and more broken up
than they should have. Don't point a finger at Computer
Modern, since the book had been set in Times Roman. There
is a large number of illustrations, many of which were
pasted in. A sensible decision, but I would have prefered
a little more consistency in caption styles. I was
occasionally irritated by the editing, or perhaps the
lack of editing. It is not clear to me whether
`digitisation' and `digitalisation' 
are the same concept or not. The switch between `digitalisation'
and `digitalization' is understandable, if annoying. The
`s' `z' change is a common American\slash English change,
but to see Tamari's paper use the `s'-form in the table of
contents, but the `z'-form in the paper\dots\ The extreme
difficulty of balancing facing pages was evident, and one
paper in particular, Fahlander's, has an unforgivable
number of widows and orphans. I could have wished an
algorithm not to be split over  pages 101--2, especially
when elsewhere, page 60 has a 3\,cm band of white space at
the bottom, for no discernable reason. I suppose the
overfull box on page 164 is understandable (I think the
overfull rules should be large and ugly for a book style),
but the ?` on page 188 shrieked out at me. The footnote
spread over pages 218--9 is unfortunate. I would have
rewritten the text to get rid of that lonely line (that's
what an editor is for). Lastly, it was easy to identify
the hand of \LaTeX. The itemized lists in Peter Karow's
paper soon gave the game away.

Why do I make these petty criticisms? Because I expect
nothing but the best from the combination of \LaTeX,
Andr\'e, Southall, Furuta and Louarn. Obviously there is
great merit in the timely production of this book. But
perhaps with a little more care these blemishes could
have been avoided.

\begintt
@proceedings{AndreHersch1989,
editor={Jacques Andr\'e and Roger Hersch},
title={Raster Imaging and Digital 
Typography},
series={The Cambridge series on Electronic
Publishing},
year={1989},
publisher={Cambridge University Press}
isbn={0 521 37490 1}}
\endtt

\smallskip\noindent
`Hyphenation' by Ronald McIntosh is a quirky little book.
In a sense it is a justification of the work that he and
Computer Hyphenation have done in creating `the
Hyphenologist' software. It is full of fascinating
details about the lowly, much abused, hyphen. It traces
its origins and evolution and gives some delicious
examples of excerable computer hyphenation. Naturally,
the Hyphenologist does it right, but you would have to be
very percipient to guess just how Hyphenologist does do
it. Nevertheless, the set of rules for hyphenating in
English, pages 60--6,  do offer a starting place for the UK
\TeX\ Users Group Working Party on hyphenation. The book
has the flavour of enthusiasm, which makes it lots of fun
and an excellent read. 

To help sow some doubt and
division in the path of would-be hyphenators, consider
the following: hyphenation for the following eight, fairly
common words are taken from six English (not American)
disctionaries. While you might optimistically expect only
eight word divisions for the eight words, there are
actually 30. How does \TeX\ fare? Not too badly. Here's
the list:

\centerline{\vbox{\halign{#\hfill&\quad#\hfill\cr
\tt\char'134showhyphens&nearest matches\cr
abid-ing & exact (3)\cr
abo-li-tion & ab-o-li-tion (2)\cr
au-toma-ton & au-tom-a-ton (2)\cr
au-ton-omy & au-ton-o-my (2)\cr
au-to-bi-og-ra-phy & exact (2)\cr
il-licit & exact (1)  il-lic-it (3)\cr
il-log-i-cal  & exact (1)\cr
blud-geon & exact (2)\cr
}}}

\noindent Since the dictionaries cannot agree among
themselves, and in fact have some strange notions, like
hyphenating after a single letter, perhaps we are being a
bit too hard on \TeX's normal rather conservative
hyphenation.

\begintt
@book{McIntosh1990,
author={Ronald McIntosh},
title={Hyphenation},
publisher={Computer Hyphenation Ltd},
address={1 Campus Road, Bradford BD7 1HR},
isbn={1 872757 01 4}
price={\pounds4.95}}
\endtt

\author{\mwc}