\title{Malcolm's Gleanings}
\author[Malcolm Clark]{Malcolm Clark\\\texttt{cudax@uk.ac.warwick.csv}}
\begin{Article}
\subsection{Book review}
\noindent \emph{Computers and Typography} `compiled by' Rosemary
Sassoon, Intellect, Oxford, 1993, 208pp, ISBN 1-871516-23-4.

On the title page of this book, the compiler notes that the customary
words ``edited by'' were omitted at her request. She goes on to say that the
book is ``an example of what this is all in aid of---typographic excellence
in the computer age''. A bold claim, and an interesting inference that
typographic excellence is not the customary bedfellow of computer `mediated'
books. As if to underline the typographic excellence, the title page faces a
reproduction of a page from Aldus Manutius' \emph{Hypnerotomachia Poliphili}
of 1499.

Before looking at whether these claims are justified, what of the content?
For whom is the book intended? The cover suggests that it is invaluable for
``all concerned with teaching, design or who produce documents of all
kinds''.  In the preface, Sassoon suggests that the purpose is to bridge the
gap between the computer people and the typography people, but mainly to
raise the awareness of letterforms and layout, rather than to educate those
in the typographic world to the appropriate use of computers. It seeks
therefore to educate the computer user to a higher level of understanding of
`typography', however widely defined.


The book is organised into five parts: each part contains two or
three contributions. Part~1 covers `Spacing and layout': the contribution by
Gunnlaugur Se Briem, \emph{Introduction to text massage}, illustrates one
recurrent difficulty in the book---typographers and designers tend to be
aware only of the desk top publishing end of computer-assisted typography. His
recommendation to search and replace the ligatured letters is a shade
risible, though on the whole his advice is sound. But how practical is it to
look at each line ending to check hyphenation, lift the baseline to adjust
parentheses (sometimes), fiddle with the leading, and so on. Should we not be
looking for better models of line and page make up which recognise the
potential problems and solve them for us? Did every jobbing printer take this
much care?  James Hartley (\emph{The layout of
computer-based text}) examines some aspects of layout, starting with a
questionnaire, and going on to more general matters of the distribution of
space, and how it can be used to enhance  content. It is indeed true that the
use of white space is poorly appreciated by many: that increased use of white
space might make something more useful (and less wasteful) is not a concept
readily grasped, until some useful and pertinent examples like these are
thrust under people's noses.  Richard Southall's \emph{Presentation rules and
rules of composition in the formatting of complex text} is a highly literate
explanation which draws together the views of `traditional' typographers from
Moxon,  Fournier, Brun, De Vinne and Tschichold in order to show how their
`rules' may, or may not be applied in computer based composition systems.
Southall's in-depth knowledge of the working of \TeX\ and \LaTeX\ gives him a
unique position, and he develops some rather telling criticisms. His remarks
are more generally applicable and help to provide a useful set of criteria
for the assessment of computer based systems.

In Part 2, \emph{Typographic choices---Latin and other alphabets},
Ari Davidow examines some of the problems facing the typesetting of
Hebrew (\emph{Digital hebrew in a monolingual world}). This is an
anecdotal discussion, with a few interesting points. Its description
of computer software (almost all Macintosh based) is a snapshot
already out of date.  He is concerned solely with \emph{wysiwyg}
type input.  The observation that italic or slanted letter forms in
Hebrew are seldom satisfactory is worth hearing, although perhaps
diminished slightly by the illustration which was inserted upside
down.  Elwyn and Michael Blacker, \emph{Spoiled for choice}, have
little to say about other alphabets, but something to say about
computer typography, and, more important, about some of the
typographic choices that were made in creating this book.  At least
they believe that fine typography is attainable with computer
technology (albeit ``in the hands of a skilled designer with mastery
of the optical considerations'', faint encouragement for the \LaTeX\ 
enthusiast). And then they mention some of the design considerations
and problems they faced with the book.  They also comment on their use
of Bembo, with additional characters chosen from the expert font. The
use of the expert Bembo font is perplexing. Although chosen in part
because it has small capitals, these seem very thin and weedy to me,
as if they have been simply optically scaled from the `normal'
capital.  Examination of the book suggests that this, and their
``detailed checking of a proof'' may have been in the realm of good
intentions rather than solid achievement.

Part 3, \emph{More technical issues involved in type design} contains
two papers. The first is \emph{Some aspects of the effects of technology on
type design} by Mike Daines, which concentrates on the advantages which Peter
Karow's Ikarus system has had on digital type. He also brings in many of the
other potential tools available, especially those for the Macintosh.  
Another useful and considered paper by Richard Southall, \emph{Character
description techniques in type manufacture}, looks at two traditional
(i.e.~non-electronic) methods of the production of type, and two digital
techniques. The objective here is to identify the strengths and weaknesses of
the changing technologies, and the areas in which they are most (or least
appropriate). From his description of the processes involved, Southall
develops a `systematic view' of the manufacture of type. This has the merit
of providing a plausible model which we can use, and may give the basis for
some qualitative comparisons. Actually, by the end of this paper I am left
surprised that any acceptable typefaces were ever produced in any technology,
given the inherent problems at each stage.

The penultimate section, \emph{Lessons to be learned from the
history of typography} includes what I found to be one of the most
demanding papers, balanced by one which I found agreeably optimistic.
Fernand Baudin's \emph{Education in the making and shaping of
  written words} is a polemic, and although it traces an argument
going back to the days of Villon, and emphasises the importance of
handwriting (along the way consigning Marshall McLuhan to one of the
outer hells), I was left unclear how the final conclusion was derived
from the route and its many byways. But one useful point which is
reiterated is that the study of type must not be to the exclusion of
the study of space. A consensus is appearing.  Alan Marshall's
contribution, \emph{A typographer by any other name} came as a
welcome relief after this fundamentalism. He puts many of the problems
in perspective, and provides a thankfully optimistic conclusion, which
seems both balanced and realistic. He appreciates that all major
technological changes have their problems, that they start with a
period of emulation, and then innovation---there are repeated
examples in the printing industry. His observation that Orwell had
argued that the advent of Penguin's paperbacks all but signalled the
end of civilization as we know it helps place in perspective similar
contemporary claims of an apocalyptic nature.  Perhaps most telling,
he suggests that the pool of typographic knowledge is not limited, but
is expanding, encouraged by the technologies becoming available.

The last section, \emph{Research and the perception of type}, I found
difficult to integrate with the stated objectives of the text. Rosemary
Sassoon's own contribution, \emph{Through the eyes of a child---perception
and type design}, is an account of designing a typeface which would aid
children learning to read. Some of her observations on legibility are
interesting and intriguing, but they are hard to relate to computer in
general, or the more specific needs of computer aided publishing for a wide
market. For educationalists and teachers there is probably much here. Perhaps
not surprisingly, she also makes a plea for handwriting. The final paper is
daunting. Roger Watt, in \emph{The visual analysis of pages of text},
describes some experiments the visual perception of printed pages. He
analyses the same text with different inter word and inter line spacing. The
technique of analysis is claimed to have some reasonable closeness to the way
in which the human visual system works. In this analysis he identifies a
number of different perceived `structures', which he then relates to the
specifics of the text, like sentence breaks, rivers, words, inter line space
and so on. Perhaps contrary to received wisdom, he suggests that rivers may
be useful, as landmarks for navigation in a text. The result is the
conclusion that it should be possible to specify the `riveriness' and
`wordiness' desired (the visual effect), and then find the appropriate
word and line spacing. This seems a little radical, and the views of some
typographers on this could be interesting. Clearly it is appropriate to
attempt to bring in a more physiological appreciation of how type is
understood, rather than the typographers' often hand waving generalisations,
but this is not a straightforward paper. It is not clear how far the
conclusions may be generalised, either to english texts in general, or to
texts in other languages, where word length, and the distribution of
ascenders and descenders may be quite different. How it would generalise to
non-Latin texts is another mystery, or, as academics say ``more work needs to
be done''.

There is the feeling that some contributors view the changes as a shock to
the system, whilst others know it has all happened before, and that while
some things will deteriorate, new possibilities will arise, and things will
become possible about which we have not yet dreamed. The curious appeals
to handwriting as the basis of success have a very luddite ring to my ears.

One of the factors which worried me about the book was the extent to which it
achieved its aim as a ``model of good typographic practices''. Frankly, it
lacks consistency, and there are far too many typos. Perhaps the erratic
application of a house style is one thing, but mistakes are something more
serious.  These blemishes and inconsistencies highlight a notable omission
from the book: discussion on the real difference between markup systems and
those which demand that the text be dealt with interactively---i.e. a {\sl
wysiwyg} system. Many of the small problems of style can be more
easily resolved through markup systems. If the goal is to produce something
which is even, markup can ensure that the rules are carried out remorselessly
each time, while the use of more `flexible' systems actually requires much
more thought and discipline right through the book production. 

Unless this volume had been presented as some model for the
typographically unkempt, it would not be appropriate to pick up on the
small faults, but sadly, it does seem to fall into the same pit in
which it sees others.  On the other hand, the overall design of the
book is pleasing. Even the very ragged right works quite well
(especially when hyphenation is all but suppressed), and the wide
central margins are used quite intelligently as a location for
captions to figures. It is obvious that the book was designed `spread
by spread', allowing for what the reader actually sees. The interplay
of white space is attractive.

It is not a book for novices; nor is it a book for power users. It falls
awkwardly between a number of stools. Taken individually, the papers are
interesting, stimulating, and often provocative. But taken as a whole I just
cannot discern the linking thread, or the theme which binds it into more than
a book of loosely connected essays. It veers from the general, or at least
broad, to the very specific, from which something more can be inferred.
Placing these side by side gives a very uneven intellectual feel to the whole
thing. It feels as if Sassoon asked some of her friends to contribute
something to a book on typography and computers, without specifying the aim
too tightly, and lo! we have the results in our hands. The central concern of
the book still worries me. Sassoon says that she hopes people will ``never
again be satisfied with second best''. Elsewhere in the book are appeals to
``fine typography''. I would have preferred to see an appeal to ``fitness for
purpose''. 

This review is based on one which appears in the \emph{Information design
journal}, vol 7, no 2, 1993, p161--6


\subsection{Information design journal}

One of the curses of the (\La)\TeX{} world is that many proponents
become infected with a thirst for matters typographic. It's an odd
affliction, since many of the victims have a scientific\slash
technical background, and the way education seems to be set up in many
countries is based on the belief that science and technology are
antithetical to anything aesthetic. And typography is largely an
aesthetic medium---or is presented as such. How do we acquire
knowledge and satisfy the hunger of our desire? There are a few books
around (in my view one of the best is Ruari McLean's
\emph{Typography}), but precious few journals. A few designerly
magazines exist (I like XYZ) but they do tend to be a little elitist
and introspective. What is there for those of us accustomed to reading
`academic' journals. I've yet to see a copy of \emph{Visible
  Language}, although Knuth has published there from time to time.
I've at last found something interesting, appropriate and local --
\emph{Information design journal}. It's not really just typography,
but there is much in it which is typographical.

The \emph{call for papers} describes the readership as
multidisciplinary and that contributions are welcomed on a range of
topics related to the communication of information of social,
technical and educational significance. Looking over the last four
issues, I note an interest in forms design (both questionnaires and
bills: this is also one of my interests---it fascinates me that it is
so difficult to design satisfactory forms), in information signing
(like directions, maps), in information symbols (like those ISO
symbols for almost anything, most of which I find odd and misleading
-- this is quite distressing for icon based computer systems\dots).
There also seems to be a wish to test comparisons---in other words,
to test hypotheses rather than make hand waving generalisations. But
there are other articles which aim to convince by qualitative
argument.

 The range of papers in each issue is broad too; not just in content, but also
in style. In a sense each issue becomes more informal as you read through it.
The key articles are refereed, as one would hope, but there are reviews of one
sort or another. Somehow it achieves a pleasant balance between rigour and
informality. I therefore commend it to you as a useful journal to read and
browse through. For more information, contact Fred Eade, Idj subscriptions, 
PO Box 1978, Gerrards Cross, Bucks, SL9 9BT.

\section{Nonsense}
The major event in the \TeX\ world over the last few weeks (nay,
months) must be the test release of \LaTeXe. To the surprise of many,
this arrived in December, just in time to disrupt family Christmases
throughout the world. Good timing.  Since it was truly a test release,
it did not have all the bits that we have been led to expect in the
\emph{Companion}. In passing, printed and bound copies of the \emph{\LaTeX\
Companion} are stated to exist. Frank Mittelbach says he has one (but
then, he would\dots). I wouldn't have thought he needed one, unlike
the rest of us.  It seems to have been relatively painless to install,
from the messages which flitted around, although running it gives you
even more file name extensions to contend with---just when you
thought you had come to grips with the profligacy of \LaTeX\ in
creating extra files for itself!

It's a relief to see something substantive like this out for use. If
there are worries though, it must be whether this will distract
attention from the serious matter in hand---\LaTeX3.  On the other
hand, it will soften us up a little, first by accustoming us to
regular upgrades/updates (just like Word for Windows!), but more
importantly ensuring that the communications channels work
consistently. To a large extent this is going to be software
distributed and supported electronically. One of the features I like
is that queries will not be entertained if you are using an `obselete'
version of \LaTeXe.

\newcommand{\LATeX}{L\kern-0.3em\raise0.6ex\hbox{A}\kern-0.15em\TeX}
\newcommand{\LslaTeX}{\Lsla\kern-0.15em\TeX}
\newcommand{\Lsla}{L\kern-0.3em\raise0.6ex\hbox{\small\sl A}}

I'm becoming confused how I should write \LaTeX!
Just the logo---mostly I can handle \LaTeX\ itself.
If I look through TTN and \TUB, I can find quite
a few instances where the preferred form is given
as \LATeX, or even \LslaTeX\---this latter form is
especially prevalent when you see it written as (\Lsla)\TeX.
Maybe consistency will return when the results of
the A-in-\LaTeX\ competition are announced.


Is the NTS project poised to take over the world?
News from the NTS project is always to be treasured,
since it has all the hallmarks of an inner cabal
composed of a secret elite: Phil Taylor's article in
\TUB\ revealed that besides trying singlehandedly to
resurrect the economies of eastern europe, it is proposing
to start to issue a `canonical \TeX\ kit' (you can
always tell when Phil is involved: `canonical' sprouts
everywhere!). This has the laudable aim to
identify what a standard (`canonical') implementation
should contain, and to liaise with developers and
implementors to ensure that this is distributed with
each \TeX\ implementation. Praiseworthy and necessary
as this step is, I'm not myself clear
how this relates to the desire to develop a new
typesetting system. In the same issue of \TUB,
Nelson Beebe encourages vendors to include
his {\tt bibclean} utilities with each distribution.
Will this be part of the NTS canon too? 

Of course there is more. The simple existence of a piece
of software does not mean that it has all the same
attributes when run on different platforms. I am minded of
Makeindex, which exists in some different incarnations
with differing capabilities in terms of size of index it
can handle. Since the aim of the canon is to ease the
transfer of documents from site to site, the support software
must be capable of handling the same sizes of problems too.
Will the project be taking on this role of guardian of
compatibility?

I suspect that underlying this is another agenda altogether.
Identify the project to implementors and developers
as the (self-selected) body in the \TeX\ world which somehow authorises
the suitability of \TeX-related applications. In this
way it makes itself the legitimate heir to Knuth as
far as this sort of software development is concerned.
It's a strategy that might work.

You may wonder how it leaves the user groups who are
already starting to produce this sort of `\TeX\ kit'.
I do.

\TUB\ readers will have noted that the journal is pretty well on
schedule. My December issue arrived at the beginning of the year. For
many people this is a welcome sign. There was a time when we felt
lucky to get \TUB\ within about 6 months of its hypothetical
publication date (even then, better than EP-odd!). There has been a
price to pay. Frequency is still a little problematic (two issues this
year came out very close together, but you could just say that one
was late and its successor on time), but more significant, one issue,
the conference proceedings, is virtually half of the total mass---in
other words, three `normal' issues constitute about the same amount of
verbal as the conference. Last year ran to about 450 pages: in 1989,
it was over 750. Even arguing that TTN is removing some `mass', then
the volume is still slimmer. We could also argue that the multiplicity
of `competing' journals has taken some articles away (but a cursory
glance of the Dutch group's MAPS will demonstrate that much is just
recycled between journals). Is there a worrying trend in motion: thin
and timely?

\end{Article}