atw: Re: Notes on Word Style Sheets

  • From: Stuart Burnfield <slb@xxxxxxxxxxxxxx>
  • To: austechwriter@xxxxxxxxxxxxx
  • Date: Mon, 13 Feb 2017 22:37:57 +0800

James, I assume you're talking about Word documents put forward as work samples by TWs we might be thinking of hiring. The usual motivation for examining the use of styles is to see if the writer knows how to produce tech docs that are usable, consistent and maintainable.

You should be able to tell from looking at the style sheets whether the document is set up to be updated by different people, including people who are neither TWs or Word experts, converted readily to another source format, and output in some medium not dreamt of in 2017.

The sort of document you describe sounds like a scientific paper or novel printed on paper in 1980, never to be revised.

Stuart

On 13/02/2017 9:06 PM, James Hunt wrote:


Some contributors have noted that they like to examine the style sheets of
Microsoft Word documents that are presented to them. The question is, what,
exactly, can you learn about a writer from such an exercise?

Microsoft Word styles apply typographical rules, which must be specified by
the author, to paragraphs of text. Microsoft Word has no typographical rules
built into it. All typographical treatments must be specified by the author,
and most authors are not typographers or book designers.

In contrast, the TeX typesetting program, widely used in scientific publishing,
contains hundreds of typographical rules, and applies them in response to a
small number of macro commands. If you don't want to know how it's done,
you don't need to know. In scientific publishing, people quite ignorant of
typographical niceties, such as electrical engineers and theoretical economists,
can produce elegantly laid-out scientific papers suitable for publication without
much, if any, further editorial intervention. If your software knows about
typography, then you don't have to.

The style sheet approach to document design is remarkably inefficient: even
a rough approximation of a typeset document can involve the creation of a few
hundred styles. Remembering them all, and training other writers in using them,
are major problems, often discussed in writing groups.

But why should it be necessary for a writer to remember all these typographical
rules, and be judged on the imperfections of their memory and inadequacy of
the software in use? Remembering and applying hundreds or thousands of
rules is a suitable task for typesetting software.

Expending a great deal of effort on the appearance of a Microsoft Word document
is an exercise in futility: since there is no method available to optimize the
grey-scale (appearance) of individual paragraphs (those interested can
Google on "Knuth-Plass algorithm"), all Word documents will always have an
air of mediocrity about them.

If you want to find out something about a writer by examining a Microsoft
Word document, it might be better to look for other inadequacies.

Not changing the default margins of a document results in very long lines of text,
which can be difficult to read. A common design rule is that a line of text should
be no more than "two and a half alphabets" long, which means that a line should
contain 65 to 80 letters and spaces (depending on the actual text). Line of this
length can be read comfortably, and the start of the next line can usually be
found easily.

My pet dislike is the use of Times Roman and other system fonts in body text.
Times Roman was designed for legibility in short lines on poor quality newsprint,
and in long lines it looks cold and unfriendly. If the writer chose another font,
I would regard that as a plus.

There are rules for specifying the sizes of headings and subheadings in relation
to body text size. These sizes are usually set by reference to the "typographical
scale", but most Word users don't bother about it, or, I suspect, even know that
it exists. As a result, sizes of headings and subheadings appear random.

A surprising number of writers do not bother to set up right and left headers
and footers, even in their own CVs. Usually, the right-hand (odd) page numbers
are correctly placed, but the left-hand (even) page numbers are often incorrectly
placed and disappear into the  gutter between the pages.

Are there any appendices, and are they lettered and numbered correctly? This
can be a clumsy business, and most Word users solve the problem by avoiding it
and not using appendices - it's Just Too Hard.

Does the document have an index? Even quite short documents can benefit from
an index, but indices seem to be rare these days.

Examining the style sheet of a Word document won't actually tell you much. The
writer may show some knowledge of the elements of typography, and there may
be clues about neat work habits, but that's about it.

JH


Other related posts: