Re: Requiring use of DocBook; LinuxDoc

[Gary Lawrence Murphy <garym@canada.com>]

>>>>> "G" == Gregory Leblanc <GLeblanc@cu-portland.edu> writes:

    G> ...  I don't know anything about writing
    G> books, but Gary Lawrence Murphy does, so I'll ask "Do authors
    G> get a say in exactly what gets published?"

Ah, er, well, (ahem), uh, no, not really.

You do if you write for a smaller publisher or if you are a superstar,
or if you are like my friend Garth who just so happened to be the
head of the art dept for the publisher who did his novel, and the
best friend of the senior editor.  Garth even designed the grocery
store cardboard stand that sold his book.

Large publishers will tell you that smaller publishers are small
because they do not understand editorial style, layout and correct
typography.  Smaller publishers will disagree, of course, but sales
figures suggest large publishers are doing *something* right.  

It's a complicated question.  Part of it is legal, part of it is
marketing perceptions and the style-d'jour of trade publishing, but
the much larger part is indeed science.

Those weened on MsWord know squat about layout and typography, and I
expect the younger readers of this email are thinking "he's nuts! it's
what you *say* that counts".  

Sorry to disappoint, lads, but, in the 400+ years of the Gutenberg
Universe, we know this is not the case; just as the slight bent and
backlight of your TV twists what you think that box says, layout,
typography and design twist an author's words to their public. 

The medium is a powerful message, one that can make or break a book.
There is good reason ORA puts the colophon at the end of their books:
They are rightly proud of their science.  Just as a Prime Minister
lets others design their stage presence, even when they wouldn't let
these people near the design of foreign or fiscal policy, a
professional tech writer stays with what they do best, and leaves
managing media to people who do that 2000 hours of each working year.

An author only recommends the semantic meanings of things like tips,
cautions, emphasis, screenshots and ... oops, sorry, I didn't mean to
start repeating that DTD snippet for <para> --- what all these
actually mean in ink on a page comes as a complete surprise, and is
often massaged to fit the typographical style of the specific line of
books (e.g. all 'Unleashed' books have similar typography for branding
reasons)

In pre-DocBook days, our books were written with arcane text codes to
flag these conceptual regions; we used MsWord (and now StarOffice)
only because it allowed editors to hide comments in the text and to
track revisions (on re-reading the twelth revision, you don't want to
sift text looking for the three lines that changed) -- the text was
otherwise pure ascii (with 2-char named macro codes) because the
typesetters and layout artists must change everything to fit their own
technology and their design ideas.

Today, with DocBook, comments are comments, revisions are revisions,
and there is no need to keep a 5 page "code breaker" chart taped to
the wall.  These codes were so awkward, so devoid of mnemonics,
standards and reference support, almost no one used them correctly,
and most often they were used inconsistently within the same book,
sometimes within the same chapter (usually mine).

The typesetters are also happier because a manuscript can be moved
quickly from their office back to the author for a quick fix and back
again to set the linotype (or whatever) without the labourious
translation of the manual codes.  At the moment, the layout and design
people and the non-tech editors are grumbling because I've obsolesced
all their favourite tools, but they'll get over it I'm sure.

-- 
Gary Lawrence Murphy <garym@linux.ca>: office voice/fax: 01 519 4222723
TCI - Business Innovations through Open Source : http://www.teledyn.com
Love Linux?  We need authors/reviewers - http://www.teledyn.com/authors



--  
To UNSUBSCRIBE, email to ldp-discuss-request@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org