[OS X TeX] Writing a book

[Martin Berggren]

This discussion made me remember an exchange from 2007 in the Usenet group comp.text.tex under the subject “Dealing with publisher". The post below is written by Peter Flynn. Not so encouraging for the use of LaTeX sources in the publishing business:

============================================ 
blm... at myrealbox.com wrote:
> In article <1178139029.... at y80g2000hsf.googlegroups.com>,
> Luis Rivera  <jlr... at gmail.com> wrote:
>> The problem seems to go back to DEK himself, who, if memory doesn't
>> fail, started working on TeX precisely because he didn't like the
>> proofs sent by his publisher. What strikes me as remarkable in this
>> thread is the sense that some publishers still seem to have little or
>> no interest into processing LaTeX source or using TeX as their
>> typesetter.
None whatsoever. They've been frightened off [La]TeX forever by the 
grotesque rubbish they were sent when they first tried it.

>> One problem I sense as a publisher is perhaps the counterpart of some
>> horror stories told here: authors with LaTeX experience have a natural
>> tendency to develop their own packages (extra command definitions,
>> shortcut macros, BibTeX and Makeindex styles, etc.), which later on
>> may cause trouble for the publisher.
Exactly. I identified this as the major cause of dissonance between 
TeXophiles and publishers in my paper at the TUG meeting in Delaware:

    ...the training technique known as ‘sitting by Daisy’continues to
    predominate.

    This comes to a head not in the production of private or internal
    documents but in the submission of LaTeX files to publishers. Although
    many of them have used LaTeX for years, most would claim that it is
    only because of pressure from authors who insist on supplying it. Few
    publishers now employ LaTeX experts to undertake the syntactical
    correction of LaTeX files necessitated by the authors’
    misunderstandings, and some even find it more profitable to print out
    an author’s document with errors and have the entire thing retyped
    from scratch in Word by keyboarding companies in the Far East.

    Most of the publishers’ misconceptions (‘LaTeX has only one font’,
    ‘LaTeX can’t do graphics’, ‘LaTeX is only for mathematics’, et cetera
    ad nauseam) stem from their experience of authors’ own misconceptions
    and lack of training. [TUGboat 22:3]

>> That's why here at work we encourage contributors to send their
>> papers in some reasonably open format (RTF or ODF lately), so that
>> we can work on conversions from those files;
> 
> As someone who contributed one of those horror stories, I wondered
> about publishers and whether they also had tales to tell.  I guess so!
> 
> About having people send you things in an open format ....  Maybe
> I just don't know enough about RTF, but I had the idea that it wasn't
> really capable of expressing the kind of logical markup that LaTeX
> is so good at, and so going this route would reduce everything to
> WYSIAYG.  I guess for simple documents that could be okay, but for
> something with a lot of references, or an index, it seems like the
> wrong approach.  
It's actually cheaper to rekey the whole document from scratch, or to 
hand-edit RTF or Word into shape, than to edit out the errors and 
misunderstandings in authors' [La]TeX sources.

> Maybe in that case you ask for things in ODF, about which I know even
> less but sounds potentially more capable of including internal
> structure.
Publishers couldn't give a tinker's fart about structure. Their job is 
to get the book out the door and onto the shelves, and so long as they 
think it looks right, they're satisfied. The typesetting houses they 
employ are sometimes asked to return XML with the PDFs or film they 
create, so that there is a marginally well-formed document in the 
archives representing what was published, but it hasn't yet dawned on 
the publishers that there is added value to be got from having the 
document in a sensible format.

///Peter



Martin Berggren
-------------------------------------------------------------------
Department of Computing Science, 
UMIT Research Lab
Umeå Universitet
Campustorget 5, S-901 87 Umeå, Sweden. Ph: +46-70-732 8111
http://www.cs.umu.se/~martinb, Martin.Berggren at cs.umu.se