Sebastian Rahtz wrote:
> in my personal view, no current authoring tools except LaTeX are of
> any real use (except to capture plain text); and unfortunately, LaTeX
> is almost totally discredited by the poor usage of its
> practitioners. yes yes yes there are millions of great LaTeX users out
> there, but what actually hits the system here is almost always
> diseased in some way.

This seems to be a much more down-to-earth problem, and much easier to
solve (maybe at least partially in future releases of LaTeX2e) than the
lofty goals of LaTeX3.

My basic claim is that base LaTeX is too narrowly defined, and therefore
causes a lot of the document exchange problems which seem to occur
frequently not only between author and publisher, but also between
different authors.  In particular:

- The set of tools which come with the core LaTeX distribution is too
  small for serious work.  Extensions exist (such as the quasi-standard
  AMS math package), but are not sufficiently promoted by third parties
  (some publishers' author packages re-implement subsets of amsmath in
  incompatible ways).

  Solution: Move the best-of-breed third-party packages into the
  LaTeX core distribution.  For mathematical publishing the choice
  seems to be rather obvious, in Physics, as I understand, things are
  much more in a mess (ReVTeX...) and may need serious revision.
  In other subject areas I don't really know, but this could and should
  be sorted out.

- The biggest mess of all is the front matter.  I remember this has
  been discussed before on this list.  Did anything ever result from
  that discussion?

  I guess a solution could be found within LaTeX2e and shipped as
default
  with future updates.  This certainly very nontrivial exercise should
  at the very least achieve the following:

  o Define a comprehensive interface (key-val syntax???) for all
marginally
    common frontmatter features.

  o Specify a way in which individual publishers are allowed to extend
    the standard if necessary.

  o Provide backwards-compatibility to common existing front-matter
    conventions (default, AMS, Elsevier, possibly others), if necessary,
    as package options.

  o Be interfaced in such a way that the standard document classes
    can be easily retrofitted with the new conventions.  (Package
    which can be loaded on top of these classes, or, better, as a
package
    option.  Personally, I would prefer if such new conventions could
    be made the default - I know the issue about the standard classes
    being frozen - if a reasonable auto-detect of the new features
    is possible without breaking existing documents.)

  o Some nice front page defaults for "report" and "book" which could
    save authors in may cases from hacking with \titlepage.

- Lack of a good single source of documentation.  There is not a good
  book on how to author documents in LaTeX.  Lamport's book is not
  inclusive enough; the Companion, or Kopka-Daly, are pretty good if it
  comes to solving concrete problems, but don't offer much guidance for
  how to write documents that are intended for submission to publishers,
  or require easy exchange in loosely organized (academic)
collaborations.

  I.e., we need the definitive reference "LaTeX for Authors", preferably
  written by someone who understands the publishing process in all
stages,
  and specifically documenting the "official", "definitive" solutions
  to the first two problems above.

  A regular update schedule would not hurt, so maybe every new LaTeX
  could be shipped as full distribution (TeX Live...) on CD together
  with this printed or electronic manual?

Am I dreaming here?

Marcel