[texhax] TeX vs LaTeX primitives

[William F Hammond]

Victor Ivrii <vivrii at gmail.com> writes:

> On Mon, Dec 30, 2013 at 2:41 PM, William F Hammond <hmwlfsr at yahoo.com> wrote:
>
>     Victor Ivrii <vivrii at gmail.com> writes:
>    
>     > ... I hoped that there could be a kind of script like LaTeX validator. ...
>    
>     Freestanding LaTeX validation for authors is part of my
>     vision for LaTeX profiles,
>    
>          http://www.tug.org/tugboat/tb31-2/tb98hammond.pdf
>
> Yes, it would be great except that most likely almost
> nobody would use it. While w3 validators have been
> available for a long time, an absolute majority of web
> pages (including those made by professional web
> developers) produce tens if not hundreds of errors

Yes.  So ...

For the far future one idea would be that a LaTeX document
beginning with \documenttype instead of \documentclass would
indicate the use of formally profiled LaTeX.  In that case
the latex engine could pass the document to something like a
perl script (or an internal lua script?) that executes the
processing wanted by the user, possibly workplace localized,
as indicated by a first line like

      %!my-processing-choice-designation(s),

for the particular profile.  The first stage of that
processing would always be validation against the formal
profile (commands used and relative context for command
usage).

The author's advantage in doing this is:

  * bullet-proof processing for a valid document

  * multiple end formats such as pdf, html (guaranteed
       correct), ebook (guaranteed correct), ...

  * source need not be customized for a particular journal
      since the journal can reliably process a valid
      document to suit its own purposes

                              -- Bill


Email: hmwlfsr at yahoo.com
       gellmu at gmail.com
https://www.facebook.com/william.f.hammond
http://www.albany.edu/~hammond/