>>>>> "RC" == Robert J Chassell <[EMAIL PROTECTED]> writes:
RC> However, I am trying to inspire someone to write a
RC> Texinfo to DocBook converter. DocBook has a
RC> docbook-to-man tool that might then do the job.
I've been thinking similar thoughts lately.
I think it is inevitable for emergence of the following
texinfo tools in the near future:
* texinfo.dtd (or texinfo.xsd if we use XML Schema rather
than DTD) capable of representing all the features of
texinfo.
* `texi2xml' program that converts texinfo to XML using
the tags specified in texinfo.dtd/texinfo.xsd.
Of course makeinfo could be added with XML support as well.
texi2xml should be much easier to write then texi2html
since there should be less `semantic gap' between
texinfo and texinfo.dtd than between texinfo and
html.dtd (i.e., HTML 4.0 spec).
* texinfo2docbook.xsl stylesheet to convert texinfo.dtd
conformant XML documents to DocBook conformant
documents.
RC> * second, remind people to create more accessible
RC> documentation. I fear that some writers who
RC> read most of their documentation on a high
RC> resolution monitor forget that documentation
RC> should be listenable by a blind person or by
RC> someone driving a car, or readable by someone
RC> working remotely over a slow modem or using a
RC> character-only terminal. Documentation should
RC> be accessible by anyone handicapped by personal
RC> or situational circumstance. At present, both
RC> DocBook and Texinfo provide for accessibility
RC> but some writers may inadvertently create
RC> inaccessible documents, as so many HTML and
RC> LaTeX authors have.
A few weeks ago, the topic of file format came up for the
FAQ file for GNU Emacs on windows platform (AKA
ntemacs). Someone suggested using texinfo. I gave a
resounding endorsement to that idea. However, there was a
lot of friction from the members of the list on the use of
texinfo. I believe the new maintainer of the FAQ file uses
HTML with some sort of preprocessor type of processing.
I believe a big reason for resistance to texinfo by many of
the people is simple ignorance. I have no doubt that if
they see the quality of HTML that makeinfo/texi2html creates
(not to mention the PDF files texi2pdf generats, and of
course my favorite, GNU info format), they will change their
mind.
Anyways, I was thinking if someone should write
Texinfo::Element and Texinfo::Parser perl modules similar to
HTML::Element and HTML::Parser modules that already exist.
These HTML::* modules parse the HTML files and provide the
perl programmer with the parsed tree of the HTML document as
an object along with many functions to operate on the
object. It makes HTML file processing really convenient.
In fact, html2texi.pl
<http://www.cs.washington.edu/homes/mernst/software/>
perl script uses these two classes.
If we had Texinfo::Element/Texinfo::Parser modules, then I
believe it would expedite the development of texi2xml and
other texinfo related programs.
Just my $0.02.
