HTML mode depends on physical, surface expression

[Robert J. Chassell]

HTML mode depends on the physical, surface expression, not the logical
expression.

The documentation for HTML mode says

    Text can be marked as <b>bold</b>, <i>italic</i>
    or <u>underlined</u> using the normal M-g or
    Edit/Text Properties/Face commands.

This means that `facemenu-set-bold' conveys erroneous information for
displays that use color or sound for different kinds of face rather
than weights.

The problem also occurs for

    facemenu-set-italic
    facemenu-set-bold-italic
    facemenu-set-underline

Texinfo provides the @strong and @emph commands, which convert to
<strong> and <em> when running `makeinfo --html test.texi'

The commands ought to be:

    facemenu-set-strong
    facemenu-set-emph
    facemenu-set-strong-emph
    facemenu-set-affirm

In order to handle the bug, in `emacspeak/lisp/voice-setup.el',
Emacspeak provides a `voice-lock-bold-personality' and other
personalities.  These personalities should only be necessary for
working with erroneous text.  Sadly, Emacspeak has now puts @strong
and @emph into a library that has been moved to its `obsolete/'
directory.  This shows how strongly situationally sighted modes have
come to dominate exposition.

(For the past 8 or 10 years I have used HTML-HELPER mode rather than
HTML mode.  However I do not wish to recommend `html-helper-mode',
since `html-mode' may well be better.  HTML mode comes from
`emacs/lisp/textmodes/sgml-mode.el' and has been updated more
recently.  It just happens that `html-helper-mode' suits my use.  The
point of this remark is that I know nothing about the relevant
libraries,  emacs/lisp/facemenu.el  and  emacs/lisp/files.el  .)

--
    Robert J. Chassell
    address@hidden                         GnuPG Key ID: 004B4AC8
    http://www.rattlesnake.com                  http://www.teak.cc