Re: HTML3 MATH tag

On Mon, 17 Aug 1998, William F. Hammond wrote:
>    ``mathematical typewriter emulation'' (MTE) standard in mass market
>    browser native rendering as part of native HTML. MTE is just emulation
>    of the mathematical typewriter prevalent in all mathematics
>    departments during the period 1960-1980.
> ...
>    
>    MTE is more ``in scale'' with ordinary HTML than is MathML, which is
>    much closer to fussy typesetting.
>    
>    All that needs to be added to basic HTML is:
>     1. the horde of character entities that we need (in scalable fonts
>...
>     2. a simple element "<lg> ... </lg>" (logical group) with attributes
>        for horizontal and/or vertical cursor motion,
>...
>     3. elements "<math>" (paragraph level) and "displaymath" (block
>        level) in which
>           + the new "lg" tag is permitted.
>           + all character level things are rendered one at a time

What you are describing is mostly available in HTML 4.0 Transitional as it
stands. 

1. Symbol entities can be accessed through <font face=symbols> (deprecated
but currently still the most functional approach)
2. Exists as the HTML <table> (with some limitations) plus judicious use
of the <sup> and <sub> tags.
3. Is just the outermost table, since you don't have to enable <lg> if
it is already just table.

TtH is my implementation of a translator from LaTeX and TeX to this
approach. [http://hutchinson.belmont.ma.us/tth/]
It is a surprisingly effective and powerful approach, and has been adopted
by lots of people. Admittedly though, it is not always beautiful.

Moreover, I am sure that the MathML champions will point out that MathML
has much more ambitious objectives than this, which is why neither they
nor I would advocate and expansion of HTML in the direction you are
suggesting. They want something that is a truly descriptive mathematical
mark-up language, independent of the rendering medium. This is a lofty
goal -- unfortunately so far seeming TOO lofty for rapid widespread
adoption and implementation. Perhaps time will prove the goal to have been
visionary AND realistic; then again, perhaps not. One thing that IS
certain, though, is that MathML is far too verbose and complicated to be a
direct authoring markup language. [Sure, you can in principle write it
directly by hand, as you can with P*stscript, but in neither case does it
make much sense.] So our best hope for MathML is that it becomes an
intermediary standard for which there are developed sufficiently
widespread renderers that it makes sense to publish in this form.

Final point though: what you describe is basically what dvi is. Dvi
renderers are freely available and already can be coupled to browsers. Of
course, I don't want to author dvi directly any more that I want to author
MathML, but I don't have to. Still, the main problem is that not everybody
has a renderer or can be bothered to install one so it is not attractive
to publish in dvi if I want everybody to be able to read what I write. 
MathML currently has this same problem. 

Ian Hutchinson.

Received on Monday, 17 August 1998 14:33:27 UTC