Re: MathML vs LaTeX

I understand the need for an accessible medium for mathematics, and the TeX
folks do not appear in any mood to develop compilers for the  purpose of
mapping a TeX/LaTeX doc to anything but PDF. But there is the workflow
problem.
We have professionals in  the field trained in using a format that  can be
made accessible and where an automated mechanism can be developed to map
into optimal formats for accessibility for each print disability.
I am asking this question. Shouldn't we focus on teaching mathematicians to
develop accessible LaTeX that we can translate to  an accessible reading
format, or formats. Remember, for years, what was considered accessible for
blind users was not accessible for me.
Right now I am composing in Word with the equation editor. And, I am using
ChatGPT to map to HTML+LaTeX. I am also getting close to getting ChatGPT to
translate one page of mathematical PDF to one page of Markdown + LaTeX for
Jupyter. Now, that could easily be mapped to whatever we like.
I guess what I'm thinking is that MathML is a good language for "under the
hood" but a poor author facing language. Maybe we need to change our
education in this respect.
Also, we need to work with the TeX Users Group to foster accessible LaTeX.
By accessible I mean capable of being mapped to an accommodating format for
people with print disabilities.
I think the center of this problem is Authoring Tools, and author
education.
Best, Wayne

On Fri, Mar 28, 2025 at 6:12 AM <kerscher@montana.com> wrote:

> Hello Wayne,
>
>
>
> I am blind and use a screen reader. It is wonderful to use linear input
> such as LaTex or the similar codes used in Microsoft Word’s equation editor
> to create expressions. Using MathCAT now available in Jaws and NVDA, I can
> read MathML using speech and braille in the most granular method I choose.
> I can copy MathML from web pages or from EPUB titles and paste it into word
> so I can show my work.
>
>
>
> LaTex is indeed wonderful, but there must be an accessible method to
> convert that linear information into professional looking math, and that is
> what MathML provides.
>
>
>
> Best
>
> George
>
>
>
> [image: Drawing of a guide dog in harness sitting looking up at a dangling
> rope in front of a mountain wall. Credits, Outdoor Life.]
>
>
>
> George Kerscher Ph.D.
>
> -In our Information Age, access to information is a fundamental human
> right.
>
> Chief Innovations Officer, DAISY Consortium
>
> http://www.daisy.org
>
> Senior Advisor, Global Literacy, Benetech
>
> http://www.benetech.org
>
> President, International Digital Publishing Forum (IDPF)
>
> http://www.idpf.org
>
> Chair Steering Council Web Accessibility Initiative (WAI), at W3C
>
> http://www.w3.org/WAI
>
> Cell:+1 406/544-2466
>
> Email: kerscher@montana.com
>
>
>
>
>
>
>
> *From:* Wayne Dick <wayneedick@gmail.com>
> *Sent:* Thursday, March 27, 2025 7:55 PM
> *To:* W3C WAI ig <w3c-wai-ig@w3.org>
> *Subject:* MathML vs LaTeX
>
>
>
> I use LaTeX. I tried with MathML, but which MathML? Is it presentational
> or semantic?
>
> The fact is that mathematicians use LaTeX. It is a language that can
> separate content from presentation. It is more syntactically concise. It
> looks like math. People with cannot author with LaTeX can use other
> products like Word to formulate mathematics, and that can be parsed into
> LaTeX.
>
> First, I must express my gratitude for the members of W3C who tried to
> develop an alternative to LaTeX, but they did not succeed. LaTeX is so much
> easier to code. There is no need to think about a presentational or
> semantic form. I think it is time to drop MathML, as much as I love sematic
> MathML and the people who developed it.
>
> Best, Wayne
>