RE: Images and alternative text

> -----Original Message-----
> From: public-html-request@w3.org [mailto:public-html-request@w3.org] On
> Behalf Of John C. Vernaleo
> Sent: Thursday, August 07, 2008 4:10 PM
> To: Ian Hickson
> Cc: Philip Taylor; public-html@w3.org
> Subject: Re: Images and alternative text
> 
> 
> On Thu, 7 Aug 2008, Ian Hickson wrote:
> 
> >
> > On Thu, 7 Aug 2008, Philip Taylor wrote:
> >>
> >>   <img src="..." alt="{x \over y} = {1 \over {y \over x}}">
> >
> > ...would be a horrific alternative text to give a screen reader.
> >
> > I don't think it's equivalent to the image at all. It's the source of
> a
> > program that was used to generate the image, but that's not the same
> > thing. Would you consider the replacement text of a fractal to be the
> C
> > source code that generated it? Or the replacement text for an SVG
> file to
> > be the raw source code of that SVG file?
> >
> > Correct alternative text for an image generated by LaTeX is a textual
> > representation of the expression generated from the same LaTeX.
> >
> > e.g.:
> >
> >   <img src="..." alt="The fraction x over y is equal to 1 divided by
> the
> >   fraction y over x.">
> >
> 
> I don't know much about screen readers, but I do know something about
> LaTeX, and I just don't see how the textual representation of equations
> scales very well past very simple equations.  Even in the example here
> that sentence is just barely unambiguous.  A more complex equation
> would
> be much worse and a matrix basically impossible.  And I'm not convinced
> a
> human could do it any better than a program could.
> 
> At least the version that was close to the LaTeX code still contained
> the
> relevant information in a way that is mostly parsable by a human.

I was thinking the same thing. I remember being in high school and people
reading, say, quadratic equations out loud. Even something as trivial as
that in the math world requires multiple levels of aural "backtracking" (the
human mind typically can remember the last 20 or less spoken words, or
"tokens", and not reliably at that) which most people are not capable of
really following.

I think that maybe what is needed is for these systems to generate the @alt
equivalent of "Greek text". You know how, when the font is smaller than a
certain size, it stops displaying the actual text and just throws out "ipsum
lorem"? Why don't these systems do that? A formula longer than a fairly
short one gets @alt = "A complex mathematical formula." And shorter ones get
the actual formula?

Alternatively, we could get our "MathML in HTML" spec nailed down, tell HTML
authors to use that and not graphic for math, and officially throw the
problem over the fence to UAs at that point. ;)

J.Ja

Received on Friday, 8 August 2008 13:52:55 UTC