# [rfc-i] Equations as graphics

Julian Reschke julian.reschke at gmx.de
Wed Apr 18 09:40:37 PDT 2012

On 2012-04-18 18:31, Joe Hildebrand wrote:
> On 4/17/12 9:56 PM, "Peter Saint-Andre"<stpeter at stpeter.im>  wrote:
>
>> On 4/17/12 9:44 PM, John Levine wrote:
>>>> Equations-shown-as-graphics are a possible simplification to the
>>>> requirements that still gets us the kind of display we want.
>>>
>>> Well, maybe.  When I think of equations, I think of the Unix "eqn"
>>> program, TeX $math$ and the like, which can be typeset inline or
>>> displayed and for which the source is greppable.  Graphics are
>>> always displayed and probably created by something external to
>>> my text editor.
>>
>> IMHO it is good to keep as much content full-text searchable as possible.
>
> I expected to say "provide an image as a fall-back to MathML", but after
> spending a few minutes in the land of MathML, wishing I was on a later
> branch of Chrome that might have picked up a little more of WebKit, I'm a
> little disappointed in the backwards-compatibility story for MathML.  If
> someone has expertise in this space and can point out the right way to do
> this, I'd appreciate it.
>
> If we were willing to allow JavaScript (and decided that the Apache 2.0
> license was adequate), MathJax (http://www.mathjax.org/) might help with the
> rendering problem.  We could also consider using this sort of JavaScript
> processing ONLY for a prototyping effort, while we push the browser vendors
> to increase the priority of MathML support.

If we decide to do MathML (which I believe would be the right approach),
then David Carlisle probably can help us understanding what we can rely
on, and in which version of browsers.

Best regards, Julian