Re: Method to get math to the most people?

Could MathML become like Acrobat? A standard of its own that one can
assume people have downloaded the viewer for? Most unlikely, I think. 

Undoubtedly it could become a niche standard, for which people who care
about it have readers. For example, I could put out my course notes in
MathML and then require the students to install a reader (if there were
one that were managable)*. But is hardly what we want. So I do my course
notes in HTML using TtH. [*I mean no unjustified slight to WebEQ, which of
course also solves the automatic download problem but at a big cost in
bandwidth and usability].

The only reason that PDF somewhat succeeds at being a standard is that
is has tremendous breadth. It is not just a niche. It aims at being THE
page description standard for the web (as PostScript is for printers).
MathML can't hope to compete on that basis. (Although Acroread is so buggy
that it seems to me to be in the process of undermining PDF's chances as a
widespread standard.)

What we want is a standard way that anyone can read our mathematics on
their favorite browser, the same way they read anything else. I sure hope
that Mozilla will give us that. If it does. Then IE will be forced to
compete. Then (and only then IMHO) MathML will become established.

And Robert, I have nothing against pointy brackets. I prefer that they
refer to quantum states, but I can live with other uses! What I am
interested in is stuff that works, more than standards for their own sake.

Regards,
	Ian Hutchinson, Plasma Science and Fusion Center, MIT. 
	http://psfc.mit.edu/~hutch/home.html

On Sat, 18 Dec 1999, Robert Miner wrote:

> 
> 
> Hi Ian,
> 
> You and Roger Blake wrote:
> 
> | > >MathML does not work with any popular browser yet, and there seems little
> | > >prospect of it being incorporated soon into IE.
> | > 
> | > Depends what you mean by soon, and on what degree of typographic perfection
> | > you require of the rendering. I know of at least one company that has worked
> | > on stylesheets for MathML in IE5 and believe this work has progressed
> | > substantially, although not yet available.
> | 
> | Depends what you mean by "incorporated". Anything that requires a reader
> | explicitly to install some new package on their browser is in practice
> | inaccessible to the general reader. If you have to send a whole stylesheet
> | to translate the MathML into native layout with every page, the mind
> | boggles as to how cumbersome it would be.
> 
> >From my interaction with people on the IE team, I think it is true
> that it will be a while before we see a native IE implementation, they
> have been really supportive of the stylesheet/plug-in route to MathML
> rendering.  For example, I believe a few extra font properties have
> been exposed just to help with Math.  IE5 is a very much improved
> platform for plug-in technologies than we have had available in the
> past.
> 
> I also think it is a bit extreme to imply that installing some sort of
> plug-in is a crippling obstacle to that method for MathML rendering.
> Automatic software delivery and installation systems have gotten quite
> sophisticated and are improving.  And although the group of people who
> have access to IE5 is admittedly somewhat selective, it is still a
> large group, and incorporates a lot of researchers and educators.  And
> as a precedent, most of these people also use Acrobat, a plug-in.
> 
> To make the discussion concrete, take a look at the beta preview of
> the new WebEQ Math Viewer for IE5 at
> 
>    http://www.webeq.com/preview
> 
> I have a bit of a math paper I translated using TtM.  (And TtM is
> great.  Thank you!  This is a real contribution to the MathML cause,
> which is doubly praiseworthy since I don't think you are really a big
> fan of pointy brackets ;-)
> 
> It still has a number of the old WebEQ weaknesses, but it also has
> some nice improvements.  There is also a new embded editor control for
> getting input back from readers.
> 
> --Robert
> 
> ----------------------------------------------------------------
> Robert Miner                          http://www.webeq.com
> Geometry Technologies, Inc.           email: rminer@geomtech.com
>                                       phone: 651-223-2884
> ----------------------------------------------------------------
> 
> 
>

Received on Sunday, 19 December 1999 10:51:26 UTC