[math-font-discuss] Unicode note on linear math representation

[Martin Schröder]

----- Forwarded message from Neil Soiffer <NeilS at DESSCI.COM> -----

Message-ID: <027701c6745d$a8c9b610$7101a8c0 at raindrops>
Date: Wed, 10 May 2006 11:15:02 -0700
Sender: PDF-Access <PDF-ACCESS at LISTSERV.AIIM.ORG>
From: Neil Soiffer <NeilS at DESSCI.COM>
Subject: Re: Unicode note on linear math representation
To: PDF-ACCESS at LISTSERV.AIIM.ORG
List-Archive: <http://listserv.aiim.org/scripts/wa.exe?LIST=PDF-ACCESS>

----- Original Message ----- 
From: "Joe Clark" <joeclark at JOECLARK.ORG>
To: <PDF-ACCESS at LISTSERV.AIIM.ORG>
Sent: Wednesday, May 10, 2006 5:12 AM
Subject: Unicode note on linear math representation


> What a coincidence that they would publish a note on that topic in April 
> just as we were starting to talk about it.
>
> <http://www.unicode.org/notes/tn28/>

A couple of things that I think should be noted.  First off this, is a 
"technical note" by Murray Sargent, not by the Unicode Consortium.  To quote 
from http://www.unicode.org/notes:

"These technical notes are independent publications, not approved by any of 
the Unicode Technical Committees, nor are they part of the Unicode Standard 
or any other Unicode specification. Publication does not imply endorsement 
by the Unicode Consortium in any way."

Murray is anti-WYSIWYG when it comes to math.  He developed his own linear 
notation years ago and does not use WYSIWYG math editors or even TeX.  This 
puts him on the fringe.  It doesn't mean his ideas are right or wrong, but 
it does mean that his ideas don't jib with current practice.

Despite this, even he admits that MathML is better for our purposes:

"The linear format is useful for 1) inputting technical text, 2) displaying 
such text by text engines that cannot display a built-up format, and 3) 
computer programs. For more general storage and interchange of math 
expressions between math-aware programs, MathML and other higher-level 
languages are preferred."

[my apologies for any typos -- the PDF he produced confuses Acrobat's 
selection so cut and paste was hit and miss and I ended up retyping and I 
ended up retyping this and other quotes by hand]

Ie, his note advocates using his "language" as an input method and agrees 
that use of MathML is preferable as "a general storage and interchange" 
method.

Later on his note says,

"In general where syntax and semantic choices were made, input convenience 
was given high priority."

Obviously for PDF, input convenience is not a consideration so choices made 
by Murray are dubious in the context of math in PDF.  Perhaps more 
importantly, Murray proposes a *language* that is meant to be processed. 
There are specific characters that are supposed to be used to (eg) make 
"sin(x)" look right -- you don't type just the ASCII letters.  Hence, 
although his idea is that you can avoid a WYSIWYG editor, you still must 
adhere to the rules of the language he proposes -- something that might look 
like math because it is a sequence of linear Unicode characters does not 
necessarily mean it is proper math as per his language.  Hence, I'm not sure 
this is really support for Joe's idea that simple math should be part of the 
standard for math.

FYI:  there is a lively discussion on the wiki about Math: 
http://pdf.editme.com/pdfua-mathml.  I've added most of the contents of the 
email as a comment there already.

    Neil


----- End forwarded message -----

-- 
                    http://www.tm.oneiros.de