Re: Accents in the operator dictionary

I forgot to send a response, but several weeks back we made changes to
chapter 3 and fixed up the operator dictionary somewhat that are along the
lines you suggested.  Take a look at:
http://monet.nag.co.uk/~dpc/draft-spec/chapter3-d.html#id.3.2.5.7

In particular, look at  3.2.5.7.2 Default value of the form attribute.

There were lots of changes/improvements to the operator dictionary:
http://monet.nag.co.uk/~dpc/draft-spec/appendixc.html
(the diff marked form of the operator dictionary was messed up when I looked
at it)

I hope these seem reasonable to you,

Neil Soiffer
Senior Scientist
Design Science, Inc.
www.dessci.com
~ Makers of MathType, MathFlow, MathPlayer, MathDaisy, WebEQ, Equation
Editor ~



2010/4/21 Neil Soiffer <neil.soiffer@gmail.com>

>
>
> 2010/4/21 Frédéric WANG <fred.wang@free.fr>
>
>  So questions/remarks:
>>
>> 1) What is the rationale for defining most accents as "postfix" operators?
>>
>
> I don't know the rationale, but I can guess... it is probably some holdover
> from when combining chars were used for accents (even though they are not
> supposed to be used as combining chars).  I was going to say that some, such
> as U+02C6 are often interchangeable with their combining char "equivalents"
> (eg, U+0302 in this case), but U+0302 is not listed as stretchy (which seems
> wrong).
>
>
> 2) The rules for determining the form and the operator dictionary make this
>> two cases behave differently:
>>   - low line: <mover><mi>abcdef</mi><mo>_</mo></mover> (form = infix, so
>> not stretchy)
>>   - overline: <mover><mi>abcdef</mi><mo>&#x203E;</mo></mover> (form =
>> postfix, so stretchy)
>>
>
> I agree that "low line" seems wrong.
>
>
>
>>  3) Why is "check mark" a stretchy accent (or even just a mathematical
>> operator)? Is it related to \widecheck (caron) or square root?
>>
>
> I'm not sure when we decided to eliminate it, but "check mark" is not in
> the current draft.  See
> http://monet.nag.co.uk/~dpc/draft-spec/appendixc.html<http://monet.nag.co.uk/%7Edpc/draft-spec/appendixc.html>
>
>
> Note:  the Operator Dictionary is non-normative, so you can ignore it as
> you please.  Nevertheless, it would be good to remove as many errors as
> possible from it.  We'll discuss these problems at our next MathML meeting.
> If you find more, please send them along.
>
> Neil Soiffer
> Senior Scientist
> Design Science, Inc.
> www.dessci.com
> ~ Makers of MathType, MathFlow, MathPlayer, MathDaisy, WebEQ, Equation
> Editor ~
>
>
>
>>
>> --
>> Frédéric Wang.
>> www.maths-informatique-jeux.com
>> www.maths-informatique-jeux.com/international
>>
>
>