W3C home > Mailing lists > Public > www-math@w3.org > March 2001

Re: W3-MathML questions: real/float, annotation, etc.

From: Stan Devitt <jsdevitt@stratumtek.com>
Date: Wed, 14 Mar 2001 13:48:44 -0500
Message-ID: <001201c0acb7$6523ede0$6961a8c0@devitt.local>
To: "Tobias Burnus" <tobias.burnus@physik.fu-berlin.de>, <www-math@w3.org>
Cc: "Tobias Burnus" <tobias.burnus@physik.fu-berlin.de>, "Hans Hagen" <pragma@wxs.nl>
For sure, the discussion of  "float" (floating-point) is inadequate.

The distinciton between float (floating-point) and real is supposed to be 
one of representation.  A real number is entered as a single 
CDATA entry consisting of sign, digits and  decimal point.

    <cn  type="real">+345.237</cn>

 while a float is represented as a mantissa. and exponent 
separated by a <sep/> entry. (see definition of cn in appendix C)

    <cn type=:"floating-point" base="10">3.45237</sep>2</cn>


The distinction between "complex" and "complex-cartesian"  comes up
in the context of identifiers.  It is perfectly reasonable for someone to
know that an identifier stands for a complex number without knowing
what representation will be used for a particular instance of such 
a number.  Thus, the type "complex" is a mathematical type while the
types "complex-cartesian" and "complex-polar" are complex numbers
with a particular data representation.

Also, note that the possible type values is not a closed list.

Stan.

p.s.  David C and I will be raising this with the working group and a more 
authoritive statement will come from the working group 
once we sort out the wording, etc.


----- Original Message ----- 
From: "Tobias Burnus" <tobias.burnus@physik.fu-berlin.de>
To: <www-math@w3.org>
Cc: "Tobias Burnus" <tobias.burnus@physik.fu-berlin.de>; "Hans Hagen" <pragma@wxs.nl>
Sent: Wednesday, March 14, 2001 4:41 AM
Subject: W3-MathML questions: real/float, annotation, etc.


> Hi,
> 
> while implementing MathML support for ConTeXt
> (http://www.w3.org/Math/#Software) we came across of something which
> seemed not to be very clear to us:
> 
> a) It is not clear what the difference between "float" and "real.
> (I cannot find "float" anymore, maybe the PDF file is/was older)
> Also the difference between "complex" and "complex-cartesian" is not
> clear. (I guess there is none.)
> 
> b) Are there any guidelines which <semantic/> presentations should be
> favoured and if it is allowed to have both xml-annotations MathML-Content
> and/or MathML-Presentation (the examples use only the latter).
> 
> c) Do you know a list of the different presentations of mathematical
> operators (such as "n over k" which I cannot find in the HTML version).
> 
> With warm regards,
> 
> Tobias
> 
> 
> 
Received on Wednesday, 14 March 2001 13:49:03 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Saturday, 20 February 2010 06:12:50 GMT