W3C home > Mailing lists > Public > www-math@w3.org > December 2000

Re: Comments on the MathML test suite - Content

From: Neil Soiffer <soiffer@wolfram.com>
Date: Wed, 27 Dec 2000 18:02:20 -0800 (PST)
Message-Id: <200012280202.SAA03256@wolfram.com>
To: emmanuel.pietriga@xrce.xerox.com (Emmanuel Pietriga)
Cc: www-math@w3.org
> Some comments on the Content section of the test suite:
> 
> *rec-fn3
> rendering is not consistent with code: code says (f+g)(z) and rendering 
> is (f+g)(x)

Fixed.

> *rec-lambda?
> the REC default rendering (uses the lambda symbol) is different from the 
> one provided here (uses the right arrow)

The renderings in the spec are suggested default renderings.  Renders
are allowed to use whatever rendering they feel is most appropriate.
In these cases, the renderings were done by WebEq and WebEq choose to use
a notation that is particularly common in CS classes.  I've decided to leave
these in, since they are not wrong in any way and because they emphasize the
fact that the renderings in the spec are only suggested defaults, and are
not required renderings.

> *rec-image1
> in order to be consistent with the provided rendering, shouldn't the 
> code be like this:
> <math> 
> <apply><eq/><apply><image/><sin/></apply><interval><cn>-1</cn><cn>1</cn></interval></apply> 
> </math>
> instead of this:
> <math> 
> <apply><eq/><apply><image/><apply><sin/><ci>x</ci></apply></apply><interval><cn>-1</cn><cn>1</cn></interval></apply> 
> </math>
> 
> (replace <sin/> by <apply><sin/><ci>x</ci></apply>)

Fixed (wrong in the spec also).

> *rec-piecewise1
> rendering is not consistent with code: code says 0 if x<0 and rendering 
> is -x if x<0

Fixed.

> *and1
> the REC default rendering is different from the one provided here ("and" 
> instead of "^")
> 
> 
> *rec-forall?
> the REC default rendering is different from the one provided here (in 
> the REC, the variable is not rendered as a subscript of forall, and 
> there is the ":" sign which is missing here)

Same as for lambda -- they are different but correct.

> *limit? and tendsto?
> -several inconsistencies between the REC and the test suite as far as 
> the rendering of tendsto is concerned when type is above
> -the code is not consistent with the rendering in some examples (tends 
> to has an attribute type with value "above" and it is rendered as a 
> right arrow (should be down arrow))

I've regenerated some images, but in I think that the spec's use of diagonal
arrows is not very standard and will be changed.

> *trigonometry8
> the code given is only made of presentation markup

Oops.  Fixed.

These fixes will take a while to show up on the W3C site.

Thanks very much for your careful work.

	Neil
Received on Friday, 29 December 2000 14:21:43 GMT

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