RE: Technical reasons for some options taken on design of MathML

Bruce Miller wrote:
> juanrgonzaleza@canonicalscience.com wrote:
> [snip]
>>
>> I said in a previous communication that the phrase "Elsevier adopts
>> MathML" was to be very debatable. It may be interesting to note here that
>> in last Elsevier’s CEP 1.1.0-1.1.3 (the core of Elsevier’s 2005 XML DTDs
>> family)
>>
>> 238
>>    U
>> 92
>
> That sure looks like chemistry to me, not math.
> Elsevier may have reasons for not adopting an explicit
> chemical markup language, but using an arbitrary
> text pre sub/superscript markup is arguably _less_
> wrong than using math markup, whether MathML or any other.
>
> Suggesting to use math to markup uranium, is
> simply using the fact that math has handy
> machinery for dealing with scripts --- the thing
> you're marking up isn't math.  It's the same
> as using math to add a "nd" superscript to the 2
> in "2nd edition".  Ie. it's an abuse, albeit
> a common one.
>

I got that example from own Elsevier’s specialists. The encoding is
defined in Elsevier’s DTD family 5. The elements <ce:inf> and <ce:sup> and
the attribute loc are defined in the *core* part of the CEP DTD.

The CEP is the central element of modular design of Elsevier’s XML
technology and the core -as its name suggests- is available for any
document, not just for Elsevier’s chemical journals.

Elsevier developed an explicit XML markup for chemical data: <ce:chem>,
<ce:compound-formula>, <ce:compound-info>, <ce:stereochem>...

> the thing
> you're marking up isn't math.

The important point of the message was the alternative script model was
"CSS- friendly", but if you are worry on the use of chemistry oriented
examples and prefer pure mathematics then look for the ISO 12083 appendix
"A.6 Mathematics"

Prescripts i and j would be encoded á la Elsevier <ce:sup loc="pre"> á la
MAIDEN <sur> or via "extended MathML code" introduced by White Lynx here

[http://lists.w3.org/Archives/Public/www-math/2006Apr/0038.html]

The rest of the discussion is a bit off-topic.

> It may well be that MathML's markup for scripts
> is less than ideal, but examples from chemistry
> aren't convincing.
>
> --
> bruce.miller@nist.gov
> http://math.nist.gov/~BMiller/
>

Elsevier's folks illustrated encoding of prescripts using an example from
chemistry probably because it is a well-known script model in science. A
practical example of the use of prescripts in mathematical notation are
Randic’s topological indices on graph theory.

For instance {}^1 gamma is defined as

1              -1/2
 X = SUM (v v )
      ij   i j


Juan R.

Center for CANONICAL |SCIENCE)

Received on Tuesday, 18 April 2006 09:57:41 UTC