Re: Reminder: MathML intent meeting 5 Nov, 2020

On 05/11/2020 08:36, Physikerwelt wrote:
> While some common attributes have more complex values like the 
> event-attributes that specify a callback function or the style attribute 
> that specifies CSS values the proposed new attributes seem even more 
> complex to me.
> 
> This makes MathML the black sheep in the HTML family once again. Does 
> the majority of this group really think it is a good idea to encode tree 
> structures as part of XML / HTML documents that are inherently trees?

As far as I can see in Deyan's proposal it is not tree structures per se 
that are being encoded but functions (we could argue the point about 
whether functions can be reduced to trees or indeed graphs but that's a 
different matter). As such there are well established formalisms such as 
XPath and XQuery that encode functions into XML attributes so MathML 
would hardly be a black sheep in that respect.

The main problem I see with Deyan's approach is not the complex 
structure of the attribute content but the fact it may be difficult to 
internationalize/localize. It might be better to have the annotation 
entirely separated (or separatable) from the element to which it refers, 
linking to that element by id or XPath etc, so that localizable 
annotations can be swapped in easily without changing the MathML 
expression or tree itself.

Received on Thursday, 5 November 2020 18:08:22 UTC