W3C home > Mailing lists > Public > www-math@w3.org > October 2003

Re: Spacing and formatting across multiple math tags

From: Stan Devitt <jsdevitt@stratumtek.com>
Date: Tue, 28 Oct 2003 12:18:53 -0500
Message-ID: <3F9EA4FD.9080408@stratumtek.com>
To: silverbanana@gmx.de
Cc: www-math@w3.org

Actually, my point was that XSLT could NOT do it even with tables
because there is not enough information in the multistep model.  The
xhtml construction I had in mind  and which captures your troublesome
cell was (Up to padding)

  <td colspan="3"> first part of first expression</td>
  <td>colspan="3"</td>second part of first expression</td>
  <td >end of first expression</td>
  <td >right hand side</td>

Basically, XSLT would not be able to use automatic
line wrapping info to determine the mathml fragments that
go into each row from a single mathml expression.

Once I have know the fragments, this could be made to look good.


Bernd Fuhrmann wrote:

> > This example helps.   The layout you prefer could be done now using
> something like 4 columns, 6 rows, some appropriate colspan attribute
> values, and some careful alignment and/or padding.
> Yes and no: it would work, but it would be bad, though it might be the 
> best solution. However, I tend to use good solutions only, so I will 
> rather risk a more confusing formating that using tables to do sth. 
> that I'm not supposed to do with them.
> [...]
> XSLT could do that. 

> But it is really dificult to get this to work since I do not know if 
> this problem will ever be treated so that tables can be avoided. If I 
> hadn't that hope, XSLT would be useless since there would be never any 
> better (feasable) solution. So let's assume MathML 3.0 (or whatever) 
> could fix this problem. This might require a significant structural 
> change in all my content which would make XSLT somehow useless.
> > For example, to mark up Term 7 entirely in a single MathML table
> cell  we would need some sort of  malign which interacted
> with  automatic line-wrapping  in  a new way
> Yep, we need a new way of treating this, exactly. Well, patience is a 
> virtue! Let's see what future versions of MathML will bring...
Received on Tuesday, 28 October 2003 12:22:43 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:27:34 UTC