From: David Carlisle <davidc@nag.co.uk>

Date: Mon, 01 Nov 2010 22:59:26 +0000

Message-ID: <4CCF464E.80701@nag.co.uk>

To: "William F. Hammond" <hammond@math.albany.edu>

CC: W3C MathML Discussion <www-math@w3.org>

Date: Mon, 01 Nov 2010 22:59:26 +0000

Message-ID: <4CCF464E.80701@nag.co.uk>

To: "William F. Hammond" <hammond@math.albany.edu>

CC: W3C MathML Discussion <www-math@w3.org>

On 01/11/2010 22:18, William F Hammond wrote: > >>> Does anyone know, is it possible to produce a bordered matrix using >>> MathML? > >> something like >> http://monet.nag.co.uk/~dpc/bm.html > > (I was offline last week.) > > Inasmuch as I prefer to look at tables recursively -- and so am not > fond of rowspan and colspan, Hmm sometimes a recursive structure is more natural but here actually I think the spanning structure is closer to teh semantic, as it places the column labels in the labelled columns, whereas the alternative you propose just has them floating free and just visually aligned rather than being tied to the column by markup. > I like this approach: > > http://math.albany.edu/~hammond/mmlmisc/bordermxt.xhtml I have to agree though this has the distinct advantage of working whereas the version I offered currently only works in mathplayer of the common browsers. You cheat a bit here (only a bit:-) by knowing the widths in advance, but perhaps with a bit of javascript measuring the widths could be automatically deduced after setting the main matrix to its normal width. > > > -- Bill > > P.S. Am I correct in perceiving the border labels as not > mathematically semantic? If so, probably the 2x2 outer array > should be an ordinary html table. > > If that is the whole display then perhaps, but if it is embedded in a larger expression then doing it all in mathml makes sense I think. DavidReceived on Monday, 1 November 2010 22:59:55 GMT

*
This archive was generated by hypermail 2.2.0+W3C-0.50
: Monday, 1 November 2010 22:59:56 GMT
*