W3C home > Mailing lists > Public > www-math@w3.org > January 2014

RE: Operator Dictionary. Non-Normative

From: Daniel Marques <dani@wiris.com>
Date: Wed, 29 Jan 2014 15:13:09 +0100
Message-ID: <bdad37eb945ddbbc15fda4d7b7124ecf@mail.gmail.com>
To: David Carlisle <davidc@nag.co.uk>, www-math@w3.org
David,

Thanks for your response. Adding the stretchy="false" seems a sensible
thing to do to avoid any ambiguity. The only drawback is that the MathML
is getting longer.

Damo



-----Original Message-----
From: David Carlisle [mailto:davidc@nag.co.uk]
Sent: martes, 28 de enero de 2014 14:46
To: Daniel Marques; www-math@w3.org
Subject: Re: Operator Dictionary. Non-Normative

On 28/01/2014 12:19, Daniel Marques wrote:
> Hello,
>
> I was wandering the exact meaning of "non-normative" appearing in the
> Operator Dictionary appendix.
>
> As you already know, at WIRIS, we do edition/input of MathML. We
> decided that parenthesis, among other symbols, are no stretchy by
> default due to practical reasons. If <mo>(</mo> and <mo>)</mo> were
> stretchy, we needed some kind of predictive tool to match opening with
> closing parenthesis. And because we just decided we did not wanted to
> pass as an oracle, parenthesis ended up being no stretchy.
>
> Daniel Marques from WIRIS
>


It means exactly that: that technically you are conformant if you do what
you say. That said, the code would be more portable if when generating you
explicitly put <mo stretchy="false">(</mo> rather than rely on a
non-standard default.

You shouldn't really have to match brackets though. MathML stretching is
not like TeX \left...\right that stretches things in pairs. In MathML each
operator stretches independently to the height of its containing mrow.

David
Received on Wednesday, 29 January 2014 14:13:30 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:27:46 UTC