W3C home > Mailing lists > Public > www-math@w3.org > January 2009

Re: Where do the "default renderings" in spec chapter 4 come from?

From: David Carlisle <davidc@nag.co.uk>
Date: Tue, 13 Jan 2009 17:06:11 GMT
Message-Id: <200901131706.n0DH6BpJ008778@edinburgh.nag.co.uk>
To: ch.lange@jacobs-university.de
Cc: www-math@w3.org, c.mueller@jacobs-university.de

> Has there ever been a debate on a particular rendering,

most of them date back to mathml1, which is 10 years, there probably
were arguments but obviously we're far too polite to remember them

As far as I know the Math working group has never actually voted on
anything, we always reach agreement in the end (even if only that all
but one person is too tired to argue any more:-)

> Sure -- but given the importance of MathML, your "suggestions" do have
> a high impact.

Actually we need to stress more that they should _not_ have as much
impact as people think. We need to have some "default default" and some
variant of English usage is as good as any, but we have to avoid any
suggestion of cultural takeover. If you (or more importantly your users)
are (say) German and the letters "lcm" don't denote whatever's the German
for Least Common Multiple, then you should feel absolutely no hesitation
in making the default rendering for <lcm/> in some system be whatever is
the local convention. Similarly ]a,b[ instead of (a,b). In an ideal
world of course all such preferences would be user-configurable but the
world is not always ideal, and there is no implication that it is more
correct for a system to default to English usage than German or Arabic.

> Why \subset for the proper subset but not \subsetneq?
Again this predates my time in the WG, but I think that the natural markup is
to use the subset symbol for proper subset, and subset eq (with a line
under it) for subset-or-equal.

Of course some people use subset symbol for subset-or-equal and then
need to use \subsetneq for proper subset but as I say sometimes you have
to make an arbitrary choice.

 > Why f' but not df/d? 

I don't think I've seen df/d (with no named variable). In the form with
differentiating a function term f so there is no bound variable I think
f' or D(f) is what I'd expect, but again there is a lot of cultural
background to people's expectations for this type of thing. Especially
the difference between the ' (or  .) markup and d/dx markup is famously due to
a notable German and a notable English mathematician not getting on too


The Numerical Algorithms Group Ltd is a company registered in England
and Wales with company number 1249803. The registered office is:
Wilkinson House, Jordan Hill Road, Oxford OX2 8DR, United Kingdom.

This e-mail has been scanned for all viruses by Star. The service is
powered by MessageLabs. 
Received on Tuesday, 13 January 2009 17:06:47 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:27:41 UTC