Re: Cougar DTD extra character entities

Adam M. Costello (amc@cs.wustl.edu)
Thu, 11 Jul 1996 23:04:27 GMT


Date: Thu, 11 Jul 1996 23:04:27 GMT
From: amc@cs.wustl.edu (Adam M. Costello)
Message-Id: <9607112304.AA03115@siesta.cs.wustl.edu>
To: www-html@w3.org
Subject: Re: Cougar DTD extra character entities

Jim Taylor <JHTaylor@videodiscovery.com> says:

> <!ENTITY emsp   SDATA "[emsp  ]"--=em space-->
> <!ENTITY ensp   SDATA "[ensp  ]"--=en space (1/2-em)-->
> <!ENTITY mdash  SDATA "[mdash ]"--=em dash-->
> <!ENTITY ndash  SDATA "[ndash ]"--=en dash-->
> 
> The fact that "em" and "en" are used for spaces but "m" and "n" are
> used for dashes could be responsible for the confusion. In retrospect
> it seems the authors of the SGML entities should have been more
> consistent. (Does anyone know if they had a reason for doing it this
> way?)

I don't know if they had a reason, but I can see a reason:

mdash and ndash are already pretty long.  emdash and endash would be
even longer.

But msp and nsp look like they want to be pronounced "em ess pee" and
"en ess pee".  The dashes don't suffer that problem.

I think I'd prefer consistency, but what is consistency in this case?
Internal consistency, or consistency with an existing internally
inconsistent spec?

AMC