Re: ISO standards -- Was: True quotes (fwd)

Murray Altheim (murray@spyglass.com)
Tue, 23 Jul 1996 10:14:18 -0500


Message-Id: <v02110102ae1a959bcc1c@[140.186.34.50]>
Date: Tue, 23 Jul 1996 10:14:18 -0500
To: "David Perrell" <davidp@earthlink.net>
From: murray@spyglass.com (Murray Altheim)
Subject: Re: ISO standards  -- Was: True quotes (fwd)
Cc: www-html@w3.org, MegaZone <megazone@livingston.com>

David Perrell <davidp@earthlink.net> writes:
>Mr Zone of the Discordian Society wrote:
>> These names are well established in the SGML community - and HTML *IS* a
>> subset of SGML.
>
>Okay! I was chastised in a private post for my misinformed discordancy, and
>sheepishly shut up. I sure wouldn't want anyone to waste time changing
>established names. I sure wish someone would spend time getting those
>entities into HTML.

David,

I sent a message July 12th [1] regarding a similar topic: the existing 8879
entity sets already part of SGML/HTML. Here's the relevant part of my
earlier message regarding availability of the SGML entity files:

I wrote:
>If you're looking for ISO entity sets, they can be found all over the net.
>The DocBook directories contain a large number of character entities
>covering math, publication symbols and languages. You can find them in
>Davenport's DocBook 2.4.1 distribution at
>
>    http://ftp.ora.com/pub/davenport/docbook/
>
>in files:
>
>    dbg241.tar.Z    20-Apr-96 15:20   878K
>    dbg241.zip      20-Apr-96 15:21   841K
>
>Or if you just want the entities, I've moved my copies into a directory you
>can get to at
>
>    http://www.stonehand.com/dtd/docbook24/ents/
>
>Theoretically, a conforming SGML system would use all of the 8879 entity
>sets. This isn't the case with HTML currently, but it sure wouldn't be a
>bad idea. Only problem would be browser and font support, which is
>something W3C is working on currently. [...]

In a sense, the SGML entity catalog is already part of HTML, being an
application of SGML. But support for them in existing browsers will
probably require more than simply recognition -- look to W3C's efforts in
Web fonts for the ability to supply the hundreds of glyphs necessary.

Beyond the issue of font/glyph availability, all this discussion of Unicode
in HTML is somewhat premature. Note that numeric entity references in the
&#8000; range (as has been bandied about here frequently) are not currently
supported in SGML under the current declaration, nor is there operating
system support for any charset to Unicode glyph mapping (please correct me
if I'm wrong) in anything but NT. And *then* there is still a slough of
issues to be worked out -- it's not simply installing a new font.

Rest assured it is being worked on -- you just won't be seeing anything
next week. The i18n discussion is quite interesting, but certainly not a
simple problem.

Murray

[1] Date: Fri, 12 Jul 1996 13:25:48 -0500
    To: "William F. Hammond" <hammond@csc.albany.edu>
    From: murray@spyglass.com (Murray Altheim)
    Subject: Re: ISO standards  -- Was:  extra character entities

```````````````````````````````````````````````````````````````````````````````
     Murray Altheim, Program Manager
     Spyglass, Inc., Cambridge, Massachusetts
     phone: +1-617-621-9701 x201
     email: <mailto:murray@spyglass.com>
     http:  <http://www.stonehand.com/murray/murray.html>
            "Give a monkey the tools and he'll eventually build a typewriter."