Re: Possible use of <INSERT> - serious inquiry (fwd)

Paul Prescod (papresco@itrc.uwaterloo.ca)
Wed, 20 Mar 1996 19:43:39 -0500 (EST)


Date: Wed, 20 Mar 1996 19:43:39 -0500 (EST)
From: Paul Prescod <papresco@itrc.uwaterloo.ca>
To: MegaZone <megazone@livingston.com>
Cc: www-html@w3.org
Subject: Re: Possible use of <INSERT> - serious inquiry (fwd)
In-Reply-To: <199603210007.QAA27860@server.livingston.com>
Message-Id: <Pine.SUN.3.91.960320193708.1133B-100000@itrc.uwaterloo.ca>

On Wed, 20 Mar 1996, MegaZone wrote:

> Once upon a time Paul Prescod shaped the electrons to say...
> >I would suggest that you look up what "valid HTML" means.
> 
> Why does everyone seem to have missed my point.

Perhaps it wasn't stated clearly enough.

> It was that some people feel that there is no reason to keep HTML a subset
> of SGML.  And if there are features that work well for the online community
> that don't conform to strict SGML - should we limit ourselves?  Everytime
> I get together at conventions and meetings with other web folk there are
> two major camps - the SGML purists and the HTML seperatists.  And a smaller
> group of people who really don't care which path it follows - maybe MU (do
> a combination of both) - as long as the power and flexibility continue to
> increase.

HTML separatists? You're the first I met. I suppose some of the folks at 
Netscape behave as if they were HTML separatists, but they've never made 
a statement to that effect.

> I'm saying lets change what 'valid HTML' means, *if* that would provide
> increased functionality on the web.
>
> This started from a comment that including other documents might violate
> the rules of SGML since you shouldn't require an SGML parser to be able to
> understand and follow URLs in order to validate a document.  My response
> was that there is obviously a demand for this - I see it almost daily on
> mailing lists and newsgroups - and the <INSERT> tag is already on the way,
> so should we avoid that capability simply because SGML parsers can't keep
> up with it?  

Let me assure you, "SGML" has no problem following URLs.  SGML _parsers_ 
may not follow URLs (most of them predate the Web!). But SGML was 
designed flexible enough to be able to support arbitrary external object 
referencing schemes.

> Maybe it is time to extend the capabilities of SGML then, or
> maybe it is time to split HTML off and remove the restraints.

As I have explained, SGML has no problem with it. _Some_ SGML tools 
might, but SGML doesn't.

> Basically what I'm saying is - lets not wear SGML blinders.

You needent worry. SGML has lots of room to support anything the Web can 
throw at it. The whole point of the other thread was that the HTML community 
has historically _reinvented_ the features that are _already_ in SGML.

 Paul Prescod