Re: INCLUDE tag

Joe English (jenglish@crl.com)
Thu, 27 Apr 1995 17:17:02 -0700


Message-Id: <199504280019.AA19836@crl.crl.com>
To: Multiple recipients of list <www-html@www10.w3.org>
Subject: Re: INCLUDE tag 
In-Reply-To: <5736020028041995/A23144/RE5/1194E0022100*@MHS> 
Date: Thu, 27 Apr 1995 17:17:02 -0700
From: Joe English <jenglish@crl.com>


Ian Higgs +44 171 510 8595 <IAN.HIGGS@reuters.com> wrote:

> Is there a proposal for anything like an INCLUDE tag?
> 
> e.g.	<include src="http://www.xyz.com/standard-footer.html">
>
> This would have the same structure as IMG but would allow something
> from the server to be included BEFORE the page is processed.

The original HTML 1 documentation suggested that <A> 
might be used for this: <A REL="EMBED" HREF="http://www.xyz.com/...">.
(I think the intent was that inlined images could be specified
this way too, before <IMG> was invented for Mosaic.)
The current status of REL and REV semantics are up in the air; 
I doubt that REL=EMBED will end up being specified
to work this way.

Netscape has invented an <EMBED> element, and deployed in 
the Windows version, but they haven't proposed it 
(or even *mentioned* it) on any of the discussion lists
as far as I know.  Not sure if it's exactly what you want.

SGML has a mechanism for doing this, namely entity references:

    <!doctype html system [
	<!entity footer SYSTEM "http://www.xyz.com/std-footer.html">
    ]>
    <HTML> 
    ... 
    &footer;
    ...
    </HTML>

but only emacs-w3 supports it and most other browsers barf
on the entity declaration and internal DTD subset.
Panorama should be able to handle this too.

Some servers will do "server-side includes", but as
you point out they defeat caching.

"Client-side includes" are probably undeployable by now.
Even if a mechanism were specified, nobody would use it 
because current browsers wouldn't process it correctly, 
and if nobody uses it, browser authors won't bother 
implementing it.



--jenglish@crl.com