RE: self-contained html file format

The feaure of embedding images and other media types into the HTML is a good idea as far as storing the information in the form of HTML pages. But we should at the same time remember that by incorporating a means the have embedded media in the HTML page are we limiting its capability to reach out to a wider spectrum of user agents. Why should an image or any other media be downloaded by a user agent if it comes to know about them before hand? That would save bandwidth and from a user agent's perspective there is not need for it to download something which is not legible for it? 
There are tool available which can be used to download the entire information from a file which can be used to archive or even store information locally. These tools download all the resources(images, media) referred to by the hyperlinks.
What do you say Josh? 
 -----Original Message----- 
From: Josh Haberman [mailto:joshua@haberman.com] 
Sent: Wed 1/22/2003 12:01 PM 
To: www-html@w3.org 
Cc: 
Subject: self-contained html file format




 HTML is probably the most portable document display format after plain
 text.  However, its use as a document exchange and archival format is
 diminished by the inability to embed binary data, most significantly
 images.
 
 It seems to me that it would be immensely useful to have some standard
 scheme for storing an html document along with all of its external data
 (such as images) in a single file.  It could use some archival format
 such as .zip or .tar, some standard file extension like .zhtml, and some
 standard mime type like application/zhtml.  The benefit would be that
 .zhtml would be a self-contained document instead of a metadocument
 containing some of the content but linking to the rest.
 
 Thoughts?  I am not subscribed, please CC me on replies.
 
 Josh Haberman