W3C home > Mailing lists > Public > www-html@w3.org > May 1998

Re: HTML Arch and Future of HTML proposals

From: Frank Boumphrey <bckman@ix.netcom.com>
Date: Fri, 1 May 1998 17:34:22 -0700
To: <roconnor@uwaterloo.ca>, <www-html@w3.org>
Message-ID: <01bd7562$0cf3f000$ceacdccf@uspppBckman>
From my understanding of architecture (which is rather rudimentary), it
could also be used to display XML files by mapping the XML elements to HTML
elements.

Please correct me if I'm wrong

Frank Boumphrey.

-----Original Message-----
From: Russell Steven Shawn O'Connor <roconnor@wronski.math.uwaterloo.ca>
To: www-html@w3.org <www-html@w3.org>
Date: Friday, May 01, 1998 9:23 AM
Subject: HTML Arch and Future of HTML proposals


>I'm going to push the HTML Architecture idea some more, because I really
>*really* think it is a good idea.  Let's see how it is related to other
>proposals.
>
>Zhengrong Song <song@cse.psu.edu>
>   Proposal for a mirror element for linking to a mirror site.
>
>With an HTML architecture any element can be used in your HTML file.
>Some elements may not be mapped to the HTML architecture and be ignored.
>Vendors may specify their own architectures with extended elements (such
>as MIRROR).  Authors may, at their own risk, use, vendor architectures
>instead.  This is much like the current system where authors use
>proprietary elements such as BLINK.
>
>One advantage is that multiple architectures can be specified for a
>document.  So an author can specify one vendor architecture (or more)
>and the official HTML architecture at the same time.  This may prevent
>conflicts some between vender implementations.
>
>Of course HTML 5.0 may or may not want to support the MIRROR element.
>(Personally I think standardizing some sort of MIRROR relationship to be
>used with A and LINK elements would be best.)
>
>
>Sebastian Rahtz <s.rahtz@elsevier.co.uk> and Herbert van Zijl
>   The future of HTML, from the perspective of Elsevier Science
>
>This proposal advocates the use of Architectures.  More smart people here.
>Not much more to say other than to agree that authors would be able to
>specify their own elements.
>
>
>Ramon Casha <rcasha@technologist.com>
>   Recommendations for a preprocessor attribute in HTML
>
>Using SGML system entities is the best method for some types of
>transculsion.  An HTML Architecture would allow the author to specify
>whatever entities e would like, including SYSTEM entities.
>
>
>Masayasu Ishikawa
>   Harmonizing the "subset" of HTML; Internationalization
>
>The DTD you choose to validate against can be a subset of HTML.  The
>Architectural mapping will still work.
>
>
>Robert S. Smith <rsmith@mitec.net>
>"I believe there needs to be a way to designate a list of HTML pages as a
>book so that the browser can search them all for a specific word instead
>of just searching one page, just like a regular book. HTML books are
>becoming increasingly common."
>
>Although this doesn't have to do with HTML architectures, I feel that the
>NEXT and PREV relationships are adequate for designating a list of HTML
>pages.  Of course the same comment about vendor implementations of elements
>that applied to the MIRROR element applies here too.
>
>
>Rob Rothenburg <wlkngowl@unix.asb.com>
>   Dictionaries
>
>Same comments about vendor implementations of elements applies.
>
>Curtis Yarvin <Curtis_Yarvin@geoworks.com>, Ian Porteous
><ian@geoworks.com>
>   Bringing the Web to Handheld Devices: Practical Issues
>
>The same comments about ``Harmonizing the "subset" of HTML;
>Internationalization'' apply here.  Also, a compact HTML architecture can
>be made.  A document can conform to both the standard and compact HTML
>architectures simultaneously.
>
>Daniel Austin <daniela@cnet.com>, Greg Sherwin <gregs@cnet.com>
>  The Role of HTML as a Display Format
>
>I disagree with this position.  Although people may move towards XML,
>HTMLs role should become an architecture defining some common *semantics*
>for the variety of elements people will create with XML.  XMLs is unable to
>communicate the semantics of many of its elements.  HTML architectures can
>give UA's more information for XML documents conforming to it.
>
>This is important.  An XML document can have an OL element, and a style
>sheet can make it look like an ordered list.  But a UA doesn't know that it
>is an ordered list.  However, if the OL is mapped to the OL element in
>the HTML architecutre, then any HTML aware UA will know that the element
>is an ordered list.
>
>
>Hidetaka Ohto <ohto@isl.mei.co.jp>
>   A Proposal for Mobile HTML Architecture
>
>See other comments about compact HTML.
>
>
>Ben Trafford <ben@aerosoftsys.com>
>    From Sunday April 26th.
>
>As I describe, XML documents can implement the HTML architecture.  Great
>isn't it.
>
>
>Joel Nava <jnava@adobe.com>, Bruce Hunt <bhunt@adobe.com>, T.V. Raman
><raman@adobe.com>, Adobe Systems Inc.
>   HTML as an XML application
>
>As I describe, XML documents can implement the HTML architecture.  Great
>isn't it.
>
>-- Russell O'Connor                           roconnor@uwaterloo.ca
>    <URL:http://www.undergrad.math.uwaterloo.ca/%7Eroconnor/>
>``And truth irreversibly destroys the meaning of its own message''
>-- Anindita Dutta, ``The Paradox of Truth, the Truth of Entropy''
>
>
>
>
Received on Friday, 1 May 1998 17:29:02 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 27 March 2012 18:15:36 GMT