HTML Plans [was: I am confused]

Daniel W. Connolly (
Fri, 10 May 1996 15:17:49 -0400

Message-Id: <>
Subject: HTML Plans [was: I am confused]
In-Reply-To: Your message of "Fri, 10 May 1996 20:33:14 -0100."
Date: Fri, 10 May 1996 15:17:49 -0400
From: "Daniel W. Connolly" <>


In message <>, schwarte@iwb.uni-st writes:
>Will <math>, <fig>, <overlay> and related tags that I am missing in
>HTML 3.2 (and so are many HTML-users) be in back in future versions,
>perhaps in HTML 4.0?

In short: yes, but possibly in a modified form.

Expanded version: please read the relavent background materials before
posting to a public forum:

| W3C continuing to work with vendors on extensions for multimedia
| objects, scripting, style sheets, layout, forms and math. W3C plans
| on incorporating this work in further versions of HTML. See The
| W3C Activity Statement on HTML for details. 

For details, see:

	W3C Activity: Hypertext Markup Language (HTML)

>Again the question: Why have they been eliminated in the
>present version, which obviously in not allready that official as 
>"Magazone" believes?
>Am I right that HTML 3.2 covers just a subset of the tags that are 
>meant to be official, whatever that means? If so, HTML 3.2 seems 
>to be a confusing interlude. 

The HTML 3.x situation has been "a confusing interlude" for some
time. (See attached supporting evidence from altavista search).

HTML 3.2 is supposed to represent stuff that the implmentors have
agreed how it works, and the information providers can safely use.
Or something like that. Broad generalizations like that seem
to lead to nothing but trouble.

The DTD last edited by hand at 23-Apr-96 (and checked into RCS at
1996/05/06 22:11:23) represents the consensus of the W3C editorial
review board.

Now we're considering input from the public (which has a bunch of
implementation experience of their own that we really should have
considered, but didn't -- YET).

The bottom line is: HTML 3.2 is supposed to represent "where we are,"
and not "where we want to be." The W3C activity statement on HTML
outlines where we want to be.

Regarding HTML 3.0 math:

HTML math was always supposed to represent enough of the structure of
the information that it could be manipulated by symbolic math
packages. From what I understand, on close examination, the
implementors of those symbolic math packages found the HTML 3.0 markup
unacceptable. It does a reasonable job representing pictures of
equasions, but not the information behind them.

>I also do not understand why the "extended HTML-tables" are not
>included in HTML 3.2.

Because we don't feel there is sufficient implementation experience,
i.e. nobody has coded the whole thing up. The HTML 3.2 DTD is
an attempt to represent what folks seem to have coded up.

> The DTD on this topic seemed to be allmost
>finished. Will this be in a future version too?

We expect so. Again, see the activity page for information about
the future.

>And what about client-side-imagemaps? The <fig> based concept seems
>to be better then the concept of Seidman that has been implemented in
>HTML 3.2. I guess that it is because Netscape and Microsoft did 
>implement it as well.  Or will there be an <object> based concept for 
>client-side-imagemaps in the future?

Again, the information is at your fingertips:


    <object data="navbar.gif" shapes>
        <a href=guide.html shape=rect coords="0,0,118,28">Access Guide</a> |
        <a href=shortcut.html shape=rect coords="118,0,184,28">Go</a> |
        <a href=search.html shape=rect coords="184,0,276,28">Search</a> |
        <a href=top10.html shape=rect coords="276,0,373,28">Top Ten</a>

>My suggestion: PLEASE stop talking about the official 
>HTML 3.2 and release some unofficial HTML X.Y as soon as possible, 
>that covers all the mentioned stuff. 

Experience shows that would be counterproductive.


Content-Type: message/news

Subject: Re: STRONG vs B (was Re: Opinions on my pages please)
Date: 01 May 1996 02:43:40 +0100
From: (Piercarlo Grandi)
Organization: Prifysgol Cymru, Aberystwyth
Newsgroups: comp.infosystems.www.authoring.misc,comp.infosystems.www.authoring.html
References: <>
     <> <4lrda2$>

>>> I agree with you that HTML should be done logically instead of
>>> physically, but it looks like HTML will be obsolete as soon as most
>>> browsers catch up to the HTML 3.0 enhancements.

But there is not such as the HTML 3.0 enhancements! There is no HTML
3.0, and there will never be one. Something called "HTML level 3" has
been promised as a future standard of rather undefined
characteristics. Here is the scoop from the W3:

  "Current situation

  (See [["Specs, Drafts, and Reports"]] in {{the HTML overview}} for a
  list of relevant publications.)

  We are frequently asked questions such as:

      o What HTML tags should I use? Which ones can I count on?
        Which ones are standard?
      o Are tables standard?
      o What about frames?
      o What happened to FIG?
      o Why isn't HTML 3.0 supported?
      o When will HTML 3.0 become a standard?

  The current situation is that the only features that are both
  specified in a published standard and universally supported are those
  in [[the HTML 2.0 specification]] (though support for the [[file
  upload RFC]] is nearly universal by now).

  "The [[HTML 3.0 draft]] has expired, and is not being maintained. The
  features proposed in that draft are in development, in other
  specification documents."

Note that this means that there is not HTML 3.0 standard, not even for
_tables_. The only thing that they state as being almost universally
supported is the file upload facility. and that there are a few
activities in progress to define piecemeal extensions to HTML 2.0, that
(presumably) will amount to various "levels" of extensions.

The Sandia HTML reference manual merely enumerates a series of common
extensions from various sources:

  "Michael J. Hannah at Sandia National Laboratories maintains an [[HTML
  Reference Manual]] (updated 2 January 1996) which does a reasonable
  job of describing the various features out there today.

  In response to the various proposed extensions to HTML, we have
  convened an editorial review board. The ERB has produced a number of
  [[working drafts]], and [[publicly committed]]to producing
  comprehensive HTML specifications."

This is the beginning of the [[publicly committed]] link above:

  "CAMBRIDGE, USA -- March 4, 1996 -- The World Wide Web Consortium at
  MIT's Laboratory for Computer Science and INRIA has reached agreement
  with market leaders to establish interoperability standards for HTML
  features such as multimedia objects, style sheets, forms, scripting,
  tables, high quality printing, and improved access for the visually

  The World Wide Web Consortium (W3C) has brought together experts from
  companies including IBM, Microsoft, Netscape Communications
  Corporation, Novell, Spyglass, and SoftQuad for joint work on these
  activities. Recently anounced work includes methods of embedding
  active objects within HTML, for which there had previously been many
  divergent proposals.

  The technical team forming the W3C's HTML editorial review board
  (HTML-ERB) expects to define new versions of HTML in the next few
  months. Specifications, once in a suitably complete form, will be made
  available for public review, including by the Internet Engineering
  Task Force. Draft documents describing work in progress are available
  from the Consortium's web site,

  The original HTML specification was written by Tim Berners-Lee, now
  director of W3C, while he was at CERN. Innovations from NCSA and other
  contributors were reviewed under the auspices of the Internet
  Engineering Task Force, and published as the HTML 2.0 specification,
  RFC 1866, edited by Dan Connolly, now at W3C. Design work on HTML
  draws from sources such as the HTML+ and HTML 3.0 drafts by Dave
  Raggett of Hewlett Packard Laboratories, and extensions proposed by
  W3C member companies."

Check out these URLs:

For details and some very partial proposals (for object embedding and
frames in CSS, notably).

>>> On Tue, 30 Apr 1996 00:25:19 PST, (Mike
>>> Meyer) said:

mwm> Most? Nuts - I'd be happy if the POPULAR browsers caught up with the
mwm> HTML 3 capable browsers. Unfortunately, the popular browsers ignore
mwm> the HTML 3 features designed to make writing UA independent documents
mwm> easy, and instead give us less capable cruft that makes writing such
mwm> documents harder.

Perhaps you have missed out on a recent trend: in the common usage, now
that the Web consortium has officially stated that there will not be any
HTML 3.0 standard, a lot of people use HTML 3.0 to indicate Netscape
Modified HTML, and quite a number of WWW sites have pages stating
proudly: "this page supports HTML 3.0 from Netscape, download here
Netscape 2.x".

For several examples of this amusing attitude see below, the results of
an Digital Altavista query. Please remember, while reading these
statements, that *there is no such thing as HTML 3.0*, therefore any
claim as to HTML 3.0 compliance is entirely bogus, and in fact it is
transparent from the entries below that a lot of people have the
impression, nurtured by Netscape, that Netscape HTML is HTML 3.0.

Netscape implements a version of HTML that here and here only resembles
the proposed, but later withdrawn, next generation HTML, and deviates
from it rather considerably. Explorer simply imitates, mostly, with
random deviations, what Netscape does.

  Search the Web and Display the Results in Standard Form
  +"html 3" +netscape +explorer__________________________ [[Submit]]

  Documents 1-10 of about 300 matching some of the query terms, best matches

      My home page is HTML 3.0 compliant. Therefore, using either Netscape 2.0
      or Microsoft Internet Explorer 2.0 is highly recommended. Netscape and
      [[]] - size 2K -
      30 Jan 96

      The Oxyfresh pages work best with Netscape or Microsoft Explorer. Some
      of our pages include tables, backgrounds, and other HTML 3.0 enhanced
      features. If...
      [[]] - size 1K - 22 Mar 96

    [[Huntsville All Star League]]
      Career Leaders. This page is HTML 3.0 enhanced and is best viewed with
      the latest versions of. Microsoft's Internet Explorer , Netscape
      Navigator, or NCSA...
      [[]] - size 12K - 20
      Mar 96

    [[Huntsville All Star League]]
      Lane Assignments. This page is HTML 3.0 enhanced and is best viewed with
      the latest versions of. Microsoft's Internet Explorer , Netscape
      Navigator, or...
      [[]] - size 10K - 28
      Mar 96

    [[Omega International, IBC Inc.]]
      This Document was created for use with html 3.0 compatible browsers,
      such as Netscape version 2, Microsoft Internet Explorer for Windows95,
      or others. To...
      [[]] - size 1K - 17 Apr 96

    [[Fichiers et utilitaires]]
      Fichiers et utilitaires. Ce site est conforme aux normes HTML 3.0
      lisible avec les versions de Netscape 1.2N et plus. Internet Explorer -
      lien plus rapide...
      [[]] - size 4K - 25 Mar 96

    [[Welcome to Rykodisc/Hannibal/Gramavision]]
      Welcome to Rykodisc/Hannibal/Gramavision. HTML 3.0 VERSION (Netscape,
      Microsoft Explorer) HTML 1.0 VERSION (AOL)
      [[]] - size 1K - 5 Apr 96

      Maybe this'll be spiffed up soon, but for now we recommend using
      Netscape 2.0b2+ or an HTML 3.0 compliant browser (NCSA's Mosaic &
      Microsoft's Internet...
      [[]] - size 1K - 13 Feb 96

    [[Lighthouse Web Solutions]]
      Our websites look best when viewed with HTML 3.0 browsers like Netscape
      Navigator Internet Explorer and Mosaic 2.0+ Now America Online users can
      [[]] - size 3K - 17 Apr 96

    [[Detecting your browser...]]
      Detecting your browser... If you're using Netscape or another browser
      such as Microsoft's Internet Explorer that supports HTML 3.0 and/or...
      [[]] - size 1K - 12 Apr 96