RE: Tag Soup (was: FW: XHTML)

On Sun, 5 Dec 1999, Kjetil Kjernsmo wrote:
> On Sun, 5 Dec 1999, Arjun Ray wrote:

> >Tag Soup is easy to implement, easy to use, and poses few if any hurdles
> >on the learning curve.  Generalized Markup requires thought, planning in
> >advance, and militates against adhockery. 
> 
> Well, I find HTML 4.0 Strict very easy to write with some exceptions.

Sure, but the point is that you basically had to *learn* this (i.e once
you had taken the trouble to RTFM, a lot of things became much clearer.
But how many people RTFM as opposed to just futzing with combinations of
tags in their favorite wowser?)

> Isn't it just a matter of shifting attention from "what should my
> document look like" to "what do I intend to say"?

It may be a bit more than that.  HTML can't be a "one size fits all".
"Intending to say" something also presupposes the means to say/express it.
Depending on the semantic domain, HTML may still require a conscious and
therefore possibly onerous/distracting process of translation (from "deep
structure" to the generic types it supports.)  The Tag Soup paradigm poses
no such constraints: the tags are supposed to be *directly* "palpable".

> I must admit that I was pretty horrified by the thought of having to
> write CSS as well as HTML in the beginning, but I think what needs
> emphasizing to people, is that structured HTML is really very simple.

Yes, and that *any* stylesheet mechanism is far more powerful than the
"smack a toolbar button" method - of which plunk-a-tag is the cognate.
Word processing packages like MS-Word or WordPerfect have had
sophisticated template/style mechanisms for years, but still only a
vanishingly small minority of users evne know about them!
 
> >> (e.g. collapsing lists would have been great!), 
> >
> >We had 'em.  In early 94. 
> 
> I was too late to experience this... :-(

Well, it's very inconvenient for the purveyors of fractured fairy tales
(to the hagiographic if not totemic effect that Mosaic/Netscape/MSIE
"invented" everything) that archives do exist... Take a look:

  http://ftp.sunet.se/ftp/pub/www/browsers/viola/

The screenDumps/ subdirectory has some interesting pictures.  Even more
interesting are the timestamps.  (On 16-May-1994, the very first line of
code for the very first version of Netscape was still to be written.)  
The chess demo actually worked, too - ignoring the unfortunately wrong
polarity. (Click on a piece, then click on a legal square to move it to,
and it would be moved.  The whole thing was about 44k in 'violascript',
and the polarity fix was trivial.)
 
> >I agree, but the point is not so much to assign blame as to understand the
> >"forces" involved.  

> > WYSIWYG *has* captured the public imagination, and so in service to
> > it, browsers are developed as extensions of the authors' will rather
> > than the readers'.
> 
> Agreed, so what do we do about it...? 

For one thing, emphasize the *distinction*.  Let Tag Soup have its day in
the sun.  Give it the legitimacy of a formal spec.  Let people *know*
(rather than have them believe, or worse, scorn standards in favor of the
"evidence in front of their own eyes") that Netploder is a Tag Soup
processor par excellence.  Given that, it's much easier to make the case
that one needs *something else* to process Generalized Markup.

Fostering the impression that squares can be circled and that tag soup
processors can, or worse should, be the vehicles to popularize structured
markup is the very *last* thing we should be doing.

XML in Netploder is a recipe for failure. 

> I guess there are still many people out there who haven't yet started
> writing web pages. The message that needs to get through is that you
> should think about structure rather than looks. How do we say that?

With a different class of software.


Arjun

Received on Sunday, 5 December 1999 09:03:44 UTC