Re: HTML is declarative on purpose [was: Web neurons ]

Len Bullard (
Tue, 28 May 1996 23:58:37 -0500

Date: Tue, 28 May 1996 23:58:37 -0500
Message-Id: <>
To: "Daniel W. Connolly" <>
From: (Len Bullard)
Subject: Re: HTML is declarative on purpose [was: Web neurons ]
Cc: (John Middlemas),

>In message <>, Len Bullard writes:
>>  The dead ends with HTML probably
>>start in using a declarative root language.
>I'll pick on Len because I know he won't take it personally.

Good to see your sig, Dan.

>But I'm seeing this claim stated more strongly in lots
>of forums. It takes the form of:
>	"Let's add <IF> and <WHILE> to HTML!"
>	"With javascript, you can do anything!"
>	"HTML is the MS-DOS of the internet"

To me, the root declarative style of HTML limits
it only in that it is a single set of productions.  

The beauty of a DTD is that it is a legal document.
The curse of a DTD is that it is a legal document.

To improve it, extensions are needed which
then become adopted after they reach some
mass in the community.  VRML has identical
issues but the 2.0 proto nodes offer a
means to experiment while maintaining
a sharable core.  2.0 also has a script
node to which a VRMLScript will be added.

Yes, to use this information, one must run it.
To author it native, one must code and well.

HTML is the text lingua franca.  It appears from 
here, the future of HTML remains extensions
and periodic standard revisions.  Good.

Frames may not be HTML, but they nonetheless 
enable multiple simultaneous targets for data display 
and manipulation, and that is an irresistable capability 
for hypermedia. Why are they declared in the same DTD?

HTML is an application notation, IMO.  It is the
language of one of the frames.  Should it also
be the frame language itself?

>HTML is declarative and limited on purpose. If it were
>turing complete, the only reliable way to consume the information
>in a document would be to "run" it. (Just ask the search
>engine vendors who are trying to deal with Javascript).

Only to make this clear, I am not asking for this. 
Please explain the problem of the search vendors
as you see it.

>In order to make folks aware of the conscious decision
>behind this limitation, I'm considering reviving
>an old essay of mine and revising it for publication
>as a W3C working draft:
>	Toward a Formalism for Communication On the Web Feb 1994

Will read.

>>   The HTML
>>model has limits that begin to show as more folks try
>>to add more functionality and cannot agree on it.  HTML
>>should be an architecture, not a DTD.
>Fair enough. Yet the burden of proof is still on the folks
>that want to see this happen. 

What do you think of James Clark's approach in SP?

len bullard