HTML+ and automated forms

Alastair Aitken CLMS <ZPALASTAIR@CLUSTER.NORTH-LONDON.AC.UK>

Mail folder: WWW Talk Apr 94-present
Next message: Gavin Nicol: ""
Previous message: Bryon Propst: "Checkbox and Forms"
Reply: Larry Masinter: "Re: HTML+ and automated forms"

Errors-To: listmaster@www0.cern.ch
Date: Mon, 27 Jun 1994 13:38:00 +0200
Errors-To: listmaster@www0.cern.ch
Message-id: <9406271134.AA00861@dxmint.cern.ch>
Errors-To: listmaster@www0.cern.ch
Reply-To: ZPALASTAIR@CLUSTER.NORTH-LONDON.AC.UK
Originator: www-talk@info.cern.ch
Sender: www-talk@www0.cern.ch
Precedence: bulk
From: Alastair Aitken CLMS <ZPALASTAIR@CLUSTER.NORTH-LONDON.AC.UK>
To: Multiple recipients of list <www-talk@www0.cern.ch>
Subject: HTML+ and automated forms
X-Listprocessor-Version: 6.0c -- ListProcessor by Anastasios Kotsikonas

(Chris Lilley, whom I quite wrongly flamed here a couple of weeks ago,
graciously accepted my private apology. I was wrong. - Thank you Chris.)

HTML + and automated documents.

Comments are due to Dave Raggett re HTML+ proposal by 23rd June.  I want to
flag these thoughts to this group.

Forms:

At the moment field inputs to forms is very limited.  The browsers don't
handle them very well - <return> does not move from field to field
terminating each input and moving the cursor to the next field.  Neither
Lynx or Mosaic seems very form-friendly yet.  I am not sure to what extent
this is due to the HTML specification but I think it is not very much.

I see HTML (++ perhaps) as being an ideal standard for the automation of
forms.  The concept of lookup tables for field inputs, code and description
is a simple structure and would need no further definition beyond a
"lookup" parameter in the field definition, any more would require a file
structure definition and may become very complicated.  This "lookup"
facility would provide the potential for data validation and verification
without recourse to the server but would lead to deeper more massive forms.
The HTML+ specification provides for notification to the server when a
field is updated but this would be slower and more cumbersome than
including, in HTML (++ ?), facilities for the automation of documents.

Electronic indexing, with dynamic display of results, is a useful addition
and gives a hint to the future shape of documents in webspace.  They look a
lot like large reference works - many k!  An automated form miight occupy
quite a large space - 100k for 20 -30 inputs but I think I would like to
see this sort of thing handled by a future version of HTML.  

Perhaps, in my ignorance I am moving into an entirely different
specification and if I am please point me at it.  I have not found a
standard for automated forms yet but believe this is might be crucial to
HTML.  Would it, for example, be possible to build forms to produce
hypertext documents with lookup tables for tags?  Might forms like this
produce also others like themselves on the web?

I will plough on with this because I believe there is demand for some
greater automation - loops, array support, etc.  This might lead to very
complex forms but might (would it?) provide a very useful and versatile
tool for webmasters, document authors, feedback, registration.

Automated forms?  Anyone?

Al. <-:< (zpalastair@grid.unl.ac.uk)
(Wearing a seat belt in case I fall off my seat!)