W3C home > Mailing lists > Public > ietf-http-wg-old@w3.org > May to August 1995

Re: any more comments? ('204 No Content' clarification)

From: Alexei Kosut <akosut@nueva.pvt.k12.ca.us>
Date: Wed, 23 Aug 1995 16:35:15 -0700 (PDT)
To: Marc Hedlund <hedlund@best.com>
Cc: Roy Fielding <fielding@beach.w3.org>, http-wg%cuckoo.hpl.hp.com@hplb.hpl.hp.com
Message-Id: <Pine.HPP.3.91.950823162253.8421A-100000@ace.nueva.pvt.k12.ca.us>
On Wed, 23 Aug 1995, Marc Hedlund wrote:

> What is a UA expected to do with form fields in this circumstance?  If a
> script is to take input and leave the form in view, it would be useful to
> clear the form fields without changing the document.  This would: (1) give
> a visual indication that something has happened; and (2) allow repeated
> entries (i.e., data entry) to the same form without reloading or <input
> type="reset">-ing.

Well, this isn't what 204 was originally for. If I remember the mailing
list archives correctly (I read them a couple months ago), it was added a
couple years ago so that imagemaps could tell the client not to do
anything when a an 'unused' portion of the map was clicked. In order to
work properly, this response code must be interpreted by the client as
causing *absolutely* no change to the document window. It should act as if
you had clicked on a part of the document where absolutely nothing exists. 
This is how Netscape and Mosaic both handle it, and it is rather useful -
I use it in imagemaps all the time. 

To do what you wish to accomplish, within the current spec, simply send a
response that redirects the client to the page they came from (i.e. 301 or
302). Ideally, the client would then send a request message, get a 304 Not
Modified back from the server, and would redisplay the document "as it was
originally received". 

> If there are no objections, I'd like to see this:

Consider this an objection (of course, I guess it doesn't really count,
since I'm nobody important). Besides, the HTTP/1.0 draft is suposed to
represent current practice, and your redefninition does not conform. Now,
for HTTP/1.1, I see no reason why a new respose code, along the lines of
"205 Reset Document" couldn't be added, which would have no content, but 
would instruct the UA to refresh the document internally, clearing any 
user-modified portions.

--/ Alexei Kosut <akosut@nueva.pvt.k12.ca.us> /--------/ Lefler on IRC
----------------------------/ <http://www.nueva.pvt.k12.ca.us/~akosut/>
The viewpoints expressed above are entirely false, and in no way
represent Alexei Kosut nor any other person or entity. /--------------
Received on Wednesday, 23 August 1995 16:36:36 EDT

This archive was generated by hypermail pre-2.1.9 : Wednesday, 24 September 2003 06:31:26 EDT