W3C home > Mailing lists > Public > www-tag@w3.org > January 2002

Re: Background information on GET and XForms (was: GET should be encouraged...)

From: Tim Berners-Lee <timbl@w3.org>
Date: Mon, 28 Jan 2002 16:04:49 -0500
Message-ID: <001201c1a83f$6c666360$f4001d12@CREST>
To: "Mike Dierken" <mike@dataconcert.com>, <w3c-forms@w3.org>, <www-tag@w3.org>

----- Original Message -----
From: "Mike Dierken" <mike@dataconcert.com>
To: <w3c-forms@w3.org>; <www-tag@w3.org>
Sent: Monday, January 28, 2002 2:09 PM
Subject: RE: Background information on GET and XForms (was: GET should be
encouraged...)


> >
> > Wouldnt the following analogy be better for GET requests:
> >
> > If you compare GET requests to a SELECT SQL query the body is
> > the WHERE
> > clause, and we need effecient mechanism to transport the WHERE clause.
> >
> That may be correct from the point of view of an individual developer, but
> it may be a more important requirement to the Web that it be easy to use
and
> re-use the WHERE clause (hence an opaque URI) than efficiently transport
it.
>
> There is real value in maintaining a URI encoding for a GET request even
at
> the cost of server-side conversion to a syntactically rich query language.
> There is also very real value in allowing for richer/structured input for
> POST type of operations - but the two alternatives and their relative
> strengths need to be very clear to designers.

Nicely put.     The trade off manages the advantages of the methods:

GET:
    +  addressable response
    +  cachable by proxies and other advantages of visibility of lack of
side effects
POST:
   + Don't have to do the yukky URL encoding for long documents

In the future, this does call for a new GET-with-body or QUERY or whatever
you like to call it, which would be defined as an operation without side
effects
(a function) of both the URI  and the message body.

It is possible that the TAG should recommend that work be started on
this in conjunction with the IETF in order to in the longer term ease this
tradeoff.
(I prefer QUERY to an adaptation of GET, myself).

Tim
Received on Monday, 28 January 2002 16:04:43 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Thursday, 26 April 2012 12:47:04 GMT