Re: pls change main validator form back to GET

Dan Connolly <connolly@w3.org> wrote:

>Eek! I just noticed the validator homepage switched from GET to POST!!!
><http://validator.w3.org/>
>
>[[[
>   The "get" method should be used when the form ....
>   causes no side-effects.
>]]] -- <http://www.w3.org/TR/1999/REC-html401-19991224
>                        /interact/forms.html#h-17.13.1>

It does cause side-effects; we just disagree on what constitutes a
"side-effect" in this context. :-)


>I suppose it's for file upload; but the earlier design was just fine for
>that: a separate POST form for file upload was provided.

I disagree. That design was suboptimal from a Human Interface
perspective[0].


>I see that even the POST form provides an addressable result; that's
>sorta nifty (how did you do that by the way?)

Magic... :-)


>but it doesn't make clear that the user of the form isn't
>committing himself to anything.

I dispute the basic premise that GET vs. POST provides for this distinction
-- no matter the wishfull thinking one chooses to apply -- as well as it's
applicability to the sucject at hand. Not the least reason for which is...

>"As we're talking about user interface here, I'd like to see a clean
> interface for making a deed, which makes it quite clear to me that I
> am committing something, and not just doing another search."
> -- http://www.w3.org/DesignIssues/UI.html

...this is a matter for the purveyors of standards for "form technology"
and implementors of such standards to resolve. The current state of the art
does not allow for this distinction to be made in practice and would not,
IMO, belong at the protocol layer if it did.


I suggest the TAG take this up with the XForms Working Group and a few of
it's Members instead of hitting Joe Q. Public about the head with their
findings. IOW, "Pick on somebody your own size!". :-)


>Please change it back ASAP.

Perhaps...

...but it is unlikely that I will make such a chance in the very near
future. Quite apart from the time required for convincing me of the
suitability -- much less the necessity -- of this, the release branch is
currently in maintenance mode and the development branch is far from the
next release branch point.


>See also
>
>TAG Finding: URIs, Addressability, and the use of HTTP GET
>http://www.w3.org/2001/tag/doc/get7

You don't feel even slightly awkward citing yourself, as Editor for that
TAG Finding, as an authority to support your argument? :-)



Thanks for your feedback on this.

-- 
"Python 2.0 beta 1 is now available from BeOpen PythonLabs.   There is a long
 list of new features since Python 1.6, released earlier today. We don't plan
 on any new releases in the next 24 hours."  - From Python 2.0b1 Announcement

Received on Monday, 2 December 2002 02:35:12 UTC