W3C home > Mailing lists > Public > www-html@w3.org > August 2000

Re: XHTML 1.1 : no frames

From: Tantek Celik <tantek@cs.stanford.edu>
Date: Thu, 31 Aug 2000 11:51:21 -0700
Message-Id: <200008311854.OAA02346@tux.w3.org>
To: Dave J Woolley <DJW@bts.co.uk>, "'www-html@w3.org'" <www-html@w3.org>
From: Dave  J Woolley <DJW@bts.co.uk>
Date: Thu, Aug 31, 2000, 11:21 AM

>> From: Tantek Celik [SMTP:tantek@cs.stanford.edu]
>> unfortunate because it would have been very simple to do (add the "name"
>> attribute to the "object" tag, and allow the "target" attribute on "a
>> href").
>   Because this would have re-introduced one of
> the reasons for removing frames, namely that a URL
> cannot reflect the full state of a page

Oh, this is WAY too funny.

You mean some people out there actually believe that URL _can_ reflect the
"full state" of a page?

What foreign substances are causing this hallucination? I am all too

So, assuming you weren't just being cynical with your statement -
 why this is a hallucination -

There is the obvious "DHTML" answer.  The javascript and DOM on a page can
pretty much turn the page into anything.

How about form elements?  As in, what is checked or unchecked, and how much
of a form is filled in?  Certainly URLs do not reflect that.

Cookies anyone?  No, they're not part of the URL, and they certainly can
affect the "full state" of the page.

And there are servers that return different content based on the
"referred-from" (sp?) in the header of the request.

And none of these are "invalid" or "incorrect".  They're all perfectly
reasonable (and useful - just keep saying 'eCommerce' in your head).

A URL (or URI) is nothing but a reference to a resource somewhere.  It's not
a "name", it's not a "full reflection of state" (except perhaps "data:"
URLs), and it's certainly not guaranteed to exist or be complete.

> - e.g. if
> a search engine gives you a pointer to the contents
> of the object, you will get the object without the
> container when you follow the link; similarly for
> personal bookmarks (= IE Favorites).

As far as how this affects the UI of user agents, yes, it does make it
challenging for us implementors, but even then we come up with creative
solutions, such as Web Archive Files and the Internet Scrapbook in IE5/Mac -
the latter of which actually captures the "current" state of a page (however
much it has been modified by DHTML etc.)

No one ever said it was supposed to be easy to implement a user agent - just
easy to use.


It's the question that brought you here.    http://www.microsoft.com/mac/ie/
Received on Thursday, 31 August 2000 14:54:24 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 15:05:54 UTC