RE: CSS-positioned OBJECT vs. frames -- Resizability

Okay, as I've feared, I've taken this discussion completely off-topic
for www-style, so I'm posting my reply over here.

>> Garth Wallace:
>> >> A while back there was a discussion about frames being replaced by
>> >> <OBJECT type="text/html"> and CSS absolute positioning. I recently
>> >> realized a problem--how would you create the equivalent of
>> resizable
>> >> frames with such an arrangement?

>> Braden McDaniel:
>> >What's more, how would you target them?

I:
>> 	That one seems easy.

Braden:
>"Seems," I think, is very appropriate.<g>

>The comment beside the definition of this attribute reads "submit as part of
>form," and there is a hyperlink to section 17.4 (The INPUT element). AFAICT,
>the spec only discusses the use of OBJECT's name attribute in conjunction
>with forms. It doesn't appear to me that the spec intends for OBJECT to be
>targetable.

	Sorry, I was thinking in terms of how one might target OBJECTs
if they replaced frames in a future version of HTML.  This discussion
is presubably about using CSS to make OBJECTs act like frames in HTML
4.0.

	BTW, I still can't see how an OBJECT would be submitted in
conjunction with a form, and so what the NAME would be used for.

>>  OBJECT already supports a NAME attribute
>> (although ID would probably be more appropriate).  TARGET="_top"
>> becomes the browser window itself, TARGET="_parent" the document
>> containing the OBJECT in which the link appears, etc.  That wouldn't
>> even require a change to the HTML 4.0 Transitional DTD, just a change
>> in interpretation.

>I don't think this is a matter of interpreting the spec a certain way, but
>would require adding language to it.

	Sorry again.  The DTD includes comments indicating what the
attributes are *supposed* to be used for, and those would have to be
changed if this functionality were added to the spec.  What I meant
was that a page which used A TARGET="foo" and OBJECT NAME="foo" to
target a frame would validate as HTML 4.0 Transitional.  It would of
course be cheating, though.  After all, <INPUT TYPE="text" CHECKED>
also validates.
					John T. Whelan
					whelan@iname.com
					http://www.slack.net/~whelan/

Received on Friday, 21 August 1998 01:49:13 UTC