Window Object 1.0 comments

[Joe D Williams]

Thanks for the link to:

http://www.w3.org/TR/Window

Reading further, it seems like some input is needed:

1.2. Not in This Specification

* ISSUE-59: Should Window 1.0 include history?
Yes, read-only of current Back url

* ISSUE-60: include access to frames by name (window.frames)?
Yes Preserve backward compatability,
Deprecate name, change to ID.
From DOM, treat 'frame' as a tagname collection

* Opening new windows
Give window api features to minimize the
need to open new window (max liveness)
XMLize the param string
Maintain difference between user-initiated and
onload or other script init

* Manipulating UI of existing windows
Let Browser makers call this
and allow maximum personal preferences

* Window size and position information
full control over child window
always send result of resize event, all is read-only
browser maker/user preference what is settable
for main window

* Client information ("window.navigator")
Hints for possible client accessiblity/capability features
Full feature testing basics

* Window events
Show a couple of current proposals

* Printing
Same as if user selects browser
Print or Print Setup or Print Preview .
Browser could provide same sort of user prefs
as for opening new windows.

* Focus
events for change of focus
controls to control focus

* Selection
pixel, character, element, down to pixel

* Editing
Think collaboration

* Networking APIs
AJAX, peer-to-peer
The objective - live interactive web window

* Handling of binary data
Same as any other encoding file type, mime type,
just pass to document helper processor, or maybe
the processor for 'The EmbeddingElement Interface' object

B. ECMAScript Language Binding

Any recommendations for this from this newsgroup?

Thank You and Best Regards,
Joe

[VK]

Joe D Williams wrote:

Thanks for the link to:

http://www.w3.org/TR/Window

Reading further, it seems like some input is needed:

1.2. Not in This Specification

* ISSUE-59: Should Window 1.0 include history?
Yes, of course. Window has to include everything from DOM 0, otherwise
it's going to be another worthless paper.

window.history.back() // no args
window.history.forward() // no args
window.history.go(+/- shift)
window.history.length

Read/write protected *including* current page (must use
window.location.href instead)
Respectively go() method accepts only relative navigation: go(-1)

* ISSUE-60: include access to frames by name (window.frames)?
Yes Preserve backward compatability,
Deprecate name, change to ID.
From DOM, treat 'frame' as a tagname collection
Nice thinking, but frames collection is based on names (besides
indexes), not id's. Not doable for legacy reasons. Just mark as an
exception.
* Opening new windows
Give window api features to minimize the
need to open new window (max liveness)
XMLize the param string
Not doable for legacy reasons.
Maintain difference between user-initiated and
onload or other script init
Not clear: more details?
* Manipulating UI of existing windows
Let Browser makers call this
and allow maximum personal preferences

Not doable for security reasons (wide open door for all kind of
spoofings).

<snip>

You have to realize that it is not an attemp to "invent the window". It
is an attempt to first time ever describe and formalize something that
existed and was in native use since Netscape 2.x
*DOM 0*
This way the primary task is to put in writing all DOM 0 features as
they are: because no one ever will drop them or change them for legacy
reasons. Only after that it may possible to start thinking what other
*existing* host object extension should be *recommended* for universal
support and which one should be marked as too proprietary.

[Jim Ley]

On Tue, 11 Apr 2006 02:14:31 GMT, "Joe D Williams"
<jo*****@earthlink.net> wrote:

Thanks for the link to:

http://www.w3.org/TR/Window

Reading further, it seems like some input is needed:

Please send this to the pu***********@w3.org mailing list, so the
working group can see it.

You may also want to read the other drafts, and the other
deliverables, which touch on many of your other points.

Cheers,

Jim.