- From: Chris Lilley <chris@w3.org>
- Date: Wed, 25 Feb 2004 09:08:18 +0100
- To: Etan Wexler <ewexler@stickdog.com>
- Cc: www-style@w3.org
On Wednesday, February 25, 2004, 8:42:03 AM, Etan wrote: EW> Dec 1, 2003, at 17:04 US/Pacific, Boris Zbarsky wrote to EW> <mailto:www-style@w3.org> on 1 Dec 2003 in "Re: [CSS 2.1] overflow of EW> replaced elements" (<mid:3FCBE515.4040101@mit.edu>): >> Yes, I realize you consider all form controls to not be replaced >> elements. Sorry, that doesn't fly. If a <textarea> were an >> inline-block, it would display the text that lives inside it (in >> textnodes). That's true only up until the point at which the user >> starts typing in it -- the question of whether those changes should be >> reflected in the page's DOM is a little vague, but last I checked with >> DOM WG people they said they should NOT be. EW> That is ridiculous. No, its sensible. The document is not changing. No mutation events are being generated. Instead, a local copy of the text contents are being edited, in the representation. These *might* be reflected back into the DOM at a later date. Or they might be sent to a server. or something. In particular, if there is a 'reset' button, the original contents of the text area will be used to get back to the original default editable contents. EW> If I edit a 'p' element, its DOM representation EW> should change, yes? What makes 'textarea' elements different? Well, if you edit a text area in the DOM its contents will change, too. If you type into a textarea, you are not editing the textarea element. EW> Boris, can you direct me to archived messages from the DOM Working EW> Group? This is normal and expected behavior; chasing up individual messages is not needed. -- Chris Lilley mailto:chris@w3.org Chair, W3C SVG Working Group Member, W3C Technical Architecture Group
Received on Wednesday, 25 February 2004 03:08:18 UTC