W3C home > Mailing lists > Public > public-html@w3.org > April 2008

Re: Nested forms

From: Henri Sivonen <hsivonen@iki.fi>
Date: Wed, 23 Apr 2008 18:26:35 +0300
Cc: HTML WG <public-html@w3.org>
Message-Id: <EB89C81E-BA3E-4AFE-A4BF-5C00183F147A@iki.fi>
To: Philip Taylor <pjt47@cam.ac.uk>

On Apr 23, 2008, at 16:31, Philip Taylor wrote:

> Henri Sivonen wrote:
>> Given the developments in unifying HTML5 and XHTML5 content models  
>> after WF2 was last revised and the outlook about the role of  
>> XHTML5, it would make sense to unify the conformance requirement  
>> regarding nested forms. That is, the spec should make nested forms  
>> non-conforming for XHTML5.
> I'd like to implement the UI described below - I did it a while ago  
> with nested forms, and couldn't think of a better way to do it, so  
> either it would be nice to know a better way or else it would be  
> nice for it to be conforming.

You could do it with non-nested forms and the form='' attribute in  
both HTML5 and XHTML5 without resorting to scripting in (X)HTML5 UAs.

Like <ul> in <p>, this is really about weighing the cost of HTML5,  
XHTML5 and DOM5 divergence against the benefit of marginal  
improvements to XHTML5 (that are only available if the author has paid  
the price of XML) and DOM5.

I tend to think that letting XHTML5 and HTML5 diverge is more  
problematic that saying that more DOM5 tree shapes are conforming,  
because systems producing HTML5 may use XHTML5 as input, so letting  
XHTML5 diverge quickly leads to having to observe an HTML5- 
serializable XHTML5 profile. I don't see harm DOM5 taking non- 
serializable shapes in apps that otherwise have so much JS state that  
the Web app state couldn't be easily serialized anyway without the  
participation of the app.

Henri Sivonen
Received on Wednesday, 23 April 2008 15:27:17 UTC

This archive was generated by hypermail 2.3.1 : Thursday, 29 October 2015 10:15:32 UTC