W3C home > Mailing lists > Public > www-dom@w3.org > July to September 2009

Re: Moving Form Events from DOM3 Events to HTML5

From: Steven Pemberton <Steven.Pemberton@cwi.nl>
Date: Wed, 23 Sep 2009 12:05:11 +0200
To: "Anne van Kesteren" <annevk@opera.com>, "Doug Schepers" <schepers@w3.org>, "www-dom@w3.org" <www-dom@w3.org>
Cc: XForms <public-forms@w3c.org>
Message-ID: <op.u0pfmxwesmjzpq@steven-750g>
On Tue, 22 Sep 2009 17:50:09 +0200, Anne van Kesteren <annevk@opera.com>  
wrote:

> On Tue, 22 Sep 2009 17:34:25 +0200, Doug Schepers <schepers@w3.org>  
> wrote:
>> Mark Birbeck wrote (on 9/22/09 10:53 AM):
>>> As you point out, XForms uses its own events, but really it should use
>>> the events you mention, and may do in a future version.
>>
>> In that case, it could refer to HTML5 for the definitions, or create a  
>> small dedicated spec for form-related events which extends DOM3 Events.  
>>   It should also contain an FormEvent interface the inherits from  
>> UIEvent, and the set of forms-centric event types.
>
> FWIW, what HTML5 does (which I think is fine) does not really amount to  
> defining the event. It just says at certain places in the algorithms of  
> the specification to dispatch an event called x. (With the appropriate  
> details being defined in each case.) If other specifications use an  
> event with the same name they would not need to refer to HTML5.
>
> E.g. HTML5 has a readystatechange event that is dispatched on the  
> Document object. It does not need to refer to XMLHttpRequest for that or  
> vice versa. They are separate events that just happen to have the same  
> name and all the same properties.

This is fine if they are identically named and have identical semantics.  
But if the semantics diverge and they then both turn up in the same DOM,  
you're in deep porridge. This is one of the values of defining one thing  
once.

Steven
Received on Wednesday, 23 September 2009 10:06:13 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Friday, 22 June 2012 06:14:03 GMT