Date: Wed, 16 Jul 1997 07:33:28 -0700 (PDT) Message-Id: <199707161433.HAA19130@germany.it.earthlink.net> To: email@example.com From: Shelley Powers <firstname.lastname@example.org> Subject: Re: HTML 4.0 events and DOM requirements draft At 11:39 AM 7/16/97 +0200, you wrote: >Oops, I just realized you were refering to the requirements draft as >specified in the subject. I don't see any inconsistency in what's >written there but no matter what, if there are any they are not >intentional and will be addressed by the DOM WG. >-- >Arnaud Le Hors - W3C, User Interface Domain - www.w3.org/People/Arnaud > > I myself am surprised to see a listing of events with the HTML 4.0 draft spec. These are better suited to the DOM, which should also model the objects, their relations, and the exposed element properties as well as events and methods. I can see the HTML working group wanting to expose a technique for attaching events as tag attributes, but a better approach would be to use some form of name/action pairs, such as name=onclick action=function(), rather than trying to list several, a list that will only change. Already I have seen two events in use that are not in the event lists: onmouseenter and ondrgdrp. Until the object model has been created this list of events is unstable. Current popular browsers attach events to tags and the HTML group may want to formalize this. However, two of the most popular browsers are also introducing new, and better, techniques to use for event trapping. Can't the HTML working group consider deprecating these "intrinsic events", as a nod of awknowledgement of their existence, and then use a more generic approach of attaching events as attribute tags? Shelley ============================================================= Shelley Powers YASD email@example.com http://www.yasd.com http://www.dynamicearth.com IDG Books Professional Series book on Dynamic HTML, authored by Shelley Powers. Out early Fall.