- From: Ian Hickson <ian@hixie.ch>
- Date: Fri, 6 May 2005 15:26:59 +0000 (UTC)
On Fri, 6 May 2005, Brad Fults wrote: > > I don't really see a case outside the datagrid dragging (which is handled > elsewhere) where there would be a need to specify an element as draggable or > as a drag target without scripting in drag behavior. You need to have something to decide if it is an element expecting to be dragged in its entirety or an element where only the selection will be dragged, because otherwise you don't know how to start the drag. > Please feel free to point out cases where another HTML attribute would > have an advantage over a DOM property that could be set directly via > script. It's easier to initialise a content attribute than a DOM attribute. > As for the dragging event model for a single DOM element, I think the > events outlined in the current draft of the spec are a good start. In > addition, however, it would be nice to add some presentational niceties > via CSS pseudo-classes. The CSS aspect of things is separate from the logic side, we'll probably have to deal with the CSS side in the CSS working group. > For the droppable model, it makes sense to have the drag target element > specify either Elements or NodeLists that are droppable. The droppable model implemented by IE and Safari has the potential drop target simply respond to events to say whether things can be dropped or not. In general, though, we want to start simple and only add complexity to this model if use cases warrant it. The model that IE has is complex enough as it is without making it much worse. :-) -- Ian Hickson U+1047E )\._.,--....,'``. fL http://ln.hixie.ch/ U+263A /, _.. \ _\ ;`._ ,. Things that are impossible just take longer. `._.-(,_..'--(,_..'`-.;.'
Received on Friday, 6 May 2005 08:26:59 UTC