[
Lists Home |
Date Index |
Thread Index
]
Mike Champion wrote:
> Default attribute values are a ugly mess in the DOM even with
> DTDs, and introduce almost undecideable behavior when schemas
> are put into the picture (because of the lack of a standard
> processing model). My favorite scenario is that deleteing an attribute
> that has a default value -- according to DOM Level 1, but not
> consistently implemented -- should cause it to magically pop
> back into existence. The effects of this on the DOM event model
> are not obvious -- is there a delete event? A delete event and a
> create event? A modify event, if the value changed?
> If the attribute already had the default value,
> nothing has really changed, should the event(s) fire? Us "experts"
> can deconstruct the specs and logically figure out what should happen,
> but I'm sure that the poor bastards who stay up
> all night debugging their scripts and ultimately RTFM and figure this
> out will be looking for someone to blame. Hmm, maybe
> that's why I'm on every spam list on the planet, "poor bastards
> who stayed up all night" wrestling with the DOM getting their revenge
> by signing me up :~)
>
> The sanest solution is to stay as far as possible from PSVI-ish things.
But what does this problem have to do with the PSVI? The problem existed
long before the PSVI. Or are you saying that now that the PSVI is being
proposed as the way forward, this should either be fixed or the PSVI
avoided?
-- Ron
|
