RE: contentformat... etc. test

The process for the Object element should be that content is tasted until an acceptable content type for display is found. 

We are saying that CSS is not an acceptable content type, so if css is found as a response to a request for an object the algorithm should presumably continue with tasting object content, trying the fall back etc. and only then FAIL if nothing is found that is acceptable.

I think the problem here is that the relevant section of OBJECTS_OR_SCRIPT refers only to the mime types in the accept which includes css. Presumably we should change the reference in OBJECTS_OR_SCRIPT specifically to exclude css. I think there's an open LCC on the wording of this bit anyway.

Either way, I think we all agree that css is not an acceptable type for OBJECT and that this should count towards the external reference count?

Jo

> -----Original Message-----
> From: public-mobileok-checker-request@w3.org [mailto:public-mobileok-
> checker-request@w3.org] On Behalf Of Dominique Hazael-Massieux
> Sent: 20 August 2007 08:29
> To: Laura Holmes
> Cc: public-mobileok-checker
> Subject: Re: contentformat... etc. test
> 
> 
> Le dimanche 19 août 2007 à 12:18 -0400, Laura Holmes a écrit :
> > Unless I hear otherwise, I'm going to assume that if an included
> > resource is of type "text/css" and referenced from within the primary
> > document, and is not referenced within the <?xml-stylesheet.... tag or
> > link tag, it will produce a fail.
> 
> I read it the same way as you do, indeed.
> 
> Dom
> 
>

Received on Monday, 20 August 2007 14:05:21 UTC