On Mon, May 23, 2005 at 04:20:44PM +0100, Graham wrote: > >---------------------------------------------------------------------- > ><atom:author> > > <atom:person><atom:name>Anne Rice</atom:name></atom:person> > > <atom:person><atom:name>Howard Allen O'Brien</atom:name></ > >atom:person> > ></atom:author> > >---------------------------------------------------------------------- > > > >then I've hacked around your restriction. That's still the same person > >listed twice. As I understand your wording, it's violating the spec - > >but it's undetectable. No way on earth any Atom processor that isn't a > >human being is going to notice, so how can we reasonably put that as a > >restriction in the spec? > > That's exactly why there's a restriction, because the processor can't > tell for itself what's going on, so it needs the publisher to provide > correct data about what's what. With the restriction, the processor > can safely treat them as separate people and tell the end user "This > entry was written by Anne Rice and Howard Allen O'Brien". Without the > restriction, all it can conclude is "The names Anne Rice and Howard > Allen O'Brien are in some way related to the authorship of this entry"
Why can't you conclude that the entry was written by Anne Rice and Howard Allen O'Brien? That's what it says (well, "authored"). It's the responsibility of the creator of the feed to ensure it makes sense - our job with Atom isn't really to try to define the semantics of authorship, just to provide a way of expressing authorship (no matter what your semantics are). I don't see a /technical/ reason for prohibiting this. None of the examples given cause me any problems, providing (as danbri says) that the spec makes it clear that we're not imposing these restrictions. Let the publishers decide what to say (because they will anyway, even if only in 0.001% of the cases and by mistake in an undetectable way). James -- /--------------------------------------------------------------------------\ James Aylett xapian.org [EMAIL PROTECTED] uncertaintydivision.org
