Tim Bray wrote:
I'm with Mnot on this one. Just because it uniquely identifies an entry, that doesn't mean you can't have two versions of the same
entry in a feed. ... I don't see any reason for ruling them out in a
single feed.
Robert Sayre wrote:
We should probably be more worried about bad implementions totally missing the point of "identifiers", than about good implementations with sophisticated notions of versioning and identifiers.
With that in mind, I'm going to agree with Graham. What will happen is that super-simple implmentations that don't keep state will show both instances, because they don't bother with GUIDs. Then Graham will get bug reports saying "This feed only has nine entries in Shrook, but ten entries in TrivialServerSideScriptX, why is Shrook so broken?" We need to make that feed invalid, and Graham's life easier. Sophisticated implementations can assert common ancestry with an extension element.
Robert Sayre
