RE: a methodical approach to defining what date elements we need

["Jeremy Gray" <jeremy@xxxxxxxxxxxxx>]

> From: owner-atom-syntax@xxxxxxxxxxxx
[mailto:owner-atom-syntax@xxxxxxxxxxxx] On Behalf Of Sam Ruby
> Sent: Tuesday, July 13, 2004 2:50 PM
> To: Norman Walsh
> Cc: Atom Syntax
> Subject: Re: a methodical approach to defining what date elements we need

-snip-

> Consider that a typical aggregator polls for data.  If you replace an 
> entry twice, and the aggregator missed the first one, then it has an 
> incomplete history to reconcile.

Worst case: the aggregator thinks it is an entirely new post.
Middle case: the aggregator notices that there is a pointer to a previous
version and visibly notifies the user that the entry is replacing one it
does not have on hand.
Best case: the aggregator notices the pointer and uses some
yet-to-be-defined mechanism to talk to the server to get the missing entry.

Even the worst case doesn't appear to be particularly problematic. Is there
something I'm missing?

Jeremy Gray