On 13/5/05 4:39 AM, "Antone Roundy" <[EMAIL PROTECTED]> wrote:

> Eric, was your point that the entry might be inheriting data from the
> feed which was updated after the entry was updated, and therefore, the
> entry might be thought of as having been updated after it's atom:updated
> timestamp?  Other than that, I don't see any problem--the entry was
> updated when at the time indicated by it's atom:updated element.

Yes. Which has implications in implementations. Consider an implementation
that stores entries in a local database, but doesn't bother doing a record
update if it later sees the same entry where /entry/updated hasn't changed.

> But back to my question--if for example, the feed/author was updated at
> the time indicated by feed/updated:
> 
> 1) If the author of the entry was NOT intended to be changed, the
> publisher should have put the old author into the entry.

true. not a problem.

> 2) If the author of the entry WAS intended to be changed, and the
> publisher considers this significant to the entry, then the publisher
> should have updated feed/entry/updated.

I agree. You can see how easy it would be to not do that though. It could
also be argued that the publisher has signalled the significance by updating
/feed/updated, and thus effectively /feed/entry/. The ambiguity bothers me.

> Because neither of those was done, from the XML shown above, we must
> assume that whatever change was made to the feed at the time in
> feed/updated did NOT affect the entry in a way that the publisher
> considered significant enough to update feed/entry/updated.  Another
> one for the implementors guide?

Yes, another for the implementors guide.

e.

Reply via email to