AnnotationPage processing about trc

[ahankinson]

I believe the proposed solution means: 'Publishers may choose to order their Annotation Pages such that embedded ones are ordered before externally-referenced ones. This will expedite client processing by avoiding a dereference operation.' (Or something along those lines).

As it stands, the text is ambiguous -- the Publishers are not the ones who need expedited processing, but they can order their annotation pages such that the rule in the first sentence assists clients in expediting their processing operations.

However, there is a further conflict (I believe) in that the ordering of the items in an Annotation Page is significant (it is an @list) so recommending that the ordering of the items be done to expedite processing may conflict with the assumption that the ordering of the items has other meanings. I think this may cause problems with Publishers, in that they are on the one hand encouraged to expedite Client processing (i.e., be a good community member), while on the other possibly wanting to make use of the ordered nature of the list to convey some meaning (e.g., the order of display for time-based annotations).

I know the proposed solution hedges its bets here with 'may choose', but the flip side to expedient processing, which is ordering is significant, is not specifically called out to Publishers.

[mikeapp]

The first sentence:
Clients _SHOULD_ process the Annotation Pages and their items in the order given in the Canvas.
gives notice to publishers that the order is very likely (but not guaranteed ) to be respected by the client.

The second sentence highlights that the client can't expedite the processing of embedded lists if it respects the order of the items, so it is up to the publisher to consider this. For example, if the Canvas references an external Annotation Page before an embedded one, and the server hosting that Annotation Page is slow or fails to respond, then the user experience may suffer, as processing of the embedded page will be delayed as well. Publishers may choose to weigh this in their decision making as they consider the ordering of the items. The notion of "be a good community member" is not at issue here, user experience is.

[azaroth42]

Issue 22 (AnnotationPage processing)

+1: 27 [Siani81 aisaac andrewgunther awead azaroth42 dismorfo emulatingkat glenrobson hadro irv jbhoward-dublin jonhartzler joshuago78 jronallo jtweed julsraemy jwd mattmcgrattan mcwhitaker mejackreed mikeapp mixterj regisrob scossu tomcrane tpendragon zimeon]
0: 2 [cubap gigamorph]
-1: 0 []

Result: 27 / 29 = 0.93

Super majority is in favor, issue is approved