Yeah good find. In our implementation we essentially have a special
way of querying a groups own index (that's what invoked via the
"with:public" or "group:group" syntax).

We have not yet worked out how to standardize this for federation.

On Nov 11, 9:53 pm, Mathijs <[email protected]> wrote:
> Hi all,
>
> I'm in the process of implementing most of the whitepaper about access
> control.
> The whitepaper mentions the INDEX permission address A can have over
> address B.
>
> I would like to know what's supposed to happen with B's history.
> Will A also receive all wavelets that have ever been sent to B?
> Or will the indexing "start" from the moment the edge gets defined?
>
> Since the index wave is generated on the fly (not stored on disk), I
> can choose to take INDEX edges into account when generating the index
> wave (optionally taking into account the date/time the edge was
> created).
>
> This will work for local wavelets, but I would like to know what would
> happen on federation.
> I know the protocol is still draft and these are still whitepapers, so
> I understand things might change.
> I just want to make sure I take a somewhat "safe" path, which is
> conceptually mostly the same as planned.
>
> So what would happen if a...@local gets INDEX, READ, ADD_ME on b.remote.
> The access edge(on which remote is authorative) will get federated to
> local.
> All wavelets on which b participates will be federated to local when
> they get updated, so local can generate A's index wave with all b-
> addressed waves in it.
> What if a wants to see all of b's history as well (if I join a group,
> I would like to browse the archive)?
> a is permitted to read all waves on which b participates.
> The only thing is: how does a know about all waves b is on?
>
> Should I write a custom protocolextension for this
> (like opening a remote index wave) ?
> Or is there a better solution?
>
> Any thoughts on this would be very welcome.
> Mathijs
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups "Wave 
Protocol" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to 
[email protected]
For more options, visit this group at 
http://groups.google.com/group/wave-protocol?hl=en
-~----------~----~----~----~------~----~------~--~---

Reply via email to