[Standards] Delayed Delivery for CSI and possibly SM

Daniel Gultsch daniel at gultsch.de
Tue May 30 13:39:20 UTC 2017


2017-05-30 15:00 GMT+02:00 Daniel Gultsch <daniel at gultsch.de>:
> I noticed that some CSI implementations (and maybe some SM
> implementations as well) add a delayed delivery tag where the from is
> set to their own domain.
>
> However that gets us into trouble in at least one scenario:
> Namely when parsing the timestamp of a unavailable presence.
> Unavailable presences come in two forms. Either they don't have a
> delayed delivery or they have one where from is their server.
> To reliably tell the the timestamp of an offline presence we have to
> be able to tell if it was delayed by our own CSI/SM or by the remote
> server. This is impossible for contacts that are on the same server.
> For this reason I propose that CSI and SM implementations if they
> choose to add a delayed delivery element it should come from the
> account and not the server.
> (Neither XEP mentions how to deal with delayed delivery but a lot of
> implementations do. So I'd propose to make it OPTIONAL to add it but
> if they add it MUST be the accounts bare jid)

To be clear: This is for server implementations and not for clients
resending messages after a SM resume.

Also it should be debated if we make adding delayed delivery tags a
SHOULD for SM and CSI since most implementations are doing this anyway
and for good reason. I just said OPTIONAL because it is the least
disruptive to the existing ecosystem. And I would really really like
to avoid a namespace bump.

cheers
Daniel


More information about the Standards mailing list