[Standards-JIG] Hop-to-hop reliablility
pavel.simerda.lists at centrum.cz
Wed May 3 18:32:41 UTC 2006
On 2006-05-01 21:34, Peter Saint-Andre wrote:
> That said, it is possible that upon further reflection the 2005-2006
> Jabber Council would accept your proposal as a JEP after this thread has
> ended (perhaps along with proposals from Dave Cridland or others).
Is anything going on about per-hop acking? AFAIK there are two proposal, one
from Justin Karneges and one from Dave Cridland.
I liked Justin's simplicity... and scalability... acking groups of messages
within reasonable time. Pings when server/client unsure about the connection
(didn't get ack or didn't get data for a long time).
Dave's synchronization points can be an alternative but... I don't understand
them yet... it seems they're something like time-by-time both-sides
aknowledgements of previous stanzas.
Keep it simple... http://www.pavlix.net/
More information about the Standards