[Standards-JIG] Re: What happened to the ACK proposal?

Jacek Konieczny jajcus at jajcus.net
Wed Aug 24 08:02:12 UTC 2005


On Tue, Aug 23, 2005 at 01:41:25PM -0600, Joe Hildebrand wrote:
> I was also not paying attention last week.  I continue to believe that
> this problem needs to be solved end-to-end, not hop-by-hop, in order
> to have been worth doing.

IMHO we need both end-to-end and hop-by-hop. And hop-by-hop as the
easier to do and giving a _big_ advantage over current implementations
should be implemented as soon as possible. I am sure that if the JEP
would be accepted (as experimental) several implementation would become
available in a matter of days. And a lot of user trouble would be solved
soon. Now we are talking, talking, arguing about who should standardize
and what and Jabber network is still lacking basic reliability.

Do we really want each client acknowledging every message and presence
sent to a MUC room? Should a directed presence or presence probe be
acknowledged by the endpoint?

Wouldn't it be good for servers to know if their s2s links work?
Shouldn't the servers (directly using a link) keep the links reliable,
restart when necessary etc. when something goes wrong or should the
client just wait for an answer for the endpoint and report an error when
it is not received (having no way to fix the problem with inter-server
links)?

Look at the encryption. We have a working hop-by-hop encryption and
people are happy with that. And we are still waiting for a good
end-to-end encryption. Should we stay with unencrypted c2s links just
because e2e encryption is better?

Greets,
        Jacek



More information about the Standards mailing list