[jdev] Communicate between two client instances of the same ID
Peter Saint-Andre
stpeter at stpeter.im
Wed Sep 3 09:19:49 CDT 2008
Kevin Smith wrote:
> On Wed, Sep 3, 2008 at 4:04 AM, Peter Saint-Andre <stpeter at stpeter.im> wrote:
>>> I think delivered and delivered once would be good enough for me.
>>> If not delivered for any reason, I'd like to be notified about it
>>> so that I'll resend it.
>> Message receipts (XEP-0184) probably give you what you want, then.
>
> No, it doesn't - we /really/ need to stop saying that 184 gives us
> reliable delivery, because it doesn't - it gives us consensual
> delivery auditing, or something similar, and that's not the same at
> all. 198 is pretty close to reliable delivery for most sane
> definitions, but 184 is a very long way off.
>
> Given that responding to the receipt requests is optional, how do you
> cope when you send a message but don't receive a receipt? Resend it
> once? Twice? To guarantee delivery, you have to keep sending it until
> you get a receipt, and infinite deliveries will take a very long time!
> Compare to 198, where the stanzas are always acked per-hop, and you
> know that your message will eventually reach its destination. If you
> want to have a chance of knowing when it reached, use 184 as well, but
> 184 on its own isn't what we want for reliable delivery.
Please see the subject of this thread: "Communicate between two client
instances of the same ID". I assume that "JLIST" wants to build some
kind of data sharing application that enables you to exchange data with
another device that you control. Since you control both ends, I further
assume that you know the other side also supports XEP-0184 and that
you don't consider receipts optional at all. In that contexts, XEP-0184
may yield the desired result. But only "JLIST" can decide for sure.
/psa
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 6751 bytes
Desc: S/MIME Cryptographic Signature
URL: <https://www.jabber.org/jdev/attachments/20080903/b5d2f8c8/attachment-0002.bin>
More information about the JDev
mailing list