[Standards-JIG] Unique room names for JEP-45

Peter Saint-Andre stpeter at jabber.org
Mon Jun 19 22:08:30 UTC 2006


-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Nolan Eakins wrote:
> Peter Saint-Andre wrote:
>>> The list consensus seems to be that this is nice but not necessary.
>>> Perhaps it would be helpful to describe a recommended algorithm for
>>> clients to use in creating room JIDs that won't collide? I have no
>>> fundamental objections to asking the service for a room JID, but right
>>> now it doesn't seem necessary to define a wire protocol for it (since
>>> clients have everything they need to create a unique room JID).
> 
> Straight forward and simple way: a true UUID. I think it's a version 1
> UUID you want. The one that takes into account your IP address and the
> current time. That type of UUID would be more effective than the other
> variants out there such as completely random.
> 
> An alternative specific to converting a 1to1 chat to MUC: room_name =
> Hash(romeos_jid + juliets_jid + some_salt)
> 
> I'll leave Hash open since it probably doesn't need to be mandated. The
> two JIDs are for the people who are chatting one-on-one, and the salt is
> just a random string.
> 
> 
> I will state that Remko's suggestion seems familiar. There's something
> nice to be able to say to a service: just make me a room, I don't care
> about the name or config.

It's familiar because we have that in pubsub.

Peter

- --
Peter Saint-Andre
Jabber Software Foundation
http://www.jabber.org/people/stpeter.shtml

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.1 (Darwin)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFElyBeNF1RSzyt3NURAhkoAJ0asQaHT1nJxo1EuCGg6/OV5V+tXQCaA93w
O8jDyn3Co1dG1JKKm9HuI9U=
=V3G6
-----END PGP SIGNATURE-----
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 3641 bytes
Desc: S/MIME Cryptographic Signature
URL: <http://mail.jabber.org/pipermail/standards/attachments/20060619/413f0eb6/attachment.bin>


More information about the Standards mailing list