[standards-jig] gateway handling of legacy contact lists

Matthias Wimmer m at tthias.net
Thu Dec 11 23:49:12 UTC 2003


Hi Maqi!

It's not a big secret, that I already prefere the jabber:x:roster
solution for a long time ...

maqi at jabberstudio.org schrieb am 2003-12-11 22:32:39:
> The typical behaviour when registering with a gateway would be
> 1. Register
> 2. Receive a jabber:x:roster packet
> 3. Add all the contacts received
> 
> Cons:
> - this is inconvenient from the user's point of view

The client can autoaccept the items on a jabber:x:roster message after
it has just subscribed to a gateway.

> - some clients do not support jabber:x:roster, silently breaking
> auto-import then

And the old way breaks all clients that are connected to a jabberd2
server. Not better ...
And with the jabber:x:roster method we can even transmit nicknames for
the items to the client.

> - there are several unhandled cases, for example what happens if a
> user only adds SOME contacts? Should the gateway remove the others then?
> How should it determine which ones have not been added?

The user has the contacts on his roster for which the transport gets a
subscription request. I don't think the transport should already add the
contacts to the list of contacts of the user when it sends the
jabber:x:roster message but when it gets the subscription from the user.

> - the gateway's contacts are not NEW contacts, they are contacts that are
> present once the user has finished gateway registration so it's not really
> an application for jabber:x:roster

I don't see anything in JEP-0093, that this namespace is only for new
contacts.


Tot kijk
    Matthias

-- 
Fon: +49-(0)70 0770 07770       http://matthias.wimmer.name/
HAM: DB1MW                      xmpp:mawis at charente.de
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 189 bytes
Desc: Digital signature
URL: <http://mail.jabber.org/pipermail/standards/attachments/20031212/95e3db8d/attachment.sig>


More information about the Standards mailing list