[JDEV] (no subject)

Chris Wilkes cwilkes-jabber at ladro.com
Mon Jan 27 17:22:25 CST 2003

On Mon, Jan 27, 2003 at 11:05:49PM -0000, Daniel MD wrote:
> hello, i was wondering what is wrong with the protocol that it can't
> take dynamic IP's ?

I'm not sure what you mean with that.  Do you mean that clients can't
have a dynamic IP address?  They can.  Servers can have it too as long
as a DNS name points to it.

> Jabber server, thus moving Jabber closer to a peer-to-peer model
> (currently this would require each device to have its own fully
> qualified domain name).

I think you're right in that if you want to send a message "directly" to
someone they will have to have a jabber server running on their end.

> I really would like to implement a p2p jabber network.

What do you mean "p2p"?  Do you mean that if A and B are talking through
a jabber server you should be able to send a message from A directly to
B, bypassing the server?

The problem with this is that if you start to go down the p2p route
you're going to introduce problems of identity.  How do I know the
identity of someone sending me a message?  Right now with a jabber
server there's some assurances (well there can be) that a person is who
they say they are.

If you can send pgp signed messages then that opens up some
possibilities of a p2p network.

Another problem I see with p2p is that if you're going to start doing
that on a larger scale you're going to get into a problem with spammers
send messages to you directly.  That's a problem right now with the
server model, but at least you can put a stop to it on the server.  With
limited bandwidth to cell phones and the like that could cause some
major problems.

Another practical problem with p2p is that if you're behind a firewall
then it isn't likely it will work without some major efforts.


More information about the JDev mailing list