But some clients already have support for message filtering and privacy lists, and also it is possible to block whole domain by filtering it's server subnet or playing with name resolving (such as put line "1.2.3.4 jabber.spammershost.com" in /etc/hosts)
Blocking of particular domains using privacy lists is unlikely to work as spammers just buy new domain names every few days, just like they do today with email spam, same with blocking IPs, if spammers realise they are getting blocked by IP they will just start using their bot networks to provide a virtually unlimited supply of sacrificial IPs to use.
Message filtering at the client might partly work, but it wont help with bandwidth wastage between the client and server or the server to server traffic.
To solve the spam problem we need to develop some kind of system that works at the s2s level like just like dialback does, either some kind of server level message filtering (aka spamassassin), or some kind of web of trust or RBL system to identify rogue servers.
Richard _______________________________________________ jdev mailing list [email protected] http://mail.jabber.org/mailman/listinfo/jdev
