There has been quite a bit of discussion on proxy JIDs on the list and there has been some concern on complexity and some consideration of alternatives.
Here is a summary of my thinking. 1. The driving requirement for Proxy JIDs is support of JID hidden channels. There was a very strong message from the summit last year that semi-anonymous was vital, particularly because of JID harvesting concerns for public channels. Review of this requirement is on my list of questions for the summit. If JID Hidden Channels (equivalent to MUC semi-anonymous rooms) are not required, proxy JIDs are not needed. 2. While burner JIDs may be helpful to provide a user with complete anonymity in a channel, I think that channel administration needs access to the real JIDs. It would not be acceptable to manage a public MUC and just have a stack of anonymous participants. So use of client provided burner JIDs is not a viable approach to JID hidden channels. 3. Use of MUC style "reference by Nick" (as noted in my message of 6th Jan) is not viable. Kev's response to this set out a several reasons for this. To me, a key issue is "stable identity". Nicks can change, and it is quite possible that different people will use different nicks. This means that when referencing history there will be no clear way for a channel participant to see which individual set what. Addressing this issue was one of the original motivations for MIX. Although there are other reasons, this single reason rules out the approach. My conclusions from this is that we need to have Proxy JIDs in MIX, as I don't anticipate a change of requirements in 1. Steve _______________________________________________ Standards mailing list Info: https://mail.jabber.org/mailman/listinfo/standards Unsubscribe: [email protected] _______________________________________________
