Jonathan Rosenberg, you raised a good point.
There are however some ISPs that are deploying NAT/Firewall (i.e. China,
Europe, Africa). In such case, DHCP might be useful. The ISP might do some
load balancing. Thus, DHCP method will provide a mean for the ISP to
configure dynamically third-end party devices. As for the security
information, this might be entered by the user to the third-end party device
(ex: same id/password as for the ADSL authentication).
There is a serious trust issue here. Is the ISP really going to issue a
username and password to every user of their network, entrusting them
with permissions to use midcom to manage port bindings on their network
wide NAT?? I certainly hope not. Thats an open invitation for
substantial denial of service attacks.