>> DKIM's ability to identify a domain owner is [also] bounded by >> whatever checks a registration authority imposes.
The most we can hope for from registration authorities is that domain names are relatively stable: whoever blah.biz is today is probably whoever it was last week. Having watched ICANN up close for the past year, I can assure you this will not change. The WHOIS data argument is deadlocked between two equal and opposing stupid positions, the IP lawyers on one side that demand that every domain contain complete info needed to sue them, and the EFF crowd on the other that demand unlimited anonymous speech using anonymous domains. (You may hope that I am making this up, but I am not.) There are some sponsored TLDs that actually do enforce registration rules, but the largest of them .TRAVEL has only about 10K registrations and all of them put together are far smaller than the 1.2 million registrations in .BIZ, the smallest of the generic TLDs and always will be. So forget that. >Precisely. We need to separate what DKIM does from what reputation >services do. Yes indeed. DKIM will be plenty useful in the short run to whitelist domains that system managers already know they like. Reputation services will broaden its utilitiy, but they're not essential. After all, how many of us have private lists of good and bad IPs for our mail servers, separate from what we get from DNSBLs? Just about all of us, I suspect. R's, John _______________________________________________ ietf-dkim mailing list http://dkim.org
