Pekka Nikander wrote:
"Good enough" ones are easy to generate without too much human intervention, for example, without any connection to the registry. OTOH, they are not necessarily unique, and therefore not "good enough" for some people. IMHO, both types are needed.
Pekka,
I still don't know if we really need the perfectly unique site local addresses. The question is: if you need perfectly unique addresses, why can't you use normal, global addresses? There we already have an existing registration process and standards. My thinking is that global addresses + statistically unique site-locals would cover most of the need we have, and it does not pay to construct complexity for the small amount of folks who would need perfect uniqueness. Of course, this is a judgement call. (You and me are perhaps the wrong persons to argue about this; input from corporate network managers and ISPs would be needed.) Jari -------------------------------------------------------------------- IETF IPng Working Group Mailing List IPng Home Page: http://playground.sun.com/ipng FTP archive: ftp://playground.sun.com/pub/ipng Direct all administrative requests to [EMAIL PROTECTED] --------------------------------------------------------------------
