Re: [OPSAWG] [OPS-AREA] Heads up - Plea for allocating a /8 to ISPs
Martin Millnert <[email protected]>
| Newsgroups | gmane.ietf.v6ops,gmane.ietf.ops |
|---|---|
| Message-ID | <[email protected]> |
On Mon, 2010-11-15 at 14:02 -1000, David Conrad wrote: > Sorry if I was unclear -- the implication of "via IPv6 only" is that > you can't also reach it via IPv4. > > The point I was trying to make is that I suspect that any commercial > site or service that intends on becoming or staying viable is going to > have to provide access via IPv4 for the foreseeable future. The > installed base of IPv4-only is simply too large to do otherwise. Ah, sure (my bad I guess). Michaels point still stands though. For the record: The globally unique IPv4 and IPv6 Internets can *in theory* overlap today, if you ignore all hosts behind v4-NATs. This theoretical overlap will soon *definitely* be over. Each consecutive day following the tipping point, the two Internets can only grow more and more distinct, and even more so as the previously non-fully-participating Internet hosts, hidden behind v4-NATs (even v4-NATs of the future, used to reach non-transitioned content servers), joins the future v6-Internet. That is, the eventual "liberation" of many v4-NAT:ed hosts will make for a much more addressable Internet. Internet will live and die by this happening. A strict eyeball/content divided Internet is to me nothing more than a very far-reaching broadcast network of the more classical 20th century designs, and I think this is very unlikely to happen. Instead, I'm content the transition will happen, even "by itself" (the market), and I look forward to the media hype of January-May next year. There are a few things that can be done to make the transition go faster/more smoothly though, even without regulation. First on that list is more content available from a IPv6-only host, something that is improving on a daily basis already. Once large content providers agree that serving content on general IPv6 is OK, we'll have come a long way! Cheers, -- Martin Millnert <[email protected]> _______________________________________________ v6ops mailing list [email protected] https://www.ietf.org/mailman/listinfo/v6ops