On 11/25/2013 10:48 PM, Lorenzo Colitti wrote:
On Tue, Nov 26, 2013 at 3:36 AM, Doug Barton <[email protected] <mailto:[email protected]>> wrote:DNS64 was a non-starter because there are always going to be IPv4 sites that hard-code IP addresses, and a non-trivial number of them are going to be critical sites for any given set of users. The authors chose to plunge ahead anyway, leaving us with yet another transition technology "cure" that is worse than the disease. Wait, what? The problem you describe is the one that 464xlat solves.
Didn't you just make my case? DNS64 didn't solve the problem, at best you can say it laid the groundwork for the (arguably) more thorough and (unarguably) dramatically more complex solution of 464xlat. And what was the delay between them?
I realize we're never going to come up with a canonical answer for the relative value of transition tech helping the transition vs. delaying it. And I can even see merit in what 464xlat provides. But IPv6 has always been particularly pathological about "let's shun the easy solutions in favor of creating increasingly complex ivory palaces," and that bothers me. Particularly when innocent folks like the OP put effort into deploying things like DNS64 because they bought the IETF snake oil that it will actually solve something for them.
Doug
