Regarding breaking backward compatibility - this compatibility affects only clients, right? Can we answer the question: exactly how would existing clients (and I'll bet we can enumerate all the available clients) be affected by a change in definition? How would the observed behavior of the clients be out-of-spec relative to the new definition?
I ask because I'd be willing to bet the observed behavior of existing clients will still be in compliance with the new definition: when the M bit is set, the client will attempt HCB and when the O bit is set the client will attempt ICB. When both are set, I'll bet the client tries HCB, which is still compliant with a new definition of M bit as a hint. - Ralph On Wed, 2005-05-25 at 10:47 +0900, JINMEI Tatuya / 神明達哉 wrote: > >>>>> On Tue, 24 May 2005 10:46:06 -0400, > >>>>> Thomas Narten <[EMAIL PROTECTED]> said: > > >> If we respect both the original sense of RFC2462 and our consensus > >> about the semantics separation of the M/O flags, I believe the right > >> solution is the following: > > > I think we should be careful NOT to get hung up on what the original > > intent of the M/O bits were, but focus on what the right behavior > > should be, given what we know now/today, and given the DHC protocols > > we actually have. > > In general, I agree (if I sound self-conflicting, note the "If" in the > cited part above). In practice, however, it should also be noted that > one major push back argument in the similar discussion for rfc2462bis > a year ago was that we should not introduce incompatible changes to > existing implementations that assume the "original intent". > > It seems to me that this is a typical case of tradeoff between "it's > never too late to do the right thing" vs "don't break compatibility". > In the previous discussion, the latter was preferred, and so we are > here. If we can now prefer the former argument, I'm personally > willing to take it. > > JINMEI, Tatuya > Communication Platform Lab. > Corporate R&D Center, Toshiba Corp. > [EMAIL PROTECTED] > > _______________________________________________ > dhcwg mailing list > dhcwg@ietf.org > https://www1.ietf.org/mailman/listinfo/dhcwg -------------------------------------------------------------------- IETF IPv6 working group mailing list ipv6@ietf.org Administrative Requests: https://www1.ietf.org/mailman/listinfo/ipv6 --------------------------------------------------------------------
