Jinmei, On Thu, 2005-01-13 at 21:48, JINMEI Tatuya / çæéå wrote: > > BTW: do we really need this level of detailed inspection to meet the > two-implementation requirement for a DS? When I raised a similar > question when we discussed how we should deal with the M/O flags in > rfc2462bis wrt this requirement, I was told that we usually only > require a rougher level (e.g. whether there are more than two > implementations that support NS/NA/RA/RS/Redirect messages without > requiring line-by-line conformance to the corresponding RFC). I > personally prefer detailed inspection (if we can do that within a > reasonable period), but I can live with the rougher version as a > real-world compromise. In any case, we should basically be consistent > on the requirement level not to make a double-standard. >
In my opinion we do not need this level of detail. I think that the current discussion of removal of unimplemented or not widely implemented implementation options is overkill. Tim Hartrick Mentat Inc. -------------------------------------------------------------------- IETF IPv6 working group mailing list [email protected] Administrative Requests: https://www1.ietf.org/mailman/listinfo/ipv6 --------------------------------------------------------------------
