> On Apr 17, 2021, at 9:49 PM, Phillip Hallam-Baker <phill@xxxxxxxxxxxxxxx> wrote: > > IPv6 isn't actually a 'protocol' as such. It might be useful to provide your definition of a protocol... > It is merely a data structure. You have described the messages only, which alone has never been enough to define a protocol. > It has a source address in a 128 bit domain and a destination address and a few flags and counters to stop loops, etc. and an extension mechanism That’s a step in the right direction, but omits quite a lot, e.g., longest prefix lookup of destination bawd tables that point to next hop interface and up address and more than a few other parts like definitions of most of those extensions and limits there upon. > allowing folk to throw pretty much anything else in. Not if you comply with the *protocol*. You can redefine the bits’ behaviors and meanings, but that would be defining a new protocol, at which point there’s little utility if any in keeping the bit patterns the same. Joe
