Hard problems [Re: Standards]

[Margaret Wasserman]

>
> > When the hard problems come along, we don't do as much
> > architecture as we should.  But where we really seem to break
> > down is in doing careful and comprehensive systems analysis.

I agree.

I think that this is one of the things we wrestled with in
IPv6.  Some of the architectural changes we've introduced
in IPv6 (scoped addressing, autoconfiguration, etc.) will
have impacts on other parts of the system (routing protocols,
applications, etc.) but those impacts weren't fully analyzed
up-front.

Margaret