problem of the Internet 

qos: mechanisms to differentiate performance based on application or network-operator requirements
or provide predictable or guaranteed performance to applications, sessions, or traffic aggregates
innovations emerged in several areas
packet scheduling, admission control, traffic shaping
successful in constrained scenarios
VOIP
empirical load-based capacity planning
wrt interdomain, it went as far as it could go technically without economic (network market) support 
a few years in the lab can often save a few hours talking to a provider
economic technology just not there
also huge sociocultural resistance to paying more 
users think the Internet should just work
they've seen it happen before
tools to separate and service differentiate topologies now emerging in protocol specs
but still insufficient market support