action items for research community

culture of & passion for sound measurement, as science & discipline
measurements need pedigrees describing them, how to navigate
audit trails, portable analysis scripting language to support reproducibility
well-managed meta-data
understand sampling implications and technology better
anonymization tools & reduction agents
simulation
need way to calibrate against real data (broken record)
safety tip: still has mostly no respect from providers.  diplomacy serves.
analysis 
find ways to assess opex versus capex of any new idea
or at least don't render it impossible to do so later
as scott encourages 'let's try to formalize some of what we (and providers) know'
as dave clark encourages 'it's about the $$$, stupid'
safety tip: providers do often quickly lose their patience w researchers 
continued/increased interaction with providers and vendors
nanog, ietf, caida 
ask for help from those who have succeeded
switch, router, measurement hardware vendors show more interest than ever
people don't trust `analyst estimates' anymore... good news in many ways

   take advantage of industry regrouping efforts & inclination to listen