panelist questions 

(1) how would you characterize past progress and current efforts to 
    characterize, understand or model behaviors of large-scale systems, 
    such as the Internet, where the artifacts being studied are complex 
    and require more detailed understanding of traffic behavior?

(2) it has been argued that `good data outlives bad theory'.
     what are impediments to acquiring a representative set of data points 
     that capture Internet behavior in some significant way and 
     what long-term vision is needed to overcome these impediments?

(3) the challenge of scale, coupled with the lack of configuration-specific  
    knowledge, calls for a more precise formulation of the measurement 
    challenge and associated research directions.  what role would the use 
    of inference techniques and simulation play in yielding insights into 
    behavior of current or future Internet protocols (transport, routing, etc)?
    what type of infrastructure must be available for resarchers not only
    to address current questions but new questions that cannot be 
    adequately addressed by current tools?