DRAFT On Changing the Culture of Empirical Internet Assessment

Over the last 20 years empirical assessment of the Internet has become a crucial task with two first-order goals. The first goal is to understand the vast global network as it exists. This informs our mental models on which we can then develop and extend the Internet in myriad ways. The second goal is to understand the problems facing the Internet in a concrete fashion such that the community can then develop solutions. While empirical assessment often appears straightforward at first blush, soundly investigating some facet of the network often results in subtle and nuanced analysis that was unforeseen at the outset of an investigation. While we are getting better as a community at thorough empirical investigation, more work in this direction is needed. In this paper we discuss two (related) problems faced by Internet empiricalists: