Current papers that propose new techniques and protocols often make assumptions about traffic characteristics that are simply not validated by real data. Hypotheses about the level of fragmented traffic, encrypted traffic, topology characteristics, traffic favoritism, path symmetry, DOS attack prevalence, address space utilization and consumption, directional balance of traffic volume, routing protocol behavior and policy, and distribution statistics of path lengths, flow sizes, packet sizes, prefix lengths, and routing announcements therefore yield questionable analytical results. Even in cases where analysis is based on data attainable by a researcher on his or her local campus, attempts to generalize typically lose integrity in the face of more complete or representative data sets.

This talk will show several examples of measurements that shed doubt on commonly assumed Internet myths. The implication is that the community could make much better use of its collective intellectual resources if we could validate ideas against a larger variety of empirical data sets before investing research and development time and energy on certain studies.