Prize committees have an easy job if they can compare submissions against each other. This committee did not have that luxury. Out of the 3 initial submissions only one made the threshold for the workshop.
It occurs to the program committee that the analysis of the interactions between the different uses of domain names within local and global contexts is almost a non-existent topic of research.
This may have to do with the lack of accessible data, the lack of a theory of root causes, or simply a lack of interest.
The best papers in the workshop (both prize and non-prize track) concentrated on data analysis without developing an overall theory for the underlying mechanism. We think this is evidence that the study of this global, centrally important, technical system that is not well understood needs to be ramped up.
We are placed in the difficult position to recognize that the best paper in the workshop is not eligible for the prize because it was not in the prize track.
That paper took a careful approach to understanding the possible systematic effects using short-term distributed measurements versus a less complete set of vantage points for a longer timeframe. The Prize Committee therefore want to acknowledge the quality of the paper by Matthew Thomas, Yannis Labrou, and Andrew Simpson on "The Effectiveness of Block Lists to Prevent Collisions".
The paper by Jim Reid, "Analysing the Use of the RA & RD bits in Queries to the Root Servers", argues that there does not seem to be demonstrable harm from entities that produce queries with the 'Recursion Desired" set, but it does not delve into the root causes for these queries. With the available data it seems that the author arrived at what is most achievable with the given data. As such the paper would be an ideal 2nd price candidate.
The rules are such that the committee cannot award a fraction of the prize money nor provide directions to the winner with respect to the prize. Even so the PC wants to stimulate further work in this area and awards the prize to Jim Reid with the hope that he will use the award towards further studying the root causes, and develop methodologies to assess how those root causes can be identified in various types of data and measurements.
Anne-Marie Eklund Löwinder
John R. Levine