Random-Sample Elections: Far Lower Cost, Better Quality and More Democratic

4 Oct
Computer Science Colloquium Series
David Chaum , Visiting Professor, KU Leuven (Belgium) and Entrepreneur
Thursday, October 4, 2012 -
4:00pm to 5:30pm
Maxwell Dworkin G125

A Random-Sample Election can be conducted regionally, nationally, or globally, with results that are more irrefutable than those of current elections but at less than one thousandth of the cost.  Example uses enabled by the ultra-low cost include petitions by non-government groups, competitively voted ladders for language to be put to vote, binding consultations of constituents by officials or parties, and even large-scale yet fine-grained direct democracy—something heretofore believed to be impossible.

The system works by randomly selecting voters and computing tallies in a novel way that lets anyone verify online that the process cannot have been manipulated. Voters are protected but are unable to sell votes.  Voters may also be better motivated and informed since each vote carries more weight and each voter can conduct research online into the single issue that voter is asked to help decide.

The approach is compared with elections today in terms of nine attributes, the technical concept along with its properties are sketched, historical context is discussed, and a variety of adoption scenarios are laid out. An example implementation, based on systems already proven in governmental elections, is sketched to confirm technical and practical feasibility.

There is no catch, this changes everything!