Computer scientists have made great strides
in how decision-making mechanisms are used.
BY VInCent ConItZeR
Based on the
pEopLE oFtEn mUst reach a joint decision even
though they have conflicting preferences over the
alternatives. Examples range from the mundane
(such as allocating chores among the members
of a household) to the sublime (such as electing
a government and thereby charting the course for
a country). The joint decision can be reached by
an informal negotiating process or by a carefully
Philosophers, mathematicians, political scientists,
economists, and others have studied the merits of
various protocols for centuries. More recently,
especially over the last decade, comput-
er scientists have also become deeply
involved in this study. The perhaps sur-
prising arrival of computer scientists
on this scene is due to a variety of rea-
sons, including the following:
Computer networks provide a new 1.
platform for communicating prefer-
ences. Examples include auction Web
sites, where preferences are commu-
nicated in the form of bids, as well as
Web sites that allow one to rate every-
thing from the quality of a product to
the attractiveness of a person.
Within computer science, there 2.
is a growing number of settings where
a decision must be made based on the
conflicting preferences of multiple parties. Examples include determining
whose job gets to run first on a machine,
whose network traffic is routed along a
particular link, or what advertisement is
shown next to a page of search results.
Greater computing power and 3.
better algorithms, as well as a more
computational mind-set in the general
public, have made it possible to run
computationally demanding protocols
that lead to much better outcomes. An
example is an auction in which bidders
can bid on arbitrary sets of items, rather than just on individual items (I will
discuss such auctions in more detail
later). Such protocols were once considered theoretical niceties that could
never be run in practice (to the extent
they were conceived of at all), but now
they are actually practical.
The paradigms of computer sci- 4.
ence give a different and useful perspective on some of the classic problems in
economics and related disciplines. For
Computer scientists are contributing
to and making use of microeconomic
Better algorithms enable new
marketplaces and other mechanisms
that lead to increased economic
Game theory and mechanism design
can be used to analyze and address the
problem of strategic users.