Editor's introduction

I am very happy to introduce Issue 7.2 of SIGecom Exchanges. This is the second issue of the Exchanges since its makeover. The makeover changed the focus from full-length articles to letters, in which authors can give a quick overview of recent research, review a recent conference or book, or lay out their opinions on where research is or should be heading. Full-length articles continue to be welcome, though. Issue 7.1 was a special issue dedicated to combinatorial auctions; in contrast, Issue 7.2 does not have a specific topic, resulting in a greater variety of topics related to e-commerce. Nevertheless, auctions continue to be well represented. The first contribution in this issue is an announcement of the book " Autonomous Bidding Agents: Strategies and Lessons from the Trading Agent Competition " by Wellman, Greenwald, and Stone. The book tells the story of the development and evolution of the Trading Agent Competition(s), gives detailed analyses of specific TAC techniques, and develops some general foundations for trading agent design. The next four letters concern incentive compatibility (also known as truthfulness or strategy-proofness). In " Truthful Opinions from the Crowds, " Jurca and Faltings review their work on incentivizing users to give truthful feedback—for example, by comparing their reports to a reference report and paying them accordingly. In " Implementable Allocation Rules, " Monderer considers in which domains (weak) monotonicity of an allocation rule is sufficient for it to be (truthfully) imple-mentable, improving on an earlier result by Saks and Yu. The letter " Characterizing Truthfulness in Discrete Domains " by Mu'alem and Schapira studies a similar topic, namely implementability in discrete domains (such as integer grid domains). Finally, in " Towards a Theory of Incentives in Machine Learning, " Procaccia explores a machine learning setting in which the labels of points in the input space are reported by different agents, and these agents must be incentivized to report them truthfully. The following three contributions concern issues in auctions (other than incentive compatibility). In their long contribution " Online Auctions and Generalized Secretary Problems, " Babaioff, Immor-lica, Kempe, and Kleinberg study auction settings where bidders arrive over time, and decisions on their bids must be made before other bidders arrive; they show how such settings can be seen as generalized versions of the well-known secretary problem, and they derive a number of results in this framework. In " A Modular Framework for Iterative …