CONEXT

It's a great pleasure to welcome everyone to ACM CoNEXT 2007 - the 3rd International Conference on emerging Networking EXperiments and Technologies (CoNEXT). With the first two CoNEXT conferences having been held in Toulouse and Lisbon and being founded as a merger of four existing EU workshops, CoNEXT has its roots in Europe. However, this year's CoNEXT is being held on the campus of Columbia University in beautiful and exciting New York City --- a locale that rivals the finest cities in Europe, and which emphasizes the global nature of both CoNEXT and our networking research community. We hope that this conference facilitates a stimulating exchange of ideas among many of the members of our international research community.

After Toulouse in 2005 and Lisbon in 2006, the 2007 edition of the Conference on emerging Networking EXperiments and Technologies (CoNEXT) has crossed the Atlantic Ocean to be held in New York, NY, USA.

This year, 138 high quality papers were submitted to the conference. We received papers written by authors from 29 different countries and covering a broad and diverse set of topics. Given the high quality of the submissions, selecting a program was both easy and challenging, as we had a plethora of worthy candidates but could not accept them all. The selection process was extensive and thorough and proceeded in three rounds during which all papers received at least three reviews, and often more.

The first round of the review process took place over the summer shortly after the submission deadline, at which time three TPC members were assigned to each paper. As one of CoNEXT's major objectives is to provide fair and accurate evaluations, TPC members were asked to review papers themselves (no delegation was allowed) and encouraged to provide detailed feedback to the authors. Each review was then cross-checked by the program chairs and a "sanity checker", who was a TPC member not among the original reviewers. The sanity check quickly assessed reviews for fairness and completeness. Based on the evaluations produced by this first round, the Program Chairs selected 84 papers for the second round that started in early September.

During the second round, the TPC members assigned to each paper read the other reviewer's opinions and debated the relative merits of the paper. The outcome of this discussion was either a unanimous recommendation on how to classify the paper in preparation for the TPC meeting, or a disagreement in which case additional reviews were requested inside and outside the TPC. Papers for which agreement could be reached were assigned to one of three categories: "accept", "discuss", and "reject". Papers for which no agreement was reached were assigned to the "discuss" category pending the completion of the additional reviews that were requested. The bulk of the papers landed in the "discuss" category, with only a handful of papers accepted up-front, and the remainder of the papers being viewed as not yet ready for publication.

The third round of the selection process was the TPC meeting itself, which was held in Philadelphia with TPC members either physically present in the room or participating through audio-conference. The bulk of the meeting was spent individually evaluating papers in the "discuss" category and ranking them as either above or below an "accept" threshold based on the calibration that emerged from the discussion phase. At the end of the day, we had identified 28 high quality papers that were selected for presentation at CoNEXT'07. The papers in the conference proceedings have been revised by the authors to account for the reviewers' comments and suggestions. The papers are organized in eight sessions that illustrate the breadth of topics that currently make up the field of networking, with not only papers on wireless, mesh, ad hoc and delay tolerant networks, but also papers proposing new architecture or techniques to better engineer or model networks and improve communication efficiency. In addition, the ever-present aspect of security, resiliency and economical issues also found their place in the program.

The CoNEXT'07 best paper award goes to the paper entitled "The Diameter of Opportunistic Mobile Networks," coauthored by A. Chaintreau, A. Mtibaa, L. Massoulie, and C. Diot from Thomson Paris Research Lab. The paper was unanimously selected from a highly competitive field by a committee consisting of three TPC members, who praised its combination of analysis and experimental validation.

As in 2006, CoNEXT'07 also awarded the "CoNEXT rising star award", which this year goes to Ion Stoica from the University of California at Berkeley. Ion was selected in recognition for outstanding research contributions early in his career on advances in Internet architecture, overlay networks and practical distributed systems. Ion will be giving the CoNEXT keynote talk in the first session of the conference. Please join us in congratulating Ion for this remarkable achievement.

In addition to the technical program, Suman Banerjee, Roger Karrer and Ashwin Sridharan have organized a wonderful workshop for students. It also attracted a record number of submissions, which both illustrates the high level of current interest in the field and bodes well for its future.

In parallel to the official CoNEXT TPC, Laurent Mathy, Philippe Owezarski and Olivier Bonaventure organized a purely educational shadow TCP whose objective was to allow young Ph.D. graduates to have an inside view into the paper selection process of a major conference. 78 of the papers submitted to CoNEXT received additional feedback from the 35 participants in the shadow TPC, who although coming primarily from Europe, also included members from USA and Asia.