Proceedings of the 2005 conference on Applications, technologies, architectures, and protocols for computer communications

摘要: There were 255 submissions to the technical program at this year's SIGCOMM. The committee and external reviewers together generated over 1,100 reviews comprising of 45,000 lines text. meeting was held USC's Information Sciences Institute in Los Angeles on 18 April, 2005. location very appropriate, being a hub much early development Internet. After 9 hours discussion, we selected 27 papers seen here program, which hope you will find stimulating and, long term, some importance field networking.We made two changes way organized year. First, reduced size twenty an attempt increase amount informed discussion during meeting. This greatly increased workload members committee. We are grateful them for their magnificent response high quality reviews. Please join us thanking rendering valuable service community.Second, past several years SIGCOMM Call Papers has included explicit call category called "position papers". broaden vision what "SIGCOMM paper" might be, bring with future not necessarily constrained by is possible today's environment or prejudices. year reverted prior practice making position papers. still received that be considered papers, wait see whether resulting meets needs community terms broader selection papers.One result smaller reliance provide names these listed elsewhere proceedings. We, network research large, owe great debt, thank them.In geographic diversity, improvement previous years' conferences. Although from across globe, bulk U.S. (23), three U.K. one France. member Japan, U.K., Germany, addition 17 U.S.-based members.This year, Anja Feldmann T.U. Munich ran experimental "shadow committee", Europe Asia, give younger colleagues introduction functioning committees. written up her experiences running shadow PC article appear ACM Computer Communications Review.As community, impose large load reviewing process number submitted. Of course, submitted accepted rejected, lifeblood conference. But, as have noted past, we, selective other communities submit our It's unclear how resolve conundrum, stifling flow could poorer However, grows, potential authors learn gauge more appropriate SIGCOMM.SIGCOMM become than just conference, but focusing been job. Luckily, Roch Guerin executive freed chairs having worry about myriad organizational issues surrounding These define experience attendees, excellent work.