Analyzing implicit group messaging: a novel messaging paradigm for group-oriented content distribution

Neil Cowzer, Paddy Nixon

Research output: A Conference proceeding or a Chapter in BookConference contributionpeer-review

1 Citation (Scopus)
16 Downloads (Pure)


Publish-subscribe systems are well suited loosely decoupled nature of the web, resulting in the messaging paradigm gaining widespread adoption and being the subject of much research. Such research has focused primarily on architectures and filtering algorithms with little evidence of performance analysis or characterization of user behavior in these widely deployed messaging paradigms. In this paper we discuss and examine implicit group messaging; an application-layer many-to-many messaging paradigm for delivering messages from publishers to specified groups of consumers. Such consumer groups are not addressed by explicit names, instead they are reached by describing the shared attributes or interests of consumers, forming easily defined implicit groups. Based on a 4 week experiment we analyze the characteristics of implicit groups and their usage. We find implicit group messaging workload to be similar to RSS in terms of group membership and update patterns; groups are typically small with few large examples and update rates vary from infrequent to more limited intervals.
Original languageEnglish
Title of host publicationProceedings of the IEEE Globecom 2010 Workshop on Pervasive Group Communications
EditorsXiaobo Zhou, Andres Kwasinsk, Liqiang Zhang
Place of PublicationUnited States
PublisherIEEE, Institute of Electrical and Electronics Engineers
Number of pages5
ISBN (Print)9781424488636
Publication statusPublished - 2010
Externally publishedYes
Event2010 IEEE Globecom Workshops - Miami, Miami, United States
Duration: 6 Dec 201010 Dec 2010


Conference2010 IEEE Globecom Workshops
Abbreviated titleGlobecom 2010
Country/TerritoryUnited States


Dive into the research topics of 'Analyzing implicit group messaging: a novel messaging paradigm for group-oriented content distribution'. Together they form a unique fingerprint.

Cite this