7.1. Introduction

Sources of information have been multiplying on the Web for several years, especially due to the success of news portals and social networks that produce information in real time. These flows of information can be kept and processed, often in RSS [RSS 03] and Atom [GRE 07] formats. However, it turns out that nowadays the amount of data which has to be analyzed daily is so large [HME 11] that a user may miss information of interest. Thus, a given user can be lost with so many sources and the frequency of updates [TRA 14]. Pub/Sub (Publish/Subscribe) systems (Redis [CAR 13], Scribe [ROW 01], Siena [CAR 01], Echo [EIS 00]) have been designed to face the problem of aggregating and delivering information of interest (bookmarks and topics) to end users.

For these reasons, we advocate a content-based Publish/Subscribe paradigm for Web 2.0 syndication in which information consumers are decoupled (in both space and time) from feeds (produced flows of items) and instead express...