To exploit the collective intelligence thanks to the co-operative replication

For sometime we have worked on information recycling obtained by implicit co-operation and in particular on the recycling of the proxy caches contents. The first phase of this work led to a system which indexed the contents of HTML pages obtained from a proxy cache. The objective was to provide a search engine implicitly centred on an intranet user's fields of interest. The promising results of this work briefly described in the paper pushed us to further investigation and in particular to recycle the reusable large size objects contained in the cache We show in this paper that the strategy of reusing already downloaded information provides interesting advantages at low cost. In particular to speed up average accesses to the Internet to improve information search and to contribute to saving bandwidth on the Internet Moreover this approach comes within the conceptual framework of informational ecology and collective intelligence.