Social Networks (SN) data has become ubiquitious. SN data is produced in
different places, in different forms,in different formats, for different
purposes. At personal scale the researcher and the user need to identify
relevant pieces of data,build out of them desired networks, and transformthem
for different purposes. This process requires the availability of techniques
and tools for data collection and integration, visualization, recording of
provenance, filtering,structured querying, etc. In this paper we present a data
management strategy toaddress these issues for SN in the range of dozens to
several millions of elements. We argue for the importance and necessity of such
an approach,and show its feasibility by presenting experimental results on
transforming and queryingSN using a dedicated data model and query language.