The server’s mechanism that collects data from participating sites is based on syndication via RSS or Atom, with the difference that the ScrutariData format is concerned with all data and puts particular emphasis on keyword indexing.

The Scrutari search engine is updated daily and keeps a log of all new references. Scrutari can therefore be used as a syndication source, making it possible to follow all web feeds of Coredem members.

With Coredem’s Scrutari search engine, two different Atom feed formats are available:

1) A reference feed: each new reference constitutes a specific entry in the syndication file; this file is the easiest to read but may result in a large number of entries at once when a new member joins Coredem. The address of this feed is:

http://sct1.scrutari.net/sct/coredem/feed/fiches-f_fr.atom

2) An update feed: a new entry is created for each search engine update (usually once a day), recording all new references. This feed is less sophisticated and comes in two versions:

• References sorted by site: http://sct1.scrutari.net/sct/coredem/feed/tree-f_fr.atom
• References sorted by document category: http://sct1.scrutari.net/sct/coredem/feed/categories-f_fr.atom

ScrutariJS, the advanced research interface on Scrutari, also provides a link to a syndication feed that includes all search results, and also provides a daily update of the search, including new results in the feed.