Draft
Copyright © 2004 by the Authors.
Permission to use, copy, modify and distribute the MPN Specification and its accompanying documentation for any purpose and without fee is hereby granted in perpetuity, provided that the above copyright notice and this paragraph appear in all copies. The copyright holders make no representation about the suitability of the specification for any purpose. It is provided "as is" without expressed or implied warranty.
The MPN-Interest module enables the connexion of individual weblogs with the goal to create a "web of interest". Within this network of weblogs, items can be passed around automatically. This module is designed for RSS 1.0 but can easily be adapted to other formats, e.g. RSS 2.0 and Atom.
It is common behaviour for weblog users to read a number of others' weblogs on a regular basis. This combination of weblogs represents the personal interests of a user. He/she is likely to publish similar items on hir own weblog. The user may also write comments on the others' entries and recommend them to hir own readers. It is the purpose of the MPN-Interest module to make this process of recommendation to happen automatically. As an effect a decentralised reputation system is created on the basis of the social network structure of the blogosphere. The "web of interest" is a central component which is defined by this module. A comprehensive explanation of the concept is given in the paper 'Be a Freeporter!': Enabling a Mobile News Publishing Community.
This module provides a value to indicate the interestingness an item has to a weblog owner. While an item travels from one weblog to another, this value is usually decreased until a threshold where it stops its travels. The value can also be maniputated by the weblog owner to give an explicit rating. This may enable the item to reach additional destinations.
A second component enables the users to monitor and thus to control this automatic process. It records the way an item took through the network. Thus, the users can open up connexions to good sources and close connexions to uninteresting (or spam) sources.
interest
Indicates the amount of interest a weblog owner has in an
item. Floating point values in the interval [0..1] are
acceptable. 0 means no interest, 1 means full interest. If
omitted a value of 1 is assumed.
derivation
indicates whether the interest value was calculated
automatically ('automatic') or that it was determined manually
('manual') by the weblog owner. If omitted 'automatic' is
assumed.
via
Contains a sequence of references to the weblogs this item came
from. The sequence is empty when an item is initially published
and grows with every hop the item travels. The references shall
point to the locations of the feeds and not to the associated
websites in general.
<item> Elements:
<?xml version="1.0" encoding="utf-8"?> <rdf:RDF xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:in="http://purl.org/rss/1.0/modules/interest/" xmlns="http://purl.org/rss/1.0/"> <channel rdf:about="http://www.bloggermatrix.com/tom.jsp"> <title>Tomsby E.</title> <link>http://www.bloggermatrix.com/tom.jsp</link> <description>Tomsby E. Moblog</description> <items> <rdf:Seq> <rdf:li resource="http://www.chaosradio.ccc.de/cr89.html" /> <rdf:li resource="http://www.bloggermatrix.com/tom.jsp?n=6" /> </rdf:Seq> </items> </channel> <item rdf:about="http://www.chaosradio.ccc.de/cr89.html"> <title>Social Software - Social Networks</title> <link>http://www.chaosradio.ccc.de/cr89.html</link> <description> Die fortschreitende Verbreitung des Internets und vor allem dessen ständige Verfügbarkeit über "Flatrates" lässt auf allen Ebenen neue Anwendungen aufsteigen, die vor allem auf die direkte Vernetzung der Anwender zielt. Nachdem Instant Messaging schon vor einiger Zeit seinen Siegeszug angetreten hatte steht jetzt die "soziale Vernetzung" zunehmend im Vordergrund. </description> <in:interest>0.6</in:interest> <in:derivation>automatic</in:derivation> <in:via> <rdf:Seq> <rdf:li resource="http://www.ccc.de/updates/chaosupdates.rdf" /> <rdf:li resource="http://somewhere/somefeed.rdf" /> </rdf:Seq> </in:via> </item> <item rdf:about="http://www.bloggermatrix.com/tom.jsp?n=6"> <title>Mobile Entertainment</title> <link>http://www.bloggermatrix.com/tom.jsp?n=6</link> <description> A conference on mobile entertainment ... </description> <in:interest>1.0</in:interest> <in:derivation>manual</in:derivation> </item> </rdf:RDF>
It would be possible to record detailed information about the path of an item in the via block. But we think that it should be sufficient to provide the interest and derivation values for only the last feed to reach the proposed goals. Thus the system is simpler. If one is interested in more detailed information it is possible to look it up in the other feeds, referenced in the via sequence.