Advogato trust metric
From TrustLet, a free, collaborative project for collecting and analyzing information about trust metrics.
The Advogato trust metric is one of the first proposed trust metric. It was created by Raph Levien. The goal is to be attack-resistant. It is described here.
Advogato is also the name of the web site (advogato.org) created by Raph Levien for testing his trust metric. Advogato web site is a community of Free Software developers.
Contents |
[edit] Advogato community
Advogato (http://advogato.org) is an online community site dedicated to free software development, created by Raph Levien. Levien created Advogato also as a research testbed for testing his own attack-resistant trust metric, the Advogato trust metric. On Advogato users can certify each other on 4 different levels: Observer, Apprentice, Journeyer, and Master and the Advogato trust metric uses this information in order to assign to every user a certification level. The goal is to be attack-resistant, i.e. to reduce the impact of attackers.
[edit] Advogato trust metric
Advogato performs certification to three different levels: Apprentice, Journeyer, and Master. This is actually done by running the basic trust metric three times, using the "level" value in the certificate as a threshold. Thus, certification of Apprentices is computed using all certificates, while Master is computed using Master certificates only.
The computation of the trust metric is performed relative to a "seed" of trusted accounts. At the time of this writing (22 Feb 2000), the seed consists of raph, miguel, federico, and alan.
The core of the trust metric is a network flow operation. Informally, if there is a rich web of interconnections, flow reaches almost all the nodes. However, only a few accounts would be accepted from a large nest of bogus accounts, as long as there are only a few certificates from the "good" web to the bogus accounts. Those certificates represent a bottleneck in the network flow.
The complete description is available on advogato site.
An implementation of the trust metric in python is available in pymmetry and another one in net_flow Python wrapper. Both of them are in our code.
[edit] Advogato dataset
The Advogato dataset is available at http://www.advogato.org/person/graph.dot as a file in DOT format. It is a very big textual file (2 or 3 Megabytes) so it is better if you don't click on it but instead right-click it and save it on your hard disk.
An almost unique characteristics of the Advogato dataset is that is it weighted: directed edges between nodes are specified on a 3 levels base, Apprentice, Journeyer, and Master, which for instance can be mapped in [0,1] in the following way: Apprentice=0.6, Journeyer=0.8, and Master=1.0
A daily snapshot of the graph.dot file, collected since October 13 2007, is available in this directory: http://www.trustlet.org/datasets/advogato/
There are other communities who run the same software as Advogato and hence they produce the same graph.dot file and have similar structures: Robots.net, Squeak Foundation, Kaitiaki, persone.softwarelibero.it
[edit] Papers analyzing Advogato dataset
- Trustlet, Open Research on Trust Metrics by Paolo Massa and Kasper Souren
* add a paper!
