20090157902 | Virtual Networks - A virtual network has a plurality of nodes. Each node has the capability to provide a service to another node. Each node maintains a list for storing entries each representing a link to another node; each entry contains the address of the other node and a label identifying a service that that other node may provide. Each node also has a store for storing messages received from other nodes, these messages serving to propose a link and containing the identity of the node originating the message, a label identifying a service that that other node may provide and a label identifying a service that that other node requires. When a node needs a service that it is not itself able to provide, it searches the link list for a link having a label that matches the service needed, and in the event that such a link is found it transmits to the node identified by the link a message requesting the service. If, however, no such link is found, it searches the message store for a message identifying another node where the label identifying a service that that other node may provide matches the service needed and the label identifying a service that that other node requires matches the service that the node needing the service has the capability to provide. In the event that such a message is found it initiates the creation of a corresponding entry in the link list. If no such message is found, the node needing the service generates a message serving to propose a link and containing its own identity, a label identifying a service that it has the capability to provide and a label identifying the service that it needs. | 06-18-2009 |
20100100600 | DISTRIBUTING DATA MESSAGES - A method of automatically distributing data messages to members of a user community involves selecting, from a list of distribution rules, a rule which meets certain criteria in terms of message distribution characteristics. The selected rule determines which members of the user community will receive the message and when. For example, in a first round, a first subset of the user community receive the message and each is prompted to provide feedback data in relation to the message, for example to say that the message is offensive, already answered or if there is someone else who can deal with the message. This feedback data is applied to the selected distribution rule to determine to whom the message is sent in the next round. The process continues over a number of rounds until a termination criterion is met. | 04-22-2010 |