Browse Prior Art Database

Self Assembling Hierarchical Index for P2P Network

IP.com Disclosure Number: IPCOM000015440D
Original Publication Date: 2002-Jun-13
Included in the Prior Art Database: 2003-Jun-20
Document File: 4 page(s) / 102K

Publishing Venue

IBM

Abstract

Self Assembling Hierarchical Index for P2P Network Self Assembling Hierarchical Index for P2P Network This concept solves the problem of how to build and maintain a distributed index of certified or authorized documents, files, data, programs etc. contained in a distributed point-to-point (P2P) network or Web. In the following description we define the document(s), files, data, programs, etc. as simply the 'document'. Documents may be copied, edited, and moved from place to place and different versions created and stored on multiple servers. It is not a replacement for existing indices or existing search technology; it is an enhancement to enable distributed indexing such as P2P networks to become more prevalent. Problems Solved: One of the biggest problems on a P2P network is validating the quality of data generated by an informal and fluid collection of information sources on a network. Nodes or servers come and go. Files can be downloaded or received from various sources but it is problematic to assess the quality of this data or information. This invention is motivated by the notion of the "primary source" or "original source" sought out by scholars when studying history. Anyone can generate information, expand on the work of a predecessor, modify a document, or create a new "original source". This invention uses voluntary collaboration (opt in) to create with every document and child document a real time index and history that records the history of that document and all children at the time of the documents creation. Furthermore, the history or index points back to the original primary source of the document. This root document contains an always up to date tiered taxonomy of the document and all of its children. A search engine which looks for this history, along with the related document, can then return the description of a document's history adding great value to those looking not just for a match to a query, but some information about the validity or origins of the information being retrieved.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 24% of the total text.

Page 1 of 4

Self Assembling Hierarchical Index for P2P Network Self Assembling Hierarchical Index for P2P Network

This concept solves the problem of how to build and maintain a distributed index of certified or authorized documents, files, data, programs etc. contained in a distributed point-to-point (P2P) network or Web. In the following description we define the document(s), files, data, programs, etc. as simply the 'document'. Documents may be copied, edited, and moved from place to place and different versions created and stored on multiple servers. It is not a replacement for existing indices or existing search technology; it is an enhancement to enable distributed indexing such as P2P networks to become more prevalent.

Problems Solved:

One of the biggest problems on a P2P network is validating the quality of data generated by an informal and fluid collection of information sources on a network. Nodes or servers come and go. Files can be downloaded or received from various sources but it is problematic to assess the quality of this data or information. This invention is motivated by the notion of the "primary source" or "original source" sought out by scholars when studying history. Anyone can generate information, expand on the work of a predecessor, modify a document, or create a new "original source". This invention uses voluntary collaboration (opt in) to create with every document and child document a real time index and history that records the history of that document and all children at the time of the documents creation. Furthermore, the history or index points back to the original primary source of the document. This root document contains an always up to date tiered taxonomy of the document and all of its children. A search engine which looks for this history, along with the related document, can then return the description of a document's history adding great value to those looking not just for a match to a query, but some information about the validity or origins of the information being retrieved.

Note that this invention is NOT a "distributed source control" system like Microsoft's Source Safe. Distributed source control retains a history of a single project being developed at multiple points on a network. The source control system exists at a single point as does the version data. In this invention, the individuals on the network are not working together to create or develop a single project. The copies and modified versions of a document need not be specifically related to the parent or enhancements of the parent (they might be parodies of the parent). Every document (that opts in) has associated with it a history and index pointing to its parent so every node on the network can, in principle, also act as a server.

This concept solves the following problems:

Where to store index information and how to distribute indices on a peer to peer network How to validate information found on a peer to peer network given a large nu...