Federated Content Management: Accessing Content from Disparate Repositories with IBM Content Federation Services and IBM Content Integrator

Readers' comments

Readers' comments (6) 

lockedThis discussion is now locked


Posted by Mr. Leonard Fox on 16 February 2010 at 9:24

Does not address the big issues in federation: Dynamic taxonomy normalization to find the same documents that have different descriptions in repositories without having to normalize all attribute values across all repositories
Cross-Domain Security –data privacy policies that entitle users to see or not see accessible repositories and its content
Allow for green-zones where repositories are unavailable because of time-zone, maintenance or other events excluding participation in the federation

Posted by Mr. Eric Edeen on 17 February 2010 at 12:38

Mr. Fox is quite correct that the issues he raised are not addressed by the book. The reason is that it is primarily concerned with IBM FileNet CFS and IBM Content Integrator, which do not provide much support for them. IBM FileNet CFS does provide limited support for the "green-zone" issue by allowing the Exporter to schedule when content discovery rules execute and the book does discuss this capability. Also, I believe IBM Content Integrator does offer some support for cross domain security.

Posted by David Werts on 18 February 2010 at 5:39

"Dynamic taxonomy normalization" is outside the scope of IBM Content Integrator. However, IBM Content Integrator could certainly be used as the communications enabler for a project willing to tackle such request.

Posted by David Werts on 18 February 2010 at 5:41

Single Sign-On capability is included with IBM Content Integrator. A user only needs to login once to the Single Sign-On system for authentication. The client application then retrieves repository credentials, as necessary, and handles authentication to the repositories transparently. A Single Sign-On SPI is included with IBM Content Integrator which allows clients to introduce credential data stores of their choice (LDAP, database, XML) if the reference implementation is not sufficient.

Posted by David Werts on 18 February 2010 at 5:44

As far "green-zones”, IBM Content Integrator provides a Federated Query Transformation (FQT) component which allows clients to develop their own before-and-after transformations of queries and/or data. Using FQT, green-zone capabilities could be easily added to the system.

Posted by David Werts on 18 February 2010 at 5:44

Even with the default product installation, federated queries complete successfully even if a participating repository is unavailable--instead of receiving results from that particular repository, the returned result set receives an exception indicating that the repository did not participate in the federated query. The results from the participating repositories will still be available.


Profile

Publish Date
21 April 2010


Rating: Not yet rated


Author(s)

ISBN-10
0738433977

IBM Form Number
SG24-7742-00