Saturday, December 1, 2001

Taking QuickPlace to the next level of collaborative knowledge sharing

KNOWLEDGE MANAGEMENT

By Bain McKay

In this article, the first of six in a series on emerging second generation Knowledge Management technology, I'll discuss how we can leverage unstructured documents in QuickPlace (at http://www.lotus.com/home.nsf/welcome/quickplace) for collaborative knowledge sharing. Using the latest computing architecture, I'll describe a highly scalable pattern-based approach that can use multidimensional hyperspace dynamically to capture and harness knowledge concept relationships in documents, leveraging how we learn to optimize knowledge reuse.

In subsequent articles, I'll address why search can't work to clear the infoglut logjam, why First generation Knowledge Management came up short, and why a pattern-based approach can deliver on the promise to help us leverage what we know for business value.

The need

First generation Knowledge Management technology used compute-expensive mathematical algorithms to improve search and classification methods. But it failed to deliver solutions for the logjam of infoglut plugging our computer networks and decision-making processes. In fact, it added to the problem, being long on promise and short on delivery to the business bottom line, because it couldn't uncover and leverage the dynamic hidden-order that drives corporate results as reflected through the usage of concepts in and across business documents.

The approach

In this article, I'll address the architecture and technology in 2G-KM (second generation Knowledge Management) in some detail and will overview how it can be used to quickly and easily build powerful, easy to use QuickPlace e-collaboration environments -- or, more specifically, QuickPlace e-collaborative knowledge sharing (kCollaborative) environments.

I'll address the integration of QuickPlace with a modern MVC (Model2)-style architecture, separating presentation from model logic through a 2G-KM engine that leverages Java, Java Servlets, J2EE (Java 2 Enterprise Edition), and SOAP (Simple Object Access Protocol) XML (eXtensible Markup Language) Knowledge Transactions as J2EE knowledge transaction carriers, and therefore intelligent knowledge brokers that transport request-response payloads to targeted knowledge sites and users and back.

I'll discuss the application of some of the powerful offerings afforded us by leveraging 2G-KM in QuickPlace through the MVC Model2 architecture, providing first, second order and higher knowledge operators, scalability, high-semantic expressiveness, active-reflexiveness, dynamic intelligence, and much more.

Back to basics

But all this is just noise, of course, unless it maps back to how we work as human beings and facilitates our productivity in a globally competitive knowledge management economy. By how we work, I mean how we assimilate information, remember, think, and learn as well as how we leverage what we know through the reuse of corporate knowledge. These are the concerns and objectives of cognitive science and what Knowledge Management is all about.