BEA to color up apps with classification Graffiti
We meet again, IBM
BEA Systems has become the latest Silicon Valley tech company to succumb to the Web 2.0 bug by adopting a search strategy.
The Java application server and middleware vendor has announced Project Graffiti, which it billed as a "collaborative information and knowledge management system."
Project Graffiti will let users classify information using a set of planned tags and then discover data held in corporate repositories such as Microsoft's SharePoint and Windows file systems, IBM Lotus Notes and Documentum. It will also pool individual classifications and make suggestions about information that are useful to the end-user by combining search pattern analytics, BEA said.
The project accompanies a planned enterprise wiki codenamed Holland that will be designed for use by business customers and developers and is expected to leverage BEA's WebLogic Portal. Both Graffiti and Holland are due in the second half of 2006.
BEA's enterprise search focus means it is opening up yet another front in its middleware war with IBM. IBM is already mining the information management, search and collaboration markets through its combined WebSphere, DB2 and Lotus Notes offerings, combined with Global Services.
In a move designed to consolidate its presence by winning partner support, IBM this summer open sourced the Unstructured Information Management Architecture (UIMA), along with UIMA tools, that are used in IBM's WebSphere Information Integrator OmniFind Edition, WebSphere Portal Server and Lotus Workplace. UIMA provides a framework to conduct context-based searches of unstructured records held in repositories such as databases and email systems.
In May, IBM also partnered with Google to provide a desktop search function for Notes.
BEA, meanwhile, also announced an integration roadmap for its WebLogic and recently acquired Plumtree portals. Portlets and other page elements will be shared between the two portals in the first half of 2006. This will be followed by componentization of features into a set of common services tackling areas like search, distributed publishing, content aggregation and knowledge management due in the second half of 2006. Unification of “common” portal elements is planned for 2007.®