Microsoft's knowledge manager adept at pulling in data.
Part portal, part document manager and yet another part search engine, Microsoft Corp.s SharePoint Portal Server (released to manufacturers late last month), the companys first venture into the knowledge management space, has no shortage of features.
Considering that SharePoint is still in its infancy and given the current amorphous state of knowledge management, we can expect to see major evolution in this field as products become more mature. Although just about every vendor has a different opinion as to what features should be in a knowledge management system, everyone agrees that the goal of these systems is to aggregate data and put it at the fingertips of knowledge workers and decision makers.
So what does SharePoint do? In a word, it sucks-information, that is. Data from file servers, mailboxes and even Web sites outside a companys network are pulled into SharePoint and automatically placed in categories based on taxonomy settings established by the IT manager.
In its current iteration, SharePoint has the ability to collect information from Public Folders in Microsofts Exchange server and from Lotus Development Corp.s Notes databases.
Through contributions that Microsoft partner TopTier Software Inc. is making to Microsofts Digital Dashboard Developer Kit, Microsoft will eventually be able to give SharePoint users access to enterprise resource planning, customer relationship management and supply chain management applications.
SharePoint Portal Server lists at $3,995 for a stand-alone server and is licensed on a per-user basis at a list price of $72.
Behind the dashboard
The configuration of employee work spaces (which in SharePoint are called dashboards) is wizard-driven and fairly easy to accomplish. With a couple of clicks of the mouse we were able to add Web sites to a work space and publish content to Portal Server.
The SharePoint search engine component seems to work fairly well, but it will be interesting to see how well it will handle extremely large data loads. According to Microsoft officials, SharePoint servers configured solely for doing data searches should have a minimum of 550MB of RAM.
The SharePoint search engine has the ability to index numerous file types including files in Office formats and HTML and TIFF files (using the optical character recognition package that is included). To scan PDF files, sites must purchase an iFilter from Adobe Systems Inc. We were disappointed that this filter for reading documents was not bundled with SharePoint.
Security within SharePoint is linked to Windows file- and user-level security, which ensures that users will not be able to locate or even learn about the existence of documents (such as payroll records) to which they shouldnt have access.
When third-party applications and databases are added, SharePoint authenticates back-end databases supported through Kerberos, Windows 2000 or Windows NT 4. For other applications, Microsoft is creating a tool (which the company expects to ship in two months) that will map Windows identification to the security of the back-end database. Until that time, we dont recommend adding home-grown applications and unsupported databases to SharePoint if they contain highly confidential data.