Margaret St. Pierre, founder and Chief Technical Officer of Blue Angel Technologies [1], introduces the MetaStar suite: an integrated web-based knowledge enabled package designed to assist organizations in making the most of their information resources. MetaStar enables organizations to forge knowledge-enabled stores and linkages, connecting employees, customers, suppliers and partners quickly, easily and cost-effectively using XML-powered solutions. The MetaStar components collectively work together to address each step of the knowledge management process. MetaStar Enterprise is comprised of several components and is essential for organizations requiring a complete information management solution rolled into a single package. MetaStar Enterprise addresses the entire knowledge-management process from data entry to automatic publication, allowing companies to maximally leverage their existing knowledge resources. Each tool in the MetaStar suite may be deployed independently or as an integrated suite of tools to best meet an organization's information needs.
 MetaStar provides a technology for organizations to establish knowledge-enabled stores and linkages, unifying data stores, employees, customers, suppliers and partners quickly, easily and cost effectively. The result is that organizations can leverage their often-obscured information resources to maximize their previously unrealized value. The MetaStar solution provides a variety of components for dealing with the entire knowledge management process. Each software component works on its own as a fully functioning application or taken together they provide a fully integrated solution (e.g.: such as is available with the MetaStar Enterprise software bundle). |
For example, the MetaStar suite provides a unifying non-proprietary interface across heterogeneous data sources spread across geographical locations without the need for cooperation. Current Internet search tools require users to access each individual data store that contains information of interest. If that's five or ten web sites or database systems, the user must spend a significant amount of time going to each of those sites and looking for the information they need. MetaStar eliminates this time consuming process. With one query, the software simultaneously searches all sites that contain the desired information and provides the user with one consolidated list of results.
MetaStar works the way the Internet was designed, that is, using a distributed model. In this model, information providers maintain their own databases locally. Information seekers directly access those local databases from their desktop rather than having to go through an information broker.
MetaStar provides high quality searching across both structured and non-structured information sources. Other search engines provide only non-structured searching or rudimentary searching of structured information.
The MetaStar suite provides information management tools that do not use proprietary solutions to lock customers into limited data sources through proprietary technology. The MetaStar tools are designed around various well-known international standards, which insures that customers will not be dependent on an individual vendor. All MetaStar tools are interfaced with and capable of managing information through the XML standard.
For those in need of a system to maintain information, MetaStar also provides a relational database solution that can manage XML data and provides insert, update, and delete capability through web browsers. In addition, MetaStar has built-in workflow capabilities.
The MetaStar suite works out of the box without programming. Therefore, customers can implement their applications within days. Competing technologies require significant installation, and implementation and on-going maintenance services. For example, the tools provide completely automatic workflow of records through simple configuration screens, all without programming.
MetaStar can automatically organize information statically or dynamically. This ability not only reduces the need for human resources, but also produces web sites very quickly with up to date information.
The MetaStar suite provides information management tools that handle not only text but such file types as graphics, sound and video.
The following case studies demonstrate how customers have used the MetaStar products to solve real-world problems:
A pharmaceutical customer was using a custom system to gather and process clinical trial data from hundreds of physicians geographically distributed across North America and Europe. The system took days to process pharmaceutical clinical trial results. Besides being slow, it was expensive to operate and maintain. In addition, each day information was delayed from being posted cost the customer hundreds of thousands of dollars, resulting in millions of dollars lost annually.
MetaStar Enterprise is replacing their current clinical trial consolidation system. MetaStar Enterprise is being used to capture, update and make information from clinical trial participants available in real-time. The primary benefit that Enterprise offers is that it completely automates the process of gathering clinical data and posts it instantaneously. Other reasons that Enterprise was selected include its ease of configuration and scalability.
U.S. Government agencies are required by law and executive order to create and maintain publicly accessible Government Information Locator (GILS) XML data that pertain to each agency's operations. The GILS data for the Defense Technical Information Center (DTIC) was maintained in a system that was labor intensive, costly and error prone. DTIC wanted a commercial off-the-shelf solution to reduce costs and to provide better service to the public. They also required that the solution work with their Oracle database.
MetaStar Enterprise enables defense department personnel located around the world to update GILS data from their web browser. When a record is updated, the system automatically sends an email to an approving authority to inform them that a record is waiting for review and approval. Once the approving authority approves the record, MetaStar Enterprise automatically publishes the record on the web for other users to search and retrieve.
If the approving authority rejects the change, an email is sent to the originator informing them that the record has been rejected and the reason for the rejection. The originator can make appropriate changes and re-submit the record if desired. The system also checks for records that have not been updated within a year and automatically sends email notification to the record owner, requesting them to update the record. If the record is not updated within the specified period, it is deleted from the system.
Because of the decision to use MetaStar Enterprise, the DTIC was able to reduce their costs and reassign employees to more productive work. DTIC is in the process of acquiring additional versions of MetaStar Enterprise.
The North Carolina State legislature passed a statute specifying that all state computer systems must be documented and the information made available to the public on the Internet. The state selected MetaStar software to meet the requirements of this law.
North Carolina Information Systems installed the MetaStar Enterprise solution along with Blue Angel's InterChange product. State personnel began by using MetaStar InterChange as a tool to populate their database. A Microsoft® Excel data spreadsheet was designed and distributed to state employees who entered data about individual systems throughout the state. This data included information related to agencies, applications, databases and data elements. The data from the spreadsheets was exported in XML format using MetaStar InterChange. The XML data was then imported into MetaStar Enterprise configured with the Microsoft SQL database.
Once the database was populated with the initial set of data, employees from around the state were able to update and modify the information from their web browser using the data entry component of MetaStar Enterprise. The workflow features available with MetaStar Enterprise enabled the state approval authorities to monitor all record changes, and to publish the records automatically on the web.
North Carolina uses MetaStar Enterprise to publish the information on the web in a number of different ways. One method is via an advanced search interface that permits a user to perform very targeted inquiries. For example, the user can search for specific values of a data element in a specific database maintained by a specific agency. The information is also made available in a browse format. Browsable trees are created dynamically based on the user's action. For example, if the user clicks on a specific state agency, all the applications for that agency are displayed. If the user clicks on one of the applications, all the databases for that application are displayed, and so on until the user gets to the specific information they need.
The Oak Ridge National Laboratory (ORNL) has scientists working throughout the world to gather scientific information about the environment and resources on the planet. The scientists frequently work in isolated locations and many do not have access to telephone, electricity or the Internet.
The challenge that ORNL faced was capturing scientist's information and making it available so that the rest of the scientific community could quickly and easily use the information. Due to budget constraints, they also needed to find a cost-effective way of accomplishing these tasks. The solution was to implement the MetaStar product suite.
Scientist's laptop computers were used to capture information in XML format. The scientists then placed the resulting files on an Oak Ridge web site and another MetaStar component crawls these web sites and harvests the information. The harvested information is automatically put into a structured format in a relational database by another MetaStar component. From there it is transmitted to a MetaStar Server which indexes the information and makes it immediately available for discovery via searching and browsing by scientists around the world.
The MetaStar suite was developed to provide an integrated system for handling information at each step in the knowledge management (KM) process [2]. Therefore organizations can knowledge-enable information stores and create linkages, connecting customers, suppliers and partners quickly, easily and cost effectively to the information they desire.
 Figure 1: Knowledge Management (KM) Framework |
Briefly, the following uses framework to describe the KM process. Later in this paper it is shown Blue Angel products relate to that framework. Typically, the KM process consists of several steps that are comprised of all or some of the following steps:
|
 Figure 2: The Blue Angel Product Components and the KM Framework |
Figure 2 outlines how Blue Angel product components relate to the KM Framework. The following briefly describes the functionality of each component. Following this description, the remainder of this paper examines in detail each step in the knowledge management process and describes how MetaStar relates to each step. Note that, Although MetaDoor products are shown in the figures, they are not discussed in this paper. Briefly, MetaDoor products are desktop applications that support the discovery and use of knowledge. Section 5 describes several facets of the capture process, and Sections 6, 7, and 8 describe the management, publication, and discovery steps, respectively. |
 Figure 3: Components forming the MetaStar Enterprise Solution |
Figure 3 shows the components that form the MetaStar Enterprise solution. Each component works together to address each step of the KM process. Information is entered, imported or gathered into the Repository. Once the information is ready to be published, Data Entry automatically publishes it to the Server. Users can then search and browse information from web browsers via the Gateway. |
Because information is typically available in a variety of forms, the MetaStar suite provides several methods to aid in the capturing of information. Each method provides a mechanism for placing information in the XML format that the MetaStar Repository imports or which can be indexed with the MetaStar Server. The methods provided by MetaStar for capturing information include:
The MetaStar Data Entry component permits users to enter records remotely over the Internet from a web browser. Data Entry is a Java Servlet, which operates with a number of web servers and operating systems. This tool has an easy-to-use interface that enables users to remotely insert, edit, and delete records in a few steps. The tool provides the following functionality:
Once the user has completed an operation, the Data Entry system delivers the entered information to the Repository's database (using the JDBC standard). The record is then processed according to the workflow process as specified by the Repository component. For example, the system can post the record immediately for general use or it can require the review and acceptance of a user(s) before it is published to the MetaStar Server. Users are notified in electronic mail that records are awaiting review. Users can review, accept, and reject records directly from their web browser.
Data Entry supports a variety of workflow options and customizations. For example, Data Entry provides a customizable interface, which can be integrated with other systems to effect various business functions.
Information is often available from a variety of off-the-shelf and custom applications, including real time data capture systems, relational database management systems, and the like. To assist users that need to import and export information to and from such systems, MetaStar provides the Translation Software Development Kit (SDK). This SDK enables users to interface external systems to the XML file format. Because a number of tools are available to process the XML format, the SDK can be used as a programmatic interface for moving knowledge to and from most external systems.
Web site harvesting entails running a robot to gather structured information from designated web sites or network directories. The MetaStar Harvester is a tool that gathers and parses HTML/XML, and extracts designated elements (e.g., title), HTML META tags (e.g., Dublin Core [6] elements), and XML tags. Using a text configuration file, the Harvester translates extracted information into either a structured or a custom file format. The configuration also has notations for instructing the Harvester to extract individual hierarchical and repeatable fields.
The value of this functionality varies depending on the application. For example, an organization, agency, or coalition with many web sites may each publish their data using HTML META tags. Once the HTML pages are gathered by the MetaStar Harvester, the Harvester could format the data for review, or the data could be imported into a management system where further operations could be performed.
Organize is the process of administering and manipulating structured information. The MetaStar Repository is as an administrative tool for managing XML data. The Repository manages XML data within an off-the-shelf ODBC-compliant RDBMS. Repository permits users to add and customize the structure of the data at anytime during its deployment. Repository also provides native support for a number of metadata standards, including the Dublin Core, FGDC, DIF and GILS.
The Repository allows users to:
The Repository also provides the ability to import XML data, and export the records in a variety of file formats, including text, SGML, XML, HTML, DIF, Text, and MARC formats. Although Repository provides default values for the import and export tags of each file format, users can customize the tags to meet their specific needs.
Using Repository requires understanding four basic concepts: templates, record sets, maps, and dictionaries. Templates are the equivalent of an XML DTD which define the properties of the XML elements. A record set stores the XML data, where each XML record is based on the elements defined by a template. The import (export) map provides a mapping between the elements in a template and the tags used in an import (export) of XML data. The dictionaries provide a vehicle for supporting controlled terms. The main feature of the Repository is its ability to perform powerful database operations that automate and maintain consistency between each component.
In addition to the above functionality, Repository supports the configuration of workflow behavior and the security that administers the MetaStar Data Entry system. In the absence of the Data Entry component, Repository operates as a standalone XML-enabled tool.
Access is the process that makes information available for others to use. MetaStar makes information accessible is several ways:
The following sections describe how MetaStar addresses publishing information using each of these methods.
MetaStar Server is a server application that enables searching and retrieving XML structured information over the Internet or a local Intranet. The server consists of three main components: the network protocol, the profile, and the search capability. The network protocol component manages the transport of information over the network using open network communications standards. The profile component ensures compliance with the various element standards. The search engine handles searching and retrieving of the elements using a full-text search engine.
The MetaStar Server makes information accessible using the ISO 23950 Search and Retrieval protocol. The MetaStar Server's architecture allows easy adaptation to new search and retrieval protocols, as they become popular.
MetaStar Server is integrated with several commercially available third-party search engines such as AltaVista, Fulcrum's SearchServer and Knowledge Network. The choice of which search engine to use is heavily influenced by a number of factors, such as its usability, features and functionality, cost, platform portability, scalability, search speed, and the quality of customer support provided. For example, if the search access points of a specific metadata standard require the ability to perform date or numeric range searching, or multi-character set support, the search engine selected will need to support this functionality.
MetaStar Server is used by the MetaStar Data Entry component to automatically publish information incrementally when a change is made by a user via their web browser.
Giving users the ability to browse information is particularly useful for users that do not know exactly what they are looking for. To aid users in navigating it is often useful to categorize the records. In this way, users can learn from a high level what information is available before descending to too low of a level where there may be large volumes of information.
To address this, the MetaStar Repository has the ability to examine the information contained in a record set and automatically generate a series of formatted HTML web pages linked together into a web Browse Tree. By forming the data into a web, it becomes possible for users to browse and find information, in addition to the search capabilities provided by the MetaStar Server.
The format of the generated web pages can be customized through the use of HTML templates. Simply modifying the HTML templates and regenerating the Browse Tree from the Repository can change the look-and-feel of the Browse Tree. The benefit of this feature is that it automates an otherwise manual process. In addition, the MetaStar Gateway can be used to generate web Browse Trees dynamically.
As described earlier, the Repository provides the ability to export data to files in a number of popular formats.
Use is the process whereby users go about finding information. There are two kinds of discovery:
MetaStar Gateway supports discovery by browse and query through dynamic web pages with its Gateway and Server components. Additionally, Repository supports static web browse pages through its export feature. The following sections describe the functionality provided by these tools.
The MetaStar Gateway lets users query one or more remote sites for information. The Gateway initiates searches concurrently to one or more databases residing on physically distributed servers on an Intranet or Internet. Because the Gateway uses a standard search protocol, it allows the user to simultaneously search any number of servers and merge the results into a single set of search results. This capability eliminates the time-consuming process of searching each data source serially and provides a uniform presentation of results. Since the Gateway connects directly to each information source, the information obtained is always the most current and up-to-date. This feature is particularly important when obtaining information from large database management systems. Like the MetaStar Server, the Gateway is engineered to provide portability across multiple protocols and XML structures. The MetaStar Gateway application is a Java Servlet that runs in conjunction with popular web Server.
A web Browse Tree is a series of HTML pages connected via hypertext links, where the leaves of the tree are the actual records. The records are organized categorically to enable users to browse the records in the subject areas in which they are most interested. A Browse Tree may be statically generated by the Repository application from a record set and a set of customized HTML templates. Alternatively, the MetaStar Gateway can be used to generate web Browse Tree dynamically. The ability to browse the resulting information is useful for users that are not sure what is available, or do not know how to formulate a query driven search.
For further information about Blue Angel Technologies' products or partner programmes contact Margaret St. Pierre at saint@BlueAngelTech.com.
If you have any comments on this article, please contact the editors (exploit-editor@ukoln.ac.uk).
Margaret St. Pierre
European Affairs Consultant
Address: P.O. Box 987, Valley Forge, Pennsylvania, USA
Telephone: 19482-0987
Email: saint@BlueAngelTech.com
 |
| Margaret St. Pierre is founder and Chief Technical Officer of Blue Angel Technologies. She is recognized as one of the leading experts in the Z39.50 protocol and related standards. Prior to founding Blue Angel, Ms. St. Pierre was a member of the Advanced Systems Development team at Thinking Machines Corporation, in Cambridge, Massachusetts where she was the lead architect and developer of expert systems. Ms. St. Pierre received her E.E. and M.S. degrees in E.E.C.S. from the Massachusetts Institute of Technology. |
For citation purposes:
Margaret St. Pierre,
"Blue Angel Introduces Metastar: An Integrated Knowledge Management Tool Suite,
" Exploit Interactive, issue 2, 20 July 1999
URL: <http://www.exploit-lib.org/issue2/blueangeltech/>
|
Issue Home | Editorial | Features | Regular Columns | News and Events | Et cetera | ||
|
| ||
| Go to Top |
A UKOLN Service. Contact Us. Copyright © 1999-2006 
|
Last Updated: 20 July 1999 |