Research and Education, especially higher education, have been key players in the development of various information technologies including the Internet. While advances in networking, computing, scientific research and education applications have been proceeding at a rapid pace, what has been lacking is a coordinated effort to capture, collect, or otherwise systematically organize the experience, knowledge and other product of the work done. We believe that a Knowledge Management Clearinghouse (aka Klearinghouse) can serve as a coordinating entity for the identification and use of tools for knowledge management in real time, any time, and over time.
The Klearinghouse will focus on three areas of information technologies:
1. People to People Communications - this is the space (kSpace) that knowledge grows and evolves from. This includes real-time (synchronous) and any-time (asynchronous) communications services such as video conferencing, web conferences, and electronic mail.
2. People to Content / Data - this is the knowledge base (kBase) or collections of information that are growing in number and size.
3. People to Work-Specific Technologies - these are the applications and technology tools (kTools) used to perform the research and educational activities. These tools are dedicated to serve specific purposes such as numeric analysis, genomic sequencing, or digital biological workbench.
The Klearinghouse will be a nexus for the identification, cataloging, evaluation, aggregation of state of the art applications, tools, information, and services in support of Research and Education processes requiring high performance, advanced networks and computational resources.
Early work has been done by the Distributed Application Support Team (DAST) of the National Laboratory for Applied Network Research (NLANR). The DAST activity developed a database schema for the first phase of a High Performance Connections Applications Database (HPC Apps Database).
The HPC Apps Database was designed to fulfill at least two functions: 1) it was to be designed as an open architecture (SQL, CGI, PERL) so that other institutions or organizations could make use of the design schema for their own data collections of advanced applications. Using like schema would then permit easier exchange of data from one collection to the other. And 2) it was developed to provide the research community a means to find out who-is-doing-what with high performance networks and computational resources as well as advanced applications.
The preliminary design of the HPC Apps Database was completed in late summer of 1999. The collection was initially populated with content about projects (meritorious applications) from the National Science Foundation High Performance Connections awards. At that time, there were approximately 140 projects loaded into the database based on the early connections awards. By September of 2000, there were more than 1100 projects identified, verified, and loaded based on information from later HPC awardees. The database team then began to provide more detail by adding in specific resources used by the projects. These resources include software applications, hardware, and other specific components used in support of the projects. There were 390 resources in the collection by the Fall of 2000. One can find the complete history, schema, and contents of the HPC Apps Clearinghouse at http://dast.nlanr.net/Clearinghouse/clearing_main.htm
The HPC Apps Database was initially intended to be part of a larger set of tools for the Research and Education community online. The Klearinghouse is a logical next-step to expand on the HPC Apps Database project. The Klearinghouse is designed to serve as a resource and tool kit for people who are either planning to use advanced technologies in their work or who are looking for other people who are already using such resources.
When initiated, The Klearinghouse will evolve to become a national center for the identification, collection, and dissemination of information about tools, resources, and projects that require high performance networks and computational environments. The Klearinghouse will work with select communities to identify computer mediated communications tools (e.g., Web base conferences, video conferencing, multicast services, etc., etc.) that would allow teams to work over distance and time. Additionally, The Klearinghouse will work with these communities to explore and develop ways to begin to establish information protocols in support of developing better knowledge management processes.
One outcome of the NLANR program has been the realization that the high performance network community as well as some of the discipline communities using the networks still do not know what other people or groups are doing with technology within or outside their own communities. As a result there is duplication of effort and missed opportunities. In addition, there has been a long-standing tradition in higher education to produce software locally on campus or in research lab. With improved knowledge management processes, The Klearinghouse will provide the Research and Education communities with the ability to identify information technology resources and to make better decisions about which tools to apply to their particular requirements.
In support of these knowledge management processes, The Klearinghouse will focus three areas:
1) People to People communications (Knowledge Space / kSpace);
2) People to Content / Data (Knowledge Base / kBase); and
3) People to work specific technologies / tools (Knowledge Tools / kTools).
A diagram of this structure is presented in Figure 1. Each of these three areas is addressed below.
"People to People communication" has been a key role of information technology for a long time. In the early days, Computer Mediated Communications was mainly electronic mail and news. Later, technology was used to build virtual laboratories for collaboration or collaboratories. Recently, much of the work of the Internet2 has been focused on the use of network-based video technologies for Research and Education.
The rapid growth of the World Wide Web has led to an explosion of tools and resources for individuals and groups to work together across distance and time with technology. Most of these tools have been developed for business communities and their customers. Currently, one can find Enterprise Information Systems (EIS), Customer Relationship Management Systems (CRMS), and Enterprise Portal Systems (EPS) and more available for purchase or license on the Web. There is not a centralized obvious source of information that provides the Research and Education community an overview or comparison of such tools. The Klearinghouse will provide that source -- enabling efficient and immediate use of these tools within the Research and Education communities.
Real-time communications continue to be a driving force in the use of high performance networks. Such applications include video which require high bandwidth along with quality of service guarantees to ensure that interactive sessions are technically acceptable to the participants. Added to these requirements are increased technical demands when other applications are used concurrently such as remote instrumentation or sharing applications across networks.
The Klearinghouse Knowledge Space proposes to identify these real-time tools -- both ready-to-use and under development. kSpace provides an easy-to-use collection of information about the tools along with links to where they can be observed or used online. Eventually, The Klearinghouse will create persistent links where its users connect with those resources identified to be the "best of breed" -- if a person is looking for a synchronous tool she will be able to find that tool in the kSpace directory and then, via the network, observe or participate in a session using the tool.
Asynchronous communication tools continue to be a mainstay of the net through email, USENET news, and World Wide Web. These resources continue to be much the same as they were in the past. However, there are aspects of even these basic tools that are subject to change such as new file formats, network optimization, and combinations of activities.
There is a recent emphasis in the business community to use Web browsers as the universal entry point and user interface for legacy computing and data systems. The Klearinghouse proposes to identify these new Web tools and recent additions in a way that will be useful and meaningful for the Research and Education community. As with the synchronous tools, we will provide links to "best of breed" with the asynchronous resources.
One additional element with asynchronous environments is the question of how to best capture synchronous events / sessions so that they then become useful in an asynchronous environment. The Klearinghouse proposes to work on this question by identifying and evaluating existing projects that are already beginning to capture and index network broadcast video or data streams such as projects at the University of California at Berkeley or the Carnegie Mellon University. A conceptual map of this process is illustrated in Figure 2.
The Klearinghouse will expand on the information collection about advanced applications that require high performance networks and computational resources. This database identifies work being done in various disciplines that is funded by a variety of agencies (federal and private). A frequently asked question is "Who is doing what with which technology?" This collection serves as a starting point in identifying the key personnel, the technologies implemented and where to go to find more information.
There are a number of information resources that relate research projects and education initiatives. However, not many of these are linked together either organizationally, technically, on the net, or in any other manner. The Klearinghouse will work to establish and improve these connections. At first, The Klearinghouse will capture / collect pointers to these collections in an adjunct collection for Technology Information Resources. During its first year, The Klearinghouse will work with other groups such as the Association for Research Libraries (ARL) and the Coalition for Networked Information (CNI) to explore implementation of dynamically linking collections via meta data, XML, or other emerging technologies.
There are a growing number of projects in Research and Education that are focused on developing applications or software tools. Examples of this include the Biology Workbench, the ACCESS Grid, the Grid Forum, and the Alliance Science Portals. In most cases, these tools are being developed to meet the needs of specific communities defined by discipline, technology, or project. At the same time, some of the underlying technologies of the tools are being generalized so that they might serve as platform for other tools. Therefore it will be important for The Klearinghouse to identify these tools to be added to the kBase.
In a process similar to kSpace, initially The Klearinghouse proposes to collect information about kTools to be added to the kBase. This information would provide discipline-based communities the ability to locate tools of special interest. The Klearinghouse will in fact have a number of the kTools in the knowledge base due to the fact that these are often listed as either projects or resources for high performance network applications.
What will be important later on is to explore how to better connect the individual to the specific community as well the reverse. In this case The Klearinghouse will need to work as a conduit or broker between individuals and communities. This activity can be done in a number of ways including: presentations at discipline oriented workshops or conferences; establishing liaison relationships with support personnel within the discipline comminutes; or developing documents for online and print based dissemination.
Rapid technological changes have impacted many aspects of how we participate in Research and Education. A major challenge has been how to "keep up" with the technological advances. Another challenge is how to keep abreast of current information in one's discipline. Also, there is the question of how can we, as researchers and educators, be more proactive in contributing to the body of knowledge in our respective specialties. These challenges are part and parcel with the implementation and integration of newly emerging software tools and networking systems into the daily activities of Research and Education.
The Klearinghouse plans to identify, develop and provide services that answer these and other questions related to the engineering and management of information / knowledge. While Knowledge Management itself is a complex environment, the Klearinghouse will set elements in place to provide a foundation for an evolving system to support Research and Education. The Klearinghouse will serve as a nexus for the identification, cataloging, evaluation, and aggregation of state-of-the-art applications, tools, information, and services in support of Research and Education processes requiring high performance, advanced networks and computational resources.
 Alliance Science Portals: http://www-pablo.cs.uiuc.edu/scienceportals/index.htm
George Brett, NCSA