Kurt Maly

Harvesting social knowledge from folksonomies

Collaborative tagging systems, or folksonomies, have the potential of becoming technological infrastructure to support knowledge management activities in an organization or a society. There are many challenges, however. This paper presents designs that enhance collaborative tagging systems to meet some key challenges: community identification, ontology generation, user and document recommendation. Design prototypes, evaluation methodology and selected preliminary results are presented.

Interactive distance learning over intranets

Many distance learning systems claim to be interactive, but few can offer two-way video, on-the-fly interaction and application sharing. To address these limitations, our research group built the Interactive Remote Instruction (IRI) system, which links sites over a high-speed intranet, allowing students at geographically dispersed satellite campuses and community colleges to take a class together. Access from home PCs through a Windows NT port is planned but not yet available. IRI improves on the Old Dominion Universitys Teletechnet system in five areas: video resolution, asymmetrical video presence, interaction, teacher support and computer simulations.

Compressed tries

This paper presents a new data structure, called a compressed trie or C-trie, to be used in information retrieval systems. It has the same underlying m-ary tree structure as a trie, where m is a parameter of the trie, but whereas the fields of the nodes in a trie have to be large enough to hold a key or at least a pointer, the fields in a C-trie are only one bit long. In the analysis part of the paper it will be shown that for a collection of n keys the retrieval time, measured in terms of bit inspections of one key, is of the order logmn and the storage requirement of the order n·(m + log2n) bits. This improvement in storage requirements and retrieval time is achieved at the cost of decreasing the flexibility of the structure, and therefore updating costs are increased. First the C-trie is analyzed as a data structure, and then several methods of its use for relatively static databases are discussed.

HiFi: a new monitoring architecture for distributed systems management

With the increasing complexity of large scale distributed (LSD) systems, an efficient monitoring mechanism has become an essential service for improving the performance and reliability of such complex applications. The paper presents a scalable, dynamic, flexible and nonintrusive monitoring architecture for managing large scale distributed (LSD) systems. This architecture, which is referred to as the HiFi monitoring system, detects and classifies interesting primitive and composite events and performs either a corrective or steering action. When appropriate, information is also disseminated to management applications, such as reactive control tools. The outlined solution offers improvements over related works by supporting new monitoring techniques such as hierarchical filtering based monitoring and filter incarnation that improve the monitoring scalability and dynamism which are required for managing large scale distributed systems. The HiFi monitoring system has been implemented and used at the Old Dominion University for monitoring and steering Interactive Remote Instruction (IRI) which is a large scale distributed multimedia system for distance learning.

The UPS Prototype: An Experimental End-User Service across E-Print Archives

A meeting was held in Santa Fe, New Mexico, October 21--22, 1999, to generate discussion and consensus about interoperability of publicly available scholarly information archives. The invitees represented several well known e-print and report archive initiatives, as well as organizations with interests in digital libraries and the transformation of scholarly communication. The central goal of the meeting was to agree on recommendations that would make the creation of end-user services--such as scientific search engines and linking systems--for data originating from distributed and dissimilar archives easier. The Universal Preprint Service (UPS) Prototype was developed in preparation for this meeting. As a proof-of-concept of a multi-discipline digital library of publicly available scholarly material, the Prototype harvested nearly 200,000 records from several different archives and created an attractive end-user environment. This paper describes the results of the project. This is done in two ways. On the one hand, the experimental end-user service that was created during the project is illustrated. On the other hand, the lessons that the project team drew from the experience of creating the Prototype are presented.

An efficient data structure for the simulation event set

Recently algorithms have been presented for the realization of event scheduling routines suitable for general purpose discrete event simulation systems. Several exhibited a performance superior to that of commonly used simple linked list algorithms. In this paper a new event scheduling algorithm is presented which improves on two aspects of the best of the previously published algorithms. First, the new algorithms performance is quite insensitive to skewed distributions, and second, its worst-case complexity is O(√n), where n is the number of events in the set. Furthermore, tests conducted to estimate the average complexity showed it to be nearly independent of n.

Buckets: smart objects for digital libraries

Current discussion of digital libraries (DLs) is often dominated by the merits of the respective storage, search and retrieval functionality of archives, repositories, search engines, search interfaces and database systems. While these technologies are necessary for information management, the information content is more important than the systems used for its storage and retrieval. Digital information should have the same long-term survivability prospects as traditional hardcopy information and should be protected to the extent possible from evolving search engine technologies and vendor vagaries in database management systems. Information content and information retrieval systems should progress on independent paths and make limited assumptions about the status or capabilities of the other. Digital information can achieve independence from archives and DL systems through the use of buckets. Buckets are an aggregative, intelligent construct for publishing in DLs. Buckets allow the decoupling of information content from information storage and retrieval. Buckets exist within the Smart Objects and Dumb Archives model for DLs in that many of the functionalities and responsibilities traditionally associated with archives are pushed down (making the archives dumber) into the buckets (making them smarter). Some of the responsibilities imbued to buckets are the enforcement of their terms and conditions, and maintenance and display of their contents.

Towards Energy Efficient Change Management in a Cloud Computing Environment

The continuously increasing cost of managing IT systems has led many companies to outsource their commercial services to external hosting centers. Cloud computing has emerged as one of the enabling technologies that allow such external hosting efficiently. Like any IT environment, a Cloud Computing environment requires high level of maintenance to be able to provide services to its customers. Replacing defective items (hardware/software), applying security patches, or upgrading firmware are just a few examples of the typical maintenance procedures needed in such environments. While taking resources down for maintenance, applying efficient change management techniques is a key factor to the success of the cloud. As energy has become a precious resource, research has been conducted towards devising protocols that minimize energy consumption in IT systems. In this paper, we propose a pro-active energy efficient technique for change management in cloud computing environments. We formulate the management problem into an optimization problem that aims at minimizing the total energy consumption of the cloud. Our proposed approach is pro-active in the sense that it takes prior SLA (Service Level Agreement) requests into account while determining time slots in which changes should take place.

Arc: an OAI service provider for cross-archive searching

The usefulness of the many on-line journals and scientific digital lib raries that exist today is limited by the lack of a service that can federate them through a unified interface. The Open Archive Initiative (OAI) is one major effort to address technical interoperability among distributed archives. The objective of OAI is to develop a framework to facilitate the discovery of content in distributed archives. In this paper, we describe our experience and lessons learned in building Arc, the first federated searching service based on the OAI protocol. Arc harvests metadata from several OAI compliant archives, normalizes them, and stores them in a search service based on a relational database (MySQL or Oracle). At present we have over 165K metadata records from 16 data providers from various domains.

Mosaic p XTV = CoReview

Abstract CoReview, an interactive document and data retrieval tool, has been developed to provide a seamless environment to assist evaluation by groups and individuals, distributed across the Internet, to interact on the progression of a project. It can also assist individuals to interactively put together a document in a collaborative manner. CoReview is based on the strengths of the World-Wide Web server, Mosaic, and XTV, an X-Window teleconferencing system. While Mosaic will be used to manage the project documents and reviewer annotation files involved in proposals and their evaluation, XTV will aid us in real-time remote collaboration among a group of users. CoReview incorporates the XTV features into a user friendly graphical interface and enables Mosaic to be shared by multiple networked participants. The system architecture embeds the concept of a chair and a set of participants for each document to be managed. The CoReview chair manages the shared resources. CoReview allows for easy creation of a pool of reviewers or proposal writers and automates the process of creating the necessary infrastructure—daemons and directories—at the needed sites.