PubMed Health⌕ Search

PubMed · 8750388

Willow: a uniform search interface.

Abstract

The objective of the Willow Project is to develop a uniform search interface that allows a diverse community of users to retrieve information from heterogeneous network-based information resources. Willow separates the user interface from the database management or information retrieval system. It provides a graphic user interface to a variety of information resources residing on diverse hosts, and using different search engines and idiomatic query languages through networked-based client-server and Transmission Control Protocol/Internet Protocol (TCP/IP) protocols. It is based on a "database driver'' model, which allows new database hosts to be added without altering Willow itself. Willow employs a multimedia extension mechanism to launch external viewers to handle data in almost any form. Drivers are currently available for a local BRS/SEARCH system and the Z39.50 protocol. Students, faculty, clinicians, and researchers at the University of Washington are currently offered 30 local and remote databases via Willow. They conduct more than 250,000 sessions a month in libraries, medical centers and clinics, laboratories, and offices, and from home. The Massachusetts Institute of Technology is implementing Willow as its uniform search interface to Z39.50 hosts.

Explore related subjects

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

D S Ketchell, M M Freedman, W E Jordan, E M Lightfoot, S Heyano, P A Libbey. Willow: a uniform search interface.. https://doi.org/10.1136/jamia.1996.96342647

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related citations

Web servlet-assisted, dial-in flow cytometry data analysis.

BACKGROUND: The obvious benefits of centralized data storage notwithstanding, the size of modern flow cytometry data files discourages their transmission over commonly used telephone modem connections. The proposed solution is to install at the central location a web servlet that can extract compact data arrays, of a form dependent on the requested display type, from the stored files and transmit them to a remote client computer program for display. METHODS: A client program and a web servlet, both written in the Java programming language, were designed to communicate over standard network connections. The client program creates familiar numerical and graphical display types and allows the creation of gates from combinations of user-defined regions. Data compression techniques further reduce transmission times for data arrays that are already much smaller than the data file itself. RESULTS: For typical data files, network transmission times were reduced more than 700-fold for extraction of one-dimensional (1-D) histograms, between 18 and 120-fold for 2-D histograms, and 6-fold for color-coded dot plots. Numerous display formats are possible without further access to the data file. CONCLUSIONS: This scheme enables telephone modem access to centrally stored data without restricting flexibility of display format or preventing comparisons with locally stored files.

Computer Communication Networks↗

Networked ICD-9 coding system for a radiology department.

OBJECTIVE. We describe a networked database system to implement a departmental list of International Classification of Diseases, 9th edition (ICD-9) codes for use by reporting radiologists. CONCLUSION. Our system is reliable, easy to use, and has increased coding accuracy and speed. In-house development allowed designing a system to specifically address the needs of the radiology group. The resulting list of codes accurately reflects the current practice environment and grows dynamically to suit changing needs.

Computer Communication Networks↗