1 Introduction

The Astrophysics Data System (ADS) provides access to the astronomical literature through the World Wide Web (WWW). It is widely used in the astronomical community. It is accessible to anybody world-wide through a forms based WWW interface. A detailed description of the history of the ADS is presented in the ADS Overview article ([Kurtz et al. 2000], hereafter OVERVIEW). The system contains information from many sources (journals, other data centers, individuals). A detailed description of the data that we get and how they are included in the ADS is presented in the ADS Data article ([Grant et al. 2000], hereafter DATA). The incoming data are processed and indexed with custom-built software to take advantage of specialized knowledge of the data and the astronomical context. A description of this processing is given in the ADS Architecture article ([Accomazzi et al. 2000], hereafter ARCHITECTURE). This article describes the development and the current status of the ADS Abstract Service user interface and search engine.

The ADS was created as a system to provide access to astronomical data ([Murray et al. 1992]). In 1993 the ADS started to provide access to a set of abstracts obtained from the NASA/STI (National Aeronautics and Space Administration/Scientific and Technical Information) project ([Kurtz et al. 1993]). The user interface was built with the proprietary software system that the ADS used at that time. The search engine of this first implementation used a commercial database system. A description of the system at that time is in [Eichhorn 1994].

In 1994, the World Wide Web (WWW, [www.w3.org 1999]) became widely useful through the NCSA Mosaic Web Browser ([Schatz & Hardin 1994]). The design of the ADS Abstract Service with a clean separation between the user interface and the search engine made it very easy to move the user interface from the proprietary ADS system to the WWW. In February 1994, a WWW interface to the ADS Abstract Service was made available publicly. The WWW interface to the ADS is described by [Eichhorn et al. (1995b)] and [Eichhorn et al. (1995a)]. Within one month of the introduction of the WWW interface, the usage of the Abstract Service tripled, and it has continued to rise ever since ([Eichhorn 1997]).

With the increased usage of the system due to the easy access through the WWW, severe limitations of the underlying commercial database system very quickly became apparent. We soon moved to an implementation of the search engine that was custom-built and tailored to the specific requirements of the data that we used.

In January 1995 we started to provide access to scanned journal articles ([Accomazzi et al. 1996]). The user interface to these scans provided the user with the capability to access the scans in various formats, both for viewing and for printing.

With time, other interfaces to the abstracts and scanned articles were developed to provide other data systems the means to integrate ADS data into their system ([Eichhorn et al. 1996b]).

With the adoption of the WWW user interface and the development of the custom-built search engine, the current version of the ADS Abstract Service was basically in place. The following sections describe the current status of the different access capabilities (Sects. 2 and 3), the search engine (Sects. 4 and 5), access statistics for the ADS system (Sect. 6), and future plans for the ADS interface and search engine (Sect. 7).