LaMont Ruley

Architectural Analysis of Systems Based on the Publisher-Subscriber Style

Architectural styles impose constraints on both the topology and the interaction behavior of involved parties. In this paper, we propose an approach for analyzing implemented systems based on the publisher subscriber architectural style. From the style definition, we derive a set of reusable questions and show that some of them can be answered statically whereas others are best answered using dynamic analysis. The paper explains how the results of static analysis can be used to orchestrate dynamic analysis. The proposed method was successfully applied on the NASA’s Goddard Mission Services Evolution Center (GMSEC) software product line. The results show that the GMSEC has a) a novel reusable vendor-independent middleware abstraction layer that allows the NASA’s missions to configure the middleware of interest without changing the publishers’ or subscribers’ source code, and b) a high-priority bug due to behavioral discrepancies, which were eluded during testing and code reviews, among different implementations of the same APIs for different vendors.

Visualization tools to support proposal submission

Many scientific observational programs require the field of view (FOV) or aperture to have a specific orientation on the sky. Since orientation requirements have a very strong impact on other aspects of the execution of the observation, an observer must have the ability to visualize the orientation of the science aperture and determine the effect of the orientation on the possible scheduling of the observation. We are prototyping an interactive, visual tool for fine-tuning the target location and orientation. To make efficient use of any instrument the user needs to understand the various modes of the instrument and then calculate exposure times or signal-to-noise ratios for many different kinds of observations. Thus, the exposure time calculator (ETC) is an essential tool that is used by various users for many different purposes. We are prototyping a more dynamic graphical ETC in which the user can simulate to some extent and determine the effect of various input parameters. This interactive exposure time calculator will not only be intuitive but will provide various users the different level of detailed information they desire. The VTT and ETC are Web-based tools that can be used by themselves or as part of the Scientists Expert Assistant, for the next generation space telescope proposal management system. Currently, the tools are being developed with the requirements of HST in mid, but will also be easily adaptable to other observatories. The underlying software for the tools is an object-oriented Java-based applet. The object-oriented nature of the design is intended to allow the tools to easily expand their features or to be customized. By making the system Java-based, we gain the ability to easily distribute the applet across a wide set of operating system and users. In addition to executing the tools as a Java applet, it can be loaded onto a users workstation and run as an application independent of a Web browser.

NGST's Scientist's Expert Assistant: evaluation results

This paper describes the approach and evaluation results of the Next Generation Space Telescope (NGST) Scientists Expert Assistant (SEA) project. The plan describes the goals, and methodology for the evaluation. The objective of this evaluation is to provide a means for the targeted user community to provide feedback to the developers, and to determine if the advanced technologies investigated as part of SEA have achieved the goals that were to be its success criteria. We can with confidence say that visual, interactive tools in SEA were found to be highly useful by the users. On a scale of 1 - 5, where 1 was excellent and 5 was poor, the SEA as a whole ranked as 1.7, i.e., between excellent and above average.

Expert assistant system to support the general observer program for NGST

One of the manually intensive efforts of HST observing is the specification and validation of the detailed proposals for scientists observing with the telescope. In order to meet the operational cost objectives for the next generation telescope, this process needs to be dramatically less time consuming and less costly. We are prototyping a new proposal development system, the Scientists Expert Assistant (SEA), using a combination of artificial intelligence and user interface techniques to reduce the time and effort involved for both scientists and the telescope operations staff. The advanced architectures and automation branch or Goddards Information Systems Center is working with the Space Telescope Science Institute to explore SEA alternatives, using an iterative prototype-review-revise cycle. We are testing the usefulness of rule-based expert systems to painlessly guide a scientist to his or her desired observation specification. We are also examining several potential user interface paradigms and explore data visualization schemes to see which techniques are more intuitive. Our prototypes will be validated using HSTs Advanced Camera for Surveys instrument as a live test instrument. Having an operational test-bed will ensure the most realistic feedback possible for the prototyping cycle. In addition, when the instruments for NGST are better defined, the SEA will already be a proven platform that simply needs adapting to NGST specific instruments.