Ethan V. Munson

Is 100 Milliseconds Too Fast

For years, software engineers have been told that applications must respond to user events within 100 milliseconds to seem instantaneous, yet this assumption has never been systematically tested. In this research, we attempt to establish thresholds of detection for changes in a graphical user interface using adaptive tracking. For keyboard interactions, subjects did not notice delays of approximately 150 milliseconds. In contrast, for mouse interactions, subjects did not notice delays of up to 195 milliseconds. Given these findings, further research is clearly needed to firmly establish lower bounds on application responsiveness so that software and operating system engineers can more precisely tune the interactive real-time responsiveness of their systems.

An infrastructure for development of object-oriented, multi-level configuration management services

In an integrated development environment, the ability to manage the evolution of a software system in terms of logical abstractions, compositions, and their interrelations is crucial to successful software development. This paper presents a novel framework and infrastructure, Molhado, upon which to build object-oriented software configuration management (SCM) services in a SCM-centered integrated development environment. Key contributions of this paper include a product versioning model, an extensible, logical, and object-oriented system model, and a reusable product versioning SCM infrastructure, that allow new types of objects to be implemented as extensions of the system models basic entities. Versions and configurations of objects are managed at different levels of abstraction and granularity. A new SCM-centered editing environment or development environment for a specific development paradigm can be rapidly realized by re-using Molhados infrastructure and implementing new object types and their associated tools. This paper also demonstrates our approach in creating prototypes of SCM-centered development environments for different paradigms.

Using peer-led team learning to increase participation and success of under-represented groups in introductory computer science

This paper describes the implementation and evaluation of a program that uses active recruiting and peer-led team learning to try to increase the participation and success of women and minority students in undergraduate computer science. These strategies were applied at eight universities starting in the fall of 2004. There have been some impressive results: We succeeded in attracting under-represented students who would not otherwise have taken a CS course. Evaluation shows that participation in our program significantly improves retention rates and grades, especially for women. Students in the program, as well as the students who served as peer leaders, are uniformly enthusiastic about their experience.

Software Configuration Management for Product Derivation in Software Product Families

A key process in software product line (SPL) engineering is product derivation, which is the process of building software products from a base set of core assets. During product derivation, the components in both core assets and derived software products are modified to meet needs for different functionality, platforms, quality attributes, etc. However, existing software configuration management (SCM) systems do not sufficiently support the derivation process in SPL. In this paper, we introduce a novel SCM system that is well-suited for product derivation in SPL. Our tool, MoSPL handles version management at the component level via its product versioning and data models. It explicitly manages logical constraints and derivation relations among components in both core assets and derived products, thus enabling the automatic propagation of changes in the core assets to their copies in derived products and vice versa. The system can also detect conflicting changes to different copies of components in software product lines.

The software concordance: a new software document management environment

In this paper, we describe the efforts of Juniper Networks to implement a Feature Guide documentation manual and discuss the usability merits of this documentation method.

Using versioned tree data structure, change detection and node identity for three-way XML merging

XML has become the standard document representation for many popular tools in various domains. When multiple authors collaborate to produce a document, they must be able to work in parallel and periodically merge their efforts into a single work. While there exist a small number of three-way XML merging tools, their performance could be improved in several areas and they lack any form of user interface for resolving conflicts. In this paper, we present an implementation of a three-way XML merge algorithm that is faster, uses less memory and is more precise than existing tools. It uses a specialized versioning tree data structure that supports node identity and change detection. The algorithm applies the traditional three-way merge found in GNU diff3 to the children of changed nodes. The editing operations it supports are addition, deletion, update, and move. A graphical interface for visualizing and resolving conflicts is also provided. An evaluation experiment was conducted comparing the proposed algorithm with three other tools on randomly generated XML data.

Assessment of factors impacting success for incoming college engineering students in a summer bridge program

A summer bridge program for incoming engineering and computer science freshmen has been used at the University of Wisconsin-Milwaukee from 2007 to 2010. The primary purpose of this program has been to improve the mathematics course placement for incoming students who initially place into a course below Calculus I on the math placement examination. The students retake the universitys math placement examination after completing the bridge program to determine if they then place into a higher-level mathematics course. If the students improve their math placement, the program is considered successful for that student. The math portion of the bridge program is designed around using the ALEKS software package for targeted, self-guided learning. In the 2007 and 2008 versions of the program, both an on-line version and an on-campus version with additional instruction were offered. In 2009 and 2010, the program was exclusively in an on-campus format, and also featured a required residential component and additional engineering activities for the students. From the results of these four programs, we are able to evaluate the success of the program in its different formats. In addition, data has been collected and analysed regarding the impact of other factors on the programs success. The factors include student preparation before the beginning of the program (as measured by math ACT scores) and the amount of time the student spent working on the material during the program. Better math preparation and the amount of time spent on the program were found to be good indicators of success. Furthermore, the on-campus version of the program is more effective than the on-line version.

Inkteractors: interacting with digital ink

Digital inking systems accept pen-based input from the user, process and archive the resulting data as digital ink. However, the reviewing techniques currently available for such systems are limited. In this paper we formalize operators that model the user interaction during digital ink capture. Such operators can be applied in situations where it is important to have a customized view of the inking activity. We describe the implementation of a player that allows the user, by selecting the desired operators, to interact with digitally annotated documents while reviewing them.

Fine-grained, structured configuration management for web projects

Researchers in Web engineering have regularly noted that existing Web application development environments provide little support for managing the evolution of Web applications. Key limitations of Web development environments include line-oriented change models that inadequately represent Web document semantics and in ability to model changes to link structure or the set of objects making up the Webapplication. Developers may find it difficult to grasp how theoverall structure of the Web application has changed over time and may respond by using ad hoc solutions that lead to problems of maintain ability, quality and reliability. Web applications are software artifacts, and as such, can benefit from advanced version control and software configuration management (SCM)technologies from software engineering. We have modified an integrated development environment to manage the evolution and maintenance of Web applications. The resulting environment is distinguished by itsfine-grained version control framework, fine-grained Web contentchange management, and product versioning configuration management, in which a Web project can be organized at the logical level and itsstructure and components are versioned in a fine-grained manner aswell. This paper describes the motivation for this environment as well as its user interfaces, features, and implementation.

A document-based approach to the generation of web applications

wVIEW is an automated system for generating Web applications that relies extensively on document representations and transformations. wVIEW adopts the widely accepted hypermedia design principle that content navigation and presentation are separate concerns. Each of these aspects of the design process is controlled by separate declarative specifications. Only the first specification the content structure specification which is described using UML must be provided. However the wVIEW user is free to add extensions and customizations to both the data and navigation models in order to make the final application suit specific needs. This paper describes the wVIEW approach and the current prototype which focuses on the data and navigation modelling aspects. The paper discusses experiences in using XSLT as the primary development tool and shows examples how the enhancements planned to XSLT address some limitations of the application generation process.