CS5340 - Computer-Human Interaction Coursework, Fall 2009

William Furr - furrwf@ccs.neu.edu

Back to Index

Table of Contents

Homework 1 - UI Critique

PDF

Good UIs

Android Home Screen - Cell Phone OS

Purpose:

This is the primary interface of an Android smartphone. It allows access to applications, notifications from applications, and status information.

Features that make it good:

Remember the Milk - To-Do List Webapp

Purpose:

This is an online application designed to help the user create and manage to-do lists and items.

Features that make it good:

Bad UIs

Datatel School ERP System UI

Purpose:

This is the interface for using the Datatel school ERP system, which includes software support for finances, human resources, student information, billing, basic customer relations management, fundraising, and more.

Features that make it bad:

Why is it so bad?

The graphical interface is translated from screen specifications written for a character-based 80x23 terminal application. The legacy holdovers and vast codebase keep the company from updating their software in a timely fashion. The company serves a fairly specialized niche, educational ERP systems. The cost of switching from one system such as Datatel to one of their competitors is very high. Many of the drawbacks of the interface can be ameliorated with training and procedures.

What could make it better?

The screen specifications need to be translated to a new engine that allows for native Windows forms or web forms to be generated with a much wider variety of WIMP interface widgets. The software also should be brought into line with other Windows desktop software in terms of how it reacts to keyboard and mouse input, as well as window management commands. Alternately, the interface could be made web-based, and made to interact in standard and predictable ways with a web browser.

Cyclo-Computer for Bicycle

Purpose:

Features that make it bad:

Why is it so bad?

In the three-constraint model, between price, features, and time; the company that manufactured this item definitely prioritized cheap and fast over full-featured. The one saving grace of this device is that it's cheap and non-essential (unlike the brakes and steering interface of the bicycle). The engineer who decided how many buttons to use and what features to map to them may have been a student of the same engineers who design cheap digital watches, with their inscrutable buttons.

What could make it better?

Clearly label the buttons and have one task per button. Have a confirmation message for destructive operations like resetting the device.

Paper Review 1

Paper Name: Acceptance and Usability of a Relational Agent Interface by Urban Older Adults - CHI'05

Presentation:

Sources:

Homework 2 - Project Ideas

Simple Wikipedia Touchscreen Browser

This would be a web browser designed for use on a touchscreen by older adults. It would be limited to certain sources of information, such as Simple English Wikipedia, or other such suitable sources containing content easy for non-native English speakers and low literacy persons. The content could be further simplified by providing an external navigation structure or ordering of the content. Readability and usability would be enhanced by using custom stylesheets with high contrast visibility, use of a colorblind-safe palette, removal of extraneous page elements, reflow of relevant inline images, and so on. A text-to-speech facility could also be provided for use by vision-impaired persons.

Touchscreen Self-Test Games

This program would provide a battery of informative self-tests for older adults interested in how their cognitive, visual, hearing, and motor skills compare with other people. This was inspired by reading about colorblindness tests (e.g. - the Ishihara color test) that could be adapted for dynamic generation and touchscreen display. Other tests could include standard tests for short-term memory, hand-eye coordination, hearing range and depth perception. The program could incorporate statistics for "typical" humans of various types or ages, and it could track an anonymous record of prior test takers for comparison (e.g. - a top 10 short term memory takers).

Touchscreen Guided Hospital Entrance Form

My least favorite part of visiting the doctor or hospital is filling out the mass of forms in the waiting room. We could potentially provide a guided method of form-filling using the touchscreen. It could either generate the form through a guided question-and-answer session and then print the final form, or do a guided tour of the physical paper form using example screenshot images. A possible extension would be to interface with the hospital information system and provide the entered information directly to the nurses, but this is likely outside the scope of a semester-long project, and there are security concerns with interfacing with hospital systems, as well as long-term maintenance concerns for keeping up with IS changes. However, even if the guided system is limited to the current version of the paper form, it could materially enhance the quality of initial data collection and ease frustration among patients entering the hospital.

Homework 3 - Ethnography Observation

I observed patients at the Boston Medical Center Geriatrics Clinic for 40 minutes and interviewed three subjects. The results are documented here: BMC Observation Writeup (PDF).

Homework 8 - Heuristic Evaluations

Heuristic Evaluation 1 for Team 4 - Digital Wall Calendar

Heuristic Evaluation 2 for Team 8 - Fit-Fit


Last Updated: February 11, 2010

Back to Index