About HFI   Certification   Tools   Services   Training   Free Resources   Media Room  
               
 Site MapUser Experience for a Better World   
Human Factors International Home
Free Resources
Bookmark and Share

Ask Eric: Questions & Answers

Each month Dr. Eric Schaffer answers selected questions on usable interface design. Recent Questions
Archived questions and answers about ...

The Usability Profession

October 22 , 2001 – submitted anonymously

Question: What is ergonomics in graphics?

Eric's response: There are a large number of ergonomic principles that apply to the design of graphic elements for Web sites. There are issues of text legibility. You have seen images were there is insufficient contrast to allow users to easily read the labels. There are issues of affordance. You have seen images where it is not obvious if they are clickable or not. There are issues of visual noise. You have seen images that clutter the page and distract the user. These are all ergonomic issues with a body of research guiding our approach.

Beyond the classical usability issues there are a set of psychological effects associated with images. The graphic work does much to create our impression of the site. At HFI we have evolved a whole methodology for systematically testing the graphics against branding objectives. We can make alternative graphic treatments and collect user data to determine which is the most "cluttered," "professional," "warm," and "cute" (or whatever branding terms apply). I consider this psychological component a major aspect of the ergonomics of graphic design.

Top

October 20 , 2001 – submitted by Fern Yates of Johannesburg, South Africa

Question: I have recently entered the world of software testing, as a tester, but because of my legal and journalistic background, I have been asked to write an article on the implications of defective software packages related directly to faulty testing. South African Web sites do not specifically address this problem. What is the legal situation in America and the rest of the world? Or could you direct me to sites that may be able to help me to gather this type of information?

Eric's response: In America there are commonly liability cases around defects in software code. Software is often sold with a warrantee period (which often means the end user gets to do the testing). Torte cases are common and most system integrators have contract language to guard against this.

Perhaps unfortunately, it is quite rare that a suit is brought for poor usability. I have worked as an expert witness in one shareholder suit that cited false claims for the good usability of a online service provider. We have also seen one suit against a system integrator that cited poor usability of the delivered interface. But the legal situation reflects the general practice. For the bulk of companies software quality assurance means that it runs, does not crash, and calculates correctly. It does NOT mean that a human can operate it. Happily, this is starting to change.

Top

May 18, 2001 – submitted by Jeff Tar of Northbrook, IL

Question: In your opinion, what are the differences between an Information Architect and a Usability Architect?

Eric's response: To begin with the obvious, you can find people with all sorts of titles. But what really matters is a person's skills and the knowledge they possess.

Having said that, let me generalize about the "Information Architect" title. Information Architect is a fairly new term. It usually means someone who organizes material, most often for presentation on the Web. Unfortunately, most of the practitioners I have seen have inadequate training. I have usually seen work based on intuition, with limited scientific basis or systematic methodology.

A "Usability Architect" is more commonly called a usability engineer, human factors specialist, software ergonomist, personal subsystem designer, or engineering psychologist. All these titles refer to a profession that goes back to World War II. The profession requires a substantial education. It deals with issues of organization, but also with a far-reaching set of issues concerning task flow design and navigational mechanisms. It uses a massive body of scientific research to address detailed issues of color, layout, operation, and wording. It provides a systematic methodology, incorporating data gathering and usability testing. Practitioners generally have advanced degrees in the field and are often board certified.

Again, I certainly do not care about titles – what counts are skills and performance. But the title "information architect," in my experience so far, has been a warning light. If I seem strong on this issue, it is because I have seen the work of many Information Architects and found it to be a poor reflection on the user-centered design world.

Top

March 23, 2001 – submitted by John Samaan of Cairo, Egypt

Question: What is new in modern ergonomics?

Eric's response: The fundamentals of ergonomics have changed little. Humans were generally single-channel processors when I was in school, and in 1,000 years (when I retire) they will still be that way. The fundamentals about vision, cognition, and motor capabilities are therefore pretty static. So are most of the methods of ergonomics.

But there are some new developments nonetheless.

Some of them are relatively superficial. There are, for example, a lot of new names for things. Interviews and task analysis became contextual inquiry. Human engineering became user-centered design and scenario-based design.

But some developments are more substantial. For one thing, we have a number of good new tools. Eye-tracking technology is better. And we now have the ability to do lots of remote testing, which saves money and works almost as well as being there.

Also, we have many wonderful new puzzles to solve. We have been working in Telematics, where systems are embedded in a vehicle and operated mostly by voice recognition. Handhelds provide another set of new challenges.

Also, as technology goes worldwide it is fascinating to work on issues of globalization and localization. For example, did you know that in India we have trouble with tabs across the top? The culture is very hierarchical, so people tend to feel uncomfortable when the choices are not stacked vertically by importance.

I am also delighted to finally see some willingness to accommodate people with disabilities. We actually have two staff members at HFI who really focus just on the issue of how to make Web pages that can be understood by the visually impaired and blind.

Top

March 1, 2001 – submitted by Raymond LaBrecque of Holyoke, MA

Question: OK, so everybody is going "Web wise" and touting all the latest buzz words and acronyms. Nothing new. The S/W industry has always been about the latest and greatest. After all if you're not on the 'bleeding edge' you get left behind... Fast.

As a S/W engineer focusing in process control and user interfaces, I've been on the practical side of usability testing and U/I design for years. The latest newsletters were real enticing (even though I am doing MMI vs. Web Pages...) But I still stutter & stumble over words/phrases like 'Heuristic Evaluation". Just what does this mean? In plain English please...

Eric's response: Hey, process control is one of my FAVORITE types of project. It is a wonderful and valuable application of usability engineering. There is often so much to offer. I remember working with a glass factory and seeing that oil, water, and air were all controlled by identical-looking valves, resembling the kind you use for a garden hose. For this factory, it was an insight that the valves should be labeled. Before, the staff would just trace the tangled tubing to see where it came from. They thought they should have markings so they could reset the valves to their original position after the equipment was turned down!

Anyway, I'll stop digressing and answer your question. Heuristic evaluation is a review of a design based on a set of rules of thumb. These general principles include things like "Be Consistent" and "Provide Feedback." You are supposed to review the application, asking yourself if it is consistent, if it gives feedback, and so forth. It is not terribly useful because it is so general, but it is better then nothing. The web-oriented principles are not bad for factory automation. But you need to include a set of issues about safety and very frequent usage.

Top

© 1996-2012 Human Factors International, Inc. All rights reserved  |  Privacy Policy  |   rss feed biber hapıbiber hapı