July 2005 Archive

Looking for the best IA advice, tricks, hacks and tools

Update: I’m now accepting submissions until Friday, 29 July

Back in June, I wrote Getting IA Done, Part I an article about IA lessons, advice, tips, tricks, hacks, tools, etc.

I’m now compiling part II of this article, which will contain the best of the same from IAs and UX practitioners the world over. I’ve already received several submissions, but I need many more ideas. Basically, I’m looking for short and meaningful IA advice, productivity tricks, tools and keyboard shortcuts that you find especially useful. To get a better idea of what I’m looking for, read part I.

So if you have any ideas that you would like to be included in part II, please send them to me by Friday, 29 July. A short, 100-150 word description is all I need. If I decide to include your submission, I’ll get back to you with any additional questions, visuals or details that I need. Accepted submissions will be included in part II, which is scheduled to be published in early August. I’m looking forward to your submission.

Brabantia waste bins: More than a guarantee

I recently bought a new pedal bin for our flat and chose the Brabantia 30 litre pedal bin.

At this point some of you might be thinking, “A pedal bin? Just go down to Target and buy one. You even get a bonus bin!” You, my friends, are the reason why most pedal bins break after 1 - 2 years of use.

Brabantia bins are more expensive than your other options (much more so in the US than in the UK), but they have a 10-year guarantee. Yes, a 10-year guarantee. For a pedal bin. According to the seductive copy on their company history page, it’s more than a guarantee, it’s what drives the company.

I bought one the very next day. After buying it I discovered a few other features that made me even happier about my purchase:

Ventilation openings in the inner bucket that prevents a vacuum from forming when removing the bin liner
Non-skid base
Plastic floor ring to keep the bin from scratching the floor

If you’re looking for a new kitchen waste bin, look no further than Brabantia. Unfortunately, they seem to be rather difficult to find in the United States, but a search on Shopping.com revealed a few online venders.

Culture and web design

Preface

This is the fourth installment of my graduate school work: culture and web design. For a background on my graduate work, please read the preface of the first installment, the interaction design process.

Introduction

From 1978 - 1983, the Dutch cultural anthropologist Geert Hofstede conducted a comprehensive study, collecting and analyzing data from over 100,000 individuals from 53 countries. From these results, and later work, he formulated a theory that identifies five primary dimensions to differentiate cultures: power-distance, collectivism vs. individualism, femininity vs. masculinity, uncertainty avoidance and long vs. short-term orientation. In this book, Cultures and Organizations, Software of the Mind, he rated each country on these dimensions, assigning normalized values from 0 - 100.

These dimensions can be applied to many aspects of cultures, one of which is information technology. Marcus et al. showed this by explaining the implications of Hofstede’s cultural dimensions for user-interface and Web design. Using a basis of this work, I analyzed four Honda websites: Honda Sweden, Honda USA, Honda Japan and Honda India.

Website Comparison

Marcus correlates a low long-term orientation rating with a desire for immediate results and achievement of goals. Both USA and Sweden have relatively low long-term orientation ratings of 33 and 29, respectively. This can be observed on both websites, but in different ways. Honda USA has a number of tools such as “Build a Honda” and “Request a Quote” that allow visitors to quickly complete common tasks. Honda Sweden does not have these types of tools, but its navigation is simple and straightforward, allowing visitors to find desired information quickly. Conversely, Japan has a very high long-term orientation rating of 80, which is correlated with patience in achieving results and goals. Even though I was not able to understand the Japanese text on the Honda Japan website, it seems that visitors might have trouble completely their goals quickly with so many available options.

The interface design aspects of power-distance are similar to those of long-term orientation. Both Sweden and USA have relatively low power-distance ratings of 33 and 40 respectively, which Marcus correlated with customer prominence. Consequently, both the Honda Sweden and Honda USA sites appear customer focused, presenting visitors with navigation options that allow visitors to quickly achieve their goals. Conversely, India has relatively high power-distance rating of 77, which Marcus correlated with leader prominence. However, this is not evident in the Honda India website, as it appears just as customer focused as the Swedish and American websites.

India was not the only exception to Marcus’s interpretations. Further analysis identified several additional misfits. For example, Japan has very high uncertainty avoidance rating of 92, which is correlated with limited choices and navigation schemes intended to prevent users from becoming lost. However, neither of these features were identified on the Honda Japan website. There are many choices available on the homepage and the navigation is very inconsistent throughout the site. Sweden has a low uncertainty avoidance rating of 29, which is correlated with maximal choices and low control of navigation. Like Honda Japan, these features are not found on the Honda Sweden website. Instead, the Honda Sweden website contains very few choices and a very high control of navigation.

Discussion

As shown above, the four Honda websites displayed cultural values in several aspects, but the application of Marcus’s work was often ambiguous and in several cases, inapplicable. This could be due to several reasons. Honda is an international brand and they may place certain design constraints on their websites and other marketing material in order to protect their brand. Chau et al. found that firms such as Amazon and Autobytel have expanded globally by providing a standardized transaction-oriented interface. Most visitors will not make transactions on Honda websites, but the same idea may be true for the websites as a whole. Just as Honda may desire to standardize their brand, they may also want to standardize other aspects of their websites.

Other misfits between Marcus’s work and Honda’s websites could simply be due to the fact that the Web is a global medium: most countries’ websites can be accessed from almost anywhere in the world. Because of this ease of access, cultures may strive to compete with other cultures and create designs similar in look and feel. In addition, cultures may be easily influenced by other cultures causing design ideas to bleed together.

Regardless of whether Hofstede’s cultural dimensions can be translated to interface and Web design, it has been shown through several studies that there are significant benefits to creating culturally specific websites. Simon found that a single global website that appeals to all cultures does not appear feasible because of differences between cultural perception and satisfaction with websites. Chau et al. found that in addition to different cultures using the Web for different purposes, these different purposes might lead them to have different impressions of the same Web sites. Therefore, I believe that in order maximize the effectiveness of websites, web designers should always consider the cultural aspects of the site they’re creating.

Honda Sweden Homepage

Honda Sweden. ↑

Honda USA Homepage

Honda USA. ↑

Honda Japan Homepage

Honda Japan. ↑

Honda India Homepage

Honda India. ↑

All installments in this series:

Designing a cashpoint

Preface

This is the third installment of my graduate school work: designing a cashpoint. For a background on my graduate work, please read the preface of the first installment, the interaction design process.

Background

I have just been hired as an ergonomics consultant to an industrial designer on the design of a new customer service machine. The machine will be located on the outside wall of a small independent building society, dispense cash and provide balance information. Customers are to be provided with plastic cash cards and a personal identification number. Additionally, customer identity will be double checked through use of a palm-print reader.

The following are the stages that I would go through to arrive at recommendations for the physical design of the machine including: further data I might need, how I might find this data, the output of each stage and how it contributes to the design. I will also describe how I would present my recommendations to the industrial designer.

Stage 1: Identifying the Users

In this stage, I will determine the key features of the new machine’s target population. This is done through a number of steps. First, if the building society has a similar machine currently in use, I will ask the society to identify customers who use this machine. Additional target users can be identified through interviews and questionnaires conducted with current customers. Next, I will gather anthropometric data from as many target users as possible in order to determine the range of users and limiting users. In a perfect world, all target customers would be surveyed. However, the chances of this happening are extremely slim. A practical alternative is to survey a large sample of target costumers focusing on the limiting users. Stature and standing elbow height will be surveyed. Additional data, such as sitting reach, will be measured for wheelchair users. Any other special needs identified in the target population will be noted. Finally, all of this information will be analyzed to determine the key features of the target population, including the range of users that needs to be accommodated.

The output of this stage will be a report that highlights these key features. This report will contribute to the design by helping the designer understand the new machine’s user population. Further into the workspace design procedure, it will also aid in stage 4, when I will be identifying limiting user requirements.

Stage 2: Identifying the Tasks

In this stage, I will identify the tasks that happening during the interaction with the new machine. This is done though a series of task analysis methods. Before the analysis can be started, it is important to consider which task analysis methods are most appropriate for this particular design. The industrial designer is primarily asking for input regarding the user interface. Two analysis methods that are particularly good at informing the user interface are hierarchical task analysis and link analysis. Data for these analyses can be acquired though a variety of methods, the most notable of which are walkthroughs, interviews and discussions.

The output of this stage will be the raw data gleaned from the hierarchical task analysis and link analysis. This data will not contribute directly do the design but will contribute indirectly through the task description which is described in the next stage.

Stage 3: Creating a Task Description

After I have the raw data from stage 2, I will create a task description, which will partially take the form of a hierarchical task diagram and link analysis diagram. Another task description diagram appropriate at this stage is an operational sequence diagram, specifically a partitioned operational sequence diagram, which is useful for showing a sequence of events, simple decision structures and flow of information between different dimensions of the task (between the user and the machine). After each diagram is constructed the tasks will be further broken in order to determine if the task is manual or cognitive. By categorizing each task as manual or cognitive, I can describe its user requirements. For example, deciding what personal identification number to enter is a cognitive task that requires previously learned information. A more detailed description of this task will be useful to the designer as she creates the user interface.

The output of this stage will be a hierarchical task, link analysis and partitioned operational sequence diagram. Each of these diagrams, along with the associated task description, will allow the designer to make more informed decisions about each interface element’s features and placement of interface elements in relation to each other and the user.

Stage 4: Identifying Limiting User Requirements

The next stage is to identify limiting user requirements for each task. Some of the most important anthropometric dimensions to consider are stature, standing elbow height and sitting reach (for wheelchair users). Bodyspace, supplemented by the data from stage 1, will provide a good starting point for many of these dimensions. In addition to anthropometry, it is also necessary to consider the deaf, blind and users that may not speak or read the local language. It will be helpful to create a mockup containing the interface elements, positioned appropriately according to the task description and limiting user requirements. If limiting users are available from the target population, a simulation or fitting trial can be carried out, which will help to further refine interface element features and placement.

The output of this stage will be a report describing features and recommendations on the placement of these features in the interface. This report, like the diagrams described in the previous stage, will contribute to the design by allowing the designer to make more informed decisions about interface element features and placement.

Stage 5: Identifying Ergonomic Concerns

In this stage, each feature from in the previous stage will be considered with reference to the Concentric Rings model, which includes the user, workstation, workspace envelope, workplace environment and external environment. By examining each feature of the machine within different contexts of the workspace, a number of concerns can be identified. Many of these concerns will be identified through a detailed analysis of the machine features with respect to the workspace. Further concerns can be identified through questionnaires and interviews with target users. For example, users may be asked in user trials if the proposed design meets their specific needs.

The output of this stage is report highlighting these ergonomic concerns. This report will help the designer to think through each feature in the larger scope of the machine.

Stage 6: Identifying Priorities and Clashes

There will be certain features that absolutely must be ergonomically correct, such as the cash dispenser. It is important to prioritize these features so that they do not get lost in the jumble of other ergonomic concerns and decisions. It is also important to identify potential clashes that may occur between design features. In order to create a comprehensive list of the prioritized features, I will carry out discussions and interviews with stakeholders and user. After features have been prioritized, clashes will be determined through a feature matrix, where all features can be listed in order of priority (from highest to lowest) along each axis. By using this method, clashes that occur towards the lower end of each axis will be noted as more serious problems. Conversely, clashes that occur towards the upper end of each axis will be noted as less serious problems but will still be considered.

The output of this stage is a report highlighting prioritized features and clashes between features. This report will contribute to the design by helping the designer understand which features must be included and which features have potential conflicts with other features.

Stage 7: Identifying Unoptimized Features

Next, any features that are not optimized will be identified. In order to successfully identify these features, it is important that a detailed mockup/simulation be created. If this is not possible, features can be isolated and tested independently. However, there are definite limits to what features can be studied in isolation. For example, because feedback affects a user’s ability to input information it is important that the keypad is tested in context with any display that the machine might use. When testing features it is also helpful to set benchmarks to test against. Some measurements to test may include success/failure within a set time, time on task and number of errors before completion. To decide a starting point, it may be appropriate to perform usability benchmarks on a similar machine.

After unoptimized features have been identified, they will be included in a report, highlighting the prioritized features. This report will inform the designer on the features that should be considered in future ergonomics work. It will also provide a basis for stage 8, which formally identifies the aspects of the design which call for further investigation.

Stage 8: Suggesting Further Investigation

It is rarely possible to address every feature and aspect to the proper extent during a design. For this reason, any features that I have not been able to address will be identified and listed. Also included on this list will be aspects of the workstation that call for further investigation by an ergonomics professional.

The output of this stage is a report outlining all features and aspects that I have not been able to address along with a discussion of potential solutions and further resources. This report will give the designer a starting point if more ergonomic input is required.

Presenting Ergonomic Recommendations

After all of the aforementioned reports have been completed, the key findings from each stage will be summarized using charts and graphs where appropriate, followed by concise ergonomic recommendations for each feature. Because these recommendations are for use by an industrial designer, recommendations will be supplemented by computer aided design (CAD) wherever possible. The summary of key findings and ergonomic recommendations will presented to the designer as a report along with a CD-ROM containing the associated CAD illustrations.

All installments in this series:

London Explosions

I’m in Horsham, England at a client site, and heard the horrible news several hours ago. There have been ~~six~~ ~~seven~~ three explosions on the Underground and one bus explosion.

Londonist and Guardian Newsblog have good, up to the minute coverage. You can also follow the story on BBC News, CNN, Google News and Flickr.

← June 2005 | Archives | August 2005 →