📘 Lecture 31 — Evaluation – Part VII
📖 Overview: This lecture provides a comprehensive definition and framework for conducting usability testing, distinguishing it from other evaluation methods like focus groups, surveys, and beta testing. It emphasizes that usability testing is an iterative process involving real users performing real tasks, with the primary goal of improving product usability rather than merely verifying functionality.
🗂️ Topics Covered
The lecture defines usability testing through five essential characteristics, explores what is and isn't required for testing, discusses when testing is appropriate throughout the design process. It covers testing applications across various product types, interfaces, and documentation aspects. The lecture also introduces special techniques like co-discovery and active intervention, and compares usability testing to beta testing, highlighting additional benefits for organizational change.
📝 Lecture Summary
What is Usability Testing?
Every usability test shares five fundamental characteristics. The primary goal is to improve the usability of a product, distinguishing it from research studies that investigate phenomena or quality assurance tests that check specifications. Within this general goal, specific concerns vary between tests—from novice versus experienced user performance to installation ease for technical staff.
The participants must represent real users who will actually use the product. Testing with programmers for a product intended for legal secretaries is not a usability test. If participants are more experienced than real users, marketplace problems may be missed; if less experienced, unnecessary changes may be made. Participants must perform real tasks that reflect actual job or home usage, which requires understanding users' work contexts.
🔑 Definition — Usability Test: An evaluation method where real users perform real tasks with a product while being observed and recorded, with the data analyzed to diagnose problems and recommend improvements.
You must observe and record what participants do and say, typically one participant at a time. This distinguishes usability testing from focus groups (opinions only), surveys (self-report only), and beta testing (unsystematic field observation). After data collection, you must analyze the data, diagnose real problems, and recommend changes to fix those problems.
📌 Example: A company testing a product for internal network deployment excluded the printed manual to simulate real conditions. Participants were told online help was available and given a help desk number. This focused the test on where people got stuck and how useful online help was, yielding actionable data for improving the interface and help system.
What Is Not Required for a Usability Test?
Features that are useful but not necessary include a laboratory with a one-way mirror, data-logging software, videotape, or a formal test report. A memorandum or meeting about results may be sufficient. Each of these has advantages, but none is an absolute requirement, allowing for low-budget testing with limited staff and equipment.
When is a Usability Test Appropriate?
Usability testing is not limited to a single summative test at project end. The five defining characteristics apply regardless of where you are in the design and development process. Testing is appropriate iteratively from predesign (testing similar products), through early design (testing prototypes), and throughout development (testing different aspects and retesting changes).
Questions that Remain in Defining Usability Testing
Some fuzzy edges remain. A test with only one participant probably would not be called a usability test—at least two or three people per subgroup are needed to avoid idiosyncratic behavior. A test without quantitative measures might also not qualify—at minimum, basic measures like number of participants with problems or task completion time are needed to substantiate reported issues.
💡 Why this matters: The evolving definition of usability testing means practitioners must make thoughtful decisions about participant numbers and measurement approaches based on their specific testing goals and constraints.
Testing Applies to All Types of Products
Usability testing works for all types of products—consumer electronics (TVs, VCRs, telephones), medical products (bedside terminals, patient monitors), engineering devices (oscilloscopes), application software, voice response systems, and automobile navigation systems. The procedures vary somewhat depending on the product but the approach remains largely consistent. Products need not involve hardware or software—forms, instructions, interviewing techniques, and questionnaires can also be tested.
Testing All Types of Interfaces
Any product people have to use has a user interface. Even an elevator's control panel can be tested by creating a simulation on a touch screen computer programmed to respond to button presses with sounds and visual feedback. Users can then be brought in one at a time with realistic scenarios.
📌 Example: An elevator interface prototype on a touch screen could simulate alarm sounds and door opening/closing. Users would interact as they would with a real elevator panel, allowing developers to identify problems before building the actual elevator—when changes would be prohibitively expensive.
Testing All Parts of the Product
Depending on development stage and concerns, usability tests can focus on specific parts including installing hardware, operating hardware, cleaning and maintaining hardware, understanding messages, installing software, navigating menus, filling out fields, recovering from errors, learning from tutorials, or finding instructions in user guides or online help.
Testing Different Aspects of the Documentation
When including documentation in a test, you must decide whether your concern is whether users go to the documentation or how well it works for them when they do. These questions are difficult to answer simultaneously. To test tutorial usability, ask participants to use it explicitly. To test whether anyone would actually open the tutorial, set up tasks with the tutorial available but not required.
📌 Example: In a test where the concern was minimizing help desk calls, the manual was excluded. Participants were told about online help and given a help desk number to call only if truly stuck. This test provided information on where people got stuck, how helpful online help was, and when people called the help desk—but gave no information to improve the printed manual.
Testing with Different Techniques
Most usability tests involve one participant at a time working alone while observed from a corner or behind a one-way mirror. Two useful alternative techniques are co-discovery and active intervention.
Co-discovery involves two participants working together to perform tasks. They are encouraged to talk to each other, which is more natural than thinking out loud alone. This yields more information about thinking processes and problem-solving strategies. Research by Hackman and Biers (1992) confirmed that co-discovery participants make useful comments without distorting other results. Co-discovery is more expensive (paying two people per session) and may be harder to observe, but is especially useful early in design.
Active intervention involves a test team member sitting with the participant and actively probing their understanding during tasks (e.g., asking them to explain what they would do next and why). This provides insights into participants' evolving mental model of the product. Active intervention is particularly useful early in design with prototypes, but should not be used when the primary concern is measuring task completion time or help desk calls.
🔑 Definition — Co-discovery: A usability testing technique where two participants work together to perform tasks, talking naturally to each other as they solve problems.
🔑 Definition — Active intervention: A usability testing technique where a team member sits with the participant and asks probing questions throughout the session to understand the participant's mental model.
Additional Benefits of Usability Testing
Usability testing has two specific benefits beyond improving the product. It helps change people's attitudes about users—watching users struggle has much greater impact than discussions about usability importance. Designers, developers, and managers often become "instant converts" after witnessing testing. It also helps change the design and development process by revealing deeper global problems with both the product and the organizational process.
Comparing Usability Testing to Beta Testing
Beta testing (field testing, clinical trials, user acceptance testing) involves giving an early-release product to customers and waiting for feedback. While real users do real tasks in real environments, beta testing seldom yields useful usability information for several specific reasons:
- The beta test site doesn't have to use the product
- Feedback is unsystematic—users report what they remember
- No one observes and records user behavior
- Tasks are not chosen by testers—critical situations may not arise
- Beta testing comes too late in development to fix usability problems
💡 Why this matters: Beta testing's major disadvantage is its timing—it occurs very close to release, when only critical functional bugs can be addressed. Usability testing can be done throughout the design process, allowing problems to be fixed and retested early when changes are less expensive.
📌 Example: Companies that include observations and interviews with beta testing, or give tape recorders to selected users to record problems as they occur, may get more useful information. However, these techniques still cannot overcome beta testing's fundamental timing disadvantage.
⭐ Key Takeaways
The five essential characteristics of usability testing—improving product usability, using real users, having them do real tasks, observing and recording behavior, and analyzing data to diagnose problems—distinguish it from research studies, quality assurance tests, focus groups, surveys, and beta testing. Usability testing can be conducted iteratively throughout the design process using prototypes, not just at the end of development. Special techniques like co-discovery and active intervention are particularly valuable early in design for gaining insights into user thinking and mental models. While tools like laboratories and videotape are useful, they are not required—effective testing can be done on a shoestring budget. Beta testing is not a substitute for usability testing because it is too late, too unsystematic, and provides no observation of actual user behavior.
🧠 Quick Revision Questions
- What are the five essential characteristics that define every usability test?
- How does a usability test differ from both a research study and a quality assurance function test?
- What is co-discovery testing, and when is it particularly useful?
- Why is beta testing generally inadequate as a primary usability evaluation method?
- What factors determine whether a test qualifies as a usability test (e.g., number of participants, use of quantitative measures)?
📘 Lecture 32 — Evaluation IV
📖 Overview: This lecture explores the critical importance of navigation in both physical spaces and Web sites. It compares the process of finding a product in a department store to navigating a Web site, emphasizing that clear, consistent navigation is essential for user satisfaction and retention.
🗂️ Topics Covered
The lecture begins by comparing navigation in physical spaces (a mall) and Web sites, highlighting similarities and differences. It then discusses the "unbearable lightness of browsing" and the unique challenges of Web space, including lack of scale, direction, and location. The overlooked purposes of navigation are examined, followed by Web navigation conventions and the concept of persistent navigation. Finally, it covers specific navigation elements including Site ID, Sections, Utilities, Low Level Navigation, Page names, "You are here" indicators, and Breadcrumbs.
📝 Lecture Summary
Scene from a mall
The lecture opens with a scenario of searching for a chainsaw in a Sears department store. You use the store's navigation systems—department names on walls and signs at the end of aisles—along with your ability to scan shelves to find what you're looking for. The process involves deciding whether to search on your own or ask for help, a decision based on familiarity with the store, trust in their organization, time pressure, and sociability.
🔑 Definition — Navigation systems: The signs and organizing hierarchy that help users find what they are looking for in a physical or digital space.
📐 Formula: Process of finding items = Look at department signs → Navigate to section → Scan aisle signs → Examine individual products → If wrong, try another aisle or start over
Web Navigation
The process of entering a Web site mirrors the physical mall experience in several ways:
- You're usually trying to find something
- You decide whether to ask first (search) or browse first
- If you choose to browse, you make your way through a hierarchy using signs
- Eventually, if you can't find what you're looking for, you'll leave
The Web equivalent of asking directions is searching—typing a description into a search box. Some people are "search-dominant" users who almost always look for a search box, while "link-dominant" users browse first and search only when necessary.
💡 Why this matters: Understanding user preferences (search vs. browse) helps designers create navigation that supports both types of users.
The unbearable lightness of browsing
While Web browsing feels similar to navigating physical spaces, the Web experience is missing many cues we rely on:
- No sense of scale: Users have little idea how large a site is
- No sense of direction: No left and right, no up and down in physical terms
- No sense of location: No physical traversal of space; you're teleported via links
The Back button accounts for 30-40 percent of all Web clicks because users lack physical sense of location. Home pages serve as fixed reference points, like the North Star.
💡 Why this matters: The lack of physicality in Web navigation makes it essential to design explicit navigation systems that compensate for this missing sense of place.
The overlooked purposes of navigation
Navigation serves several important functions beyond helping users find things and telling them where they are:
- It gives us something to hold on to — Done right, navigation makes users feel grounded
- It tells us what's here — Navigation reveals content by making the hierarchy visible
- It tells us how to use the site — Good navigation implicitly tells users where to begin and what options are available
- It gives us confidence in the people who built it — Clear navigation creates a good impression and influences whether users return
Web navigation conventions
Physical spaces have navigation conventions that have evolved over time (street signs, page numbers, chapter titles). Web navigation conventions have emerged quickly, mostly adapted from print conventions. The basic elements include Site ID, Sections, Utilities, and Pages.
Persistent navigation (or global navigation) describes the set of navigation elements that appear on every page of a site. It should include five elements users most need at all times.
Exceptions to the persistent navigation rule:
- The Home page — has different burdens and sometimes shouldn't use persistent navigation
- Forms — on payment or registration pages, persistent navigation can be a distraction; use a minimal version with Site ID, Home link, and helpful Utilities
Site ID
The Site ID or logo is like the building name for a Web site. Users expect to see it at the top of the page, usually in or near the upper left corner. The Site ID represents the whole site and should:
- Be at the top to frame the entire page
- Look like a Site ID with a distinctive typeface and recognizable graphic
The Sections
The Sections (primary navigation) are links to the main sections of the site—the top level of the site's hierarchy. The persistent navigation should also display secondary navigation: the list of subsections in the current section.
The Utilities
Utilities are links to important elements that aren't part of the content hierarchy. These include Help, Site Map, Shopping Cart, About Us, and Contact Us. The Utilities list should be slightly less prominent than Sections. Only four or five Utilities should be in the persistent navigation.
🔑 Definition — Utilities: Links to important elements of the site that help users use the site or provide information about its publisher.
Low Level Navigation
A common problem in Web design is failing to give lower-level navigation the same attention as the top. This happens because:
- Good multi-level navigation is hard with limited space
- Designers often don't have time to figure out beyond two levels
- Designers assume users will understand by the time they get deeper
The reality is that users spend as much time on lower-level pages as at the top. It's vital to have sample pages showing navigation for all potential levels before arguing about the Home page color scheme.
Page names
Page names are the street signs of the Web. Four things to know:
- Every page needs a name — Don't rely solely on navigation highlighting
- The name needs to be in the right place — Should frame the unique content
- The name needs to be prominent — Usually the largest text on the page
- The name needs to match what I clicked — Sites make an implicit social contract: if you click "Hot mashed potatoes," you should get a page named "Hot mashed potatoes"
"You are here"
A "You are here" indicator shows users where they are in the scheme of things. On the Web, this is accomplished by highlighting the current location in navigational bars, lists, or menus. The most common failing is that indicators are too subtle; they need to stand out, ideally using more than one visual distinction (e.g., different color and bold text).
Breadcrumbs
Breadcrumbs show users where they are by displaying the path from the Home page to the current location. Unlike "You are here" indicators that show position in context of the hierarchy, Breadcrumbs only show the path taken. They're reminiscent of the trail of crumbs Hansel dropped in the woods.
🔑 Definition — Breadcrumbs: Navigation elements that show the path from the Home page to the current page, helping users understand where they are and how they got there.
⭐ Key Takeaways
Navigation is critical for Web sites because users lack the physical cues they rely on in real-world spaces (sense of scale, direction, and location). Persistent navigation should include Site ID, Sections, and Utilities, appearing consistently on every page except Home pages and forms. Every page needs a clear, prominent name that matches what users clicked, and "You are here" indicators must be sufficiently obvious to be useful. Lower-level navigation deserves as much attention as top-level navigation since users spend significant time there.
🧠 Quick Revision Questions
- What are the three "senses" that Web users lack compared to navigating physical spaces?
- What are the five elements that persistent navigation should include?
- What are the two exceptions to using persistent navigation on every page?
- Why is it important that page names match what users click on?
- What is the difference between a "You are here" indicator and Breadcrumbs?
📘 Lecture 33 — Evaluation V
📖 Overview: This lecture focuses on evaluating web navigation through two specific lenses: the effective use of tabs as a navigation device and the "trunk test" for assessing how well a page communicates its context. It explains why tabs work so well, using Amazon as a case study, and provides a practical, quick method for evaluating whether a web page answers the essential questions a lost user would have.
🗂️ Topics Covered
The lecture begins by outlining four key reasons to use tabs for site navigation, highlighting their self-evident nature and visual distinctiveness. It then provides a detailed analysis of Amazon's implementation of tabs, focusing on four critical attributes: correct drawing, fast loading, color coding, and having a tab selected on entry. Finally, it introduces the "trunk test," a quick method where a user is dropped onto a random page and must be able to instantly identify the site, page name, sections, local navigation, location, and search function.
📝 Lecture Summary
Four reasons to use tabs
Tabs are a rare example of a physical metaphor that works well in a user interface. They divide content into sections and make it easy to navigate by clicking on the tab. The lecture gives four reasons why tabs are an excellent navigation choice for large sites. First, they are self-evident; even the most "computer illiterate" user understands their function. Second, they are hard to miss, as they are visually distinctive and unlikely to be overlooked like a button bar. Third, they are slick, adding polish and visual interest without significantly increasing page load times if implemented correctly. Fourth, they suggest a physical space, creating the illusion that the active tab is in front of the others, which provides a strong visual cue about the current section.
💡 Why this matters: Effective navigation is critical for user experience. Tabs leverage a familiar real-world metaphor to make large, complex websites feel more manageable and intuitive.
Why Amazon is good?
Amazon is presented as the gold standard for tab implementation. The lecture identifies four key attributes of Amazon's tabs to imitate. First, they’re drawn correctly, meaning the active tab must visually appear "in front of" the others by using a different color or contrasting shade and physically connecting with the content below. Second, they load fast, using only two graphics per page, totaling less than 6k, which avoids the longer initial load times and "crazy-quilt" effect caused by using separate graphics for each tab. Third, they’re color coded, where each section uses a distinct color that is also used in other navigational elements, although the lecture warns that color should never be the only cue due to color-blindness and a general lack of awareness of color coding in a significant percentage of users. Fourth, there's a tab selected when you enter the site, which ensures the impact of the tabs is felt from the very first second, though this can create a dilemma when a site's identity evolves (e.g., Amazon creating a "Welcome" tab to move beyond being just a bookstore).
🔑 Definition — Color Coding: The use of different colors for different sections of a website to help users visually distinguish between them. It is an additional cue and should not be the only differentiator.
📐 Formula: Effective Tabs = Correct Drawing (active tab in front) + Fast Loading (minimal graphics) + Color Coding (as a secondary cue) + Active Tab on Entry
📌 Example: Amazon uses a single row of tabs. Each tab (e.g., Books, Electronics) has a distinct color. The active tab is drawn in a contrasting shade and connects to the content below. When a user lands on the Amazon home page, the "Welcome" tab is selected.
Here's how you perform the trunk test:
The trunk test is a quick acid test for good web navigation. The scenario is that you are blindfolded, locked in a trunk, driven around, and dumped on a random page deep in a website. The test is whether you can instantly answer six questions: (1) What site is this? (Site ID), (2) What page am I on? (Page name), (3) What are the major sections? (Sections), (4) What are my options at this level? (Local navigation), (5) Where am I? ("You are here" indicators), and (6) How can I search? The procedure is to print a random page, squint to see only the overall appearance, and quickly try to find and circle each of these six items. The test emphasizes that users often enter a site from a search engine link, not the home page, so the page must be self-explanatory at a glance.
🔑 Definition — Trunk Test: A usability evaluation method where you test if a web page instantly and clearly communicates the site identity, the user's current page, the major sections, local navigation options, the user's location within the site, and search functionality.
💡 Why this matters: The trunk test addresses the reality of web use, where users are often "dumped" onto a page without context. A well-designed page should not require close scrutiny to be understood.
33.1 Try the trunk test
The lecture provides a visual example of the trunk test using a sample website. It demonstrates how to circle each element on the page. The Site ID is typically a logo or brand name at the top-left. The Page Name is usually the title of the current page. Sections and subsections are the main navigation categories, such as tabs. Local Navigation refers to links that are specific to the current section. A "You are here" indicator shows the user's current location, often by highlighting the active tab or using a breadcrumb trail. Finally, the Search function is a utility that should be easy to find.
📌 Example: In the sample website provided, the Site ID is the company logo. The Page Name is "Home". The Sections are represented by tabs (e.g., Home, About, Services). The Local Navigation would be links within the "Home" section. The "You are here" indicator is shown by the "Home" tab having the same background color as the page content. The Search utility is a text box located at the top right of the page.
⭐ Key Takeaways
Tabs are an exceptionally effective navigation tool because they are self-evident, visually distinctive, and create a strong sense of physical space. A successful tab implementation, as exemplified by Amazon, requires the active tab to visually appear in front, fast loading via minimal graphics, and the use of a tab selected upon entering the site. Color coding of tabs is helpful but must be supplemented with other cues to account for color-blindness and users who do not notice color differences. The trunk test is a crucial evaluation method that simulates the common user experience of arriving on a deep page from a search engine, demanding that a page's core navigation elements be instantly identifiable at a glance without close scrutiny.
🧠 Quick Revision Questions
- What are the four specific reasons given in the lecture for why tabs are an excellent navigation choice for large websites?
- Describe the visual illusion that must be created for tabs to work effectively. What are the two key graphic requirements for the active tab to create this illusion?
- According to the lecture, what is the main problem with using a separate graphic for each tab or button on a navigation bar?
- What is the "trunk test" and what are the six questions it is designed to answer about any given web page?
- Why does the lecture recommend squinting or holding a printed page at arm's length when performing the trunk test?
📘 Lecture 34 — Evaluation – Part VI
📖 Overview: This lecture focuses on performing heuristic evaluations to identify usability issues in a web interface. Through a practical example, it demonstrates how to systematically detect common design problems by applying usability heuristics, covering issues related to readability, navigation, content clarity, and user expectations.
🗂️ Topics Covered
The lecture walks through a heuristic evaluation of a website, identifying specific usability problems such as unclear fonts, dim colors, misleading browser titles, excessive banner ads, improper use of highlighted tabs, confusing breadcrumbs, deceptive “sign up” links, and horizontal scrolling. Each issue is tied to a usability heuristic and explained with visual annotations from the site.
📝 Lecture Summary
Identifying Usability Issues Using Heuristics
The lecture presents a practical heuristic evaluation case study. A website is examined for common usability violations. Each issue is identified with a screenshot annotation, linking the visual problem to a specific heuristic principle. The examples cover readability, navigation, content labeling, and page layout.
Clear and readable fonts not being used
The site uses fonts that are not clear or readable. This violates the heuristic of consistency and standards or aesthetic and minimalist design. Dim color further reduces readability, making text hard to see.
🔑 Definition — Heuristic Evaluation: A usability inspection method where experts review an interface against a set of usability principles (heuristics) to identify potential problems.
Browser Title always contains the word ‘Home’
Every browser title on the site includes the word “Home,” which is misleading. Different pages should have unique, descriptive titles. This confuses users and violates the heuristic of user control and freedom or match between system and the real world.
Banner ads take up too much space
Large banner ads dominate the page layout, pushing content aside. This violates aesthetic and minimalist design by cluttering the interface and distracting users from primary tasks.
Invalid browser title characters
The browser title contains invalid or special characters that may not display correctly in all browsers, reducing usability and accessibility.
Use of highlighted tabs in global navigation bar shows this is the ‘Home’ page
In one screenshot, a highlighted tab correctly indicates the current section (Home). However, absence of a highlighted tab in another view confuses the user about which section is being viewed. This violates the heuristic visibility of system status, as users need constant feedback about their location.
📌 Example: On one page, the “Home” tab is visually highlighted, clearly showing the user’s location. On another page, no tab is highlighted, leaving the user uncertain which section they are in.
Version numbers should not be given on the main page
Displaying version numbers (e.g., “v2.1”) on the main page is irrelevant to most users and adds unnecessary clutter. This violates aesthetic and minimalist design — only information relevant to the user’s task should be visible.
Breadcrumbs format do not follow standard conventions
The site’s breadcrumb trail uses an unusual format, confusing users who expect standard separators (e.g., “>” or “/”). This violates consistency and standards, as users rely on learned conventions to navigate.
‘Sign up now’ link appears to refer to free report...
The text “Sign up now” suggests a free offer (e.g., a free report), but the link actually takes the user to the online store. This violates consistency and standards and user control and freedom — the link label does not match its destination, misleading the user.
💡 Why this matters: Misleading links break the trust between the user and the system. If a user expects one action and gets another, they may feel tricked, increasing frustration and abandonment rates.
The page has horizontal scrolling
The page requires horizontal scrolling to view all content. This is a severe usability issue because it breaks the standard vertical scrolling pattern, forcing users to scroll in two dimensions. It violates flexibility and efficiency of use and aesthetic and minimalist design — users expect content to fit within the viewport width.
⭐ Key Takeaways
This lecture demonstrates that heuristic evaluation is a powerful, low-cost method for catching usability problems before user testing. Key violations to remember include: misleading or absent navigation cues (like highlighted tabs and breadcrumbs), deceptive link labels that mismatch destinations, irrelevant content (like version numbers), and layout problems (like horizontal scrolling). Every evaluation should check for consistency, clarity, feedback, and user expectations. These heuristics apply broadly across all digital interfaces.
🧠 Quick Revision Questions
- What is the purpose of a highlighted tab in a global navigation bar, and what problem arises when it is missing?
- Why is displaying a version number on a website’s main page considered a usability issue?
- How does a misleading link label (e.g., “Sign up now” leading to an online store) violate usability heuristics?
- What two specific heuristics are violated when a page requires horizontal scrolling?
- List three examples of heuristic violations identified in this lecture that relate to the principle of consistency and standards.
📘 Lecture 35 — Evaluation – Part VII
📖 Overview: This lecture explores the strategic nature of usability by examining key questions about users, technology, medium, and commitment. It then delves deeply into the unique nature of the Web as a super-medium, introducing a five-plane conceptual framework (Strategy, Scope, Structure, Skeleton, Surface) for understanding user experience and the tools used to solve design problems.
🗂️ Topics Covered
The lecture begins by establishing the relationship between evaluation and usability, posing critical questions about understanding users, technology, the medium, and organizational commitment. It then defines the Web as a super-medium combining print, video, audio, and software. The core of the lecture introduces a five-plane model (Surface, Skeleton, Structure, Scope, Strategy) built from bottom to top, and explores the historical duality of the Web as both an application and a hypertext information space. Finally, it maps the elements of user experience onto each plane, detailing functional specifications, content requirements, interaction design, information architecture, interface design, navigation design, and visual design.
📝 Lecture Summary
35.1 The relationship between evaluation and usability?
Evaluation helps uncover problems in the interface that improve the usability of the product. To achieve this, several fundamental questions must be asked: Do you understand the users? Do you understand the medium? Do you understand the technologies? Do you have commitment?
Regarding Technologies, you must understand the constraints of technology, what can be implemented using current technologies, and that building a good system requires a good understanding of technology constraints and potentials. For Users, you must know your users, their goals and behaviors, how they can be satisfied, and use goals and personas. Commitment involves ensuring that building usable systems has commitment at every level in your organization. Finally, for the Medium, you must understand the medium you are working in to build a good usable system.
Nature of the Web Medium
The World Wide Web is a combination of many different mediums of communication. It is a super medium which incorporates all of the following media: Print (newspapers, magazines, books), Video (TV, movies), Audio (radio, CDs, etc.), and Traditional software applications. Because of its very diverse nature, the Web is a unique medium and presents many challenges for its designers.
We can more clearly understand the nature of the Web by looking at a conceptual framework consisting of five planes.
The Surface Plane On the surface you see a series of Web pages, made up of images and text. Some of these images are things you can click on, performing some sort of function such as taking you to a shopping cart. Some of these images are just illustrations, such as a photograph of a book cover or the logo of the site itself.
The Skeleton Plane Beneath that surface is the skeleton of the site: the placement of buttons, tabs, photos, and blocks of text. The skeleton is designed to optimize the arrangement of these elements for maximum effect and efficiency—so that you remember the logo and can find that shopping cart button when you need it.
The Structure Plane The skeleton is a concrete expression of the more abstract structure of the site. The skeleton might define the placement of the interface elements on our checkout page; the structure would define how users got to that page and where they could go when they were finished there. The skeleton might define the arrangement of navigational items allowing the users to browse categories of books; the structure would define what those categories actually were.
The Scope Plane The structure defines the way in which the various features and functions of the site fit together. Just what those features and functions are constitutes the scope of the site. Some sites that sell books offer a feature that enables users to save previously used addresses. The question of whether that feature—or any feature—is included on a site is a question of scope.
The Strategy Plane The scope is fundamentally determined by the strategy of the site. This strategy incorporates not only what the people running the site want to get out of it but what the users want to get out of the site as well. In the case of our bookstore example, users want to buy books, and we want to sell them.
Building from Bottom to Top These five planes—strategy, scope, structure, skeleton, and surface—provide a conceptual framework for talking about user experience problems and the tools we use to solve them. When the Web started, it was just about hypertext. Tim Berners-Lee created it as a way for researchers to share and refer to each other's findings. People originally seized on the Web as a new publishing medium, but as technology advanced, the Web took on new capabilities, becoming more interactive and responding to user input like traditional desktop applications.
When the Web user experience community started to form, its members spoke two different languages. One group saw every problem as an application design problem. The other group saw the Web in terms of information distribution and retrieval. To address this basic duality, let's split our five planes down the middle. On the left, we'll put those elements specific to using the Web as a software interface, mainly concerned with tasks. On the right, we'll put the elements specific to hypertext information spaces, where our concern is information.
The Elements of User Experience Now we can map the whole confusing array of terms into the model.
The Strategy Plane: The same strategic concerns come into play for both software products and information spaces. User needs are the goals for the site from outside our organization. Balanced against user needs are our own site objectives (e.g., business goals like "Make $1 million in sales").
The Scope Plane: On the software side, strategy is translated into scope through functional specifications (a description of the "feature set"). On the information space side, scope takes the form of content requirements (a description of various content elements).
The Structure Plane: Scope is given structure on the software side through interaction design (defining how the system behaves in response to the user). For information spaces, the structure is information architecture (the arrangement of content elements).
The Skeleton Plane: The skeleton plane breaks down into three components. On both sides, we must address information design (presenting information to facilitate understanding). For software products, the skeleton includes interface design (arranging elements to interact with functionality). The interface for an information space is its navigation design (screen elements allowing users to move through the information architecture).
The Surface Plane: Regardless of whether we are dealing with a software product or an information space, our concern is the same: visual design, or the look of the finished product.
Using the Elements: Few sites fall exclusively on one side of this model. Within each plane, the elements must work together. For example, information design, navigation design, and interface design jointly define the skeleton of a site. In reality, the lines between these areas are not clearly drawn. A couple of additional factors go into shaping the final user experience, most importantly content—the single most important thing most Web sites can offer.
⭐ Key Takeaways
The lecture establishes that strategic usability requires a deep understanding of users, technologies, the medium, and organizational commitment. The Web is a unique super-medium that combines print, video, audio, and software applications, presenting distinct design challenges. The five-plane model (strategy, scope, structure, skeleton, surface) provides a comprehensive framework for analyzing user experience from abstract goals down to concrete visual design. A critical duality exists in Web design between the software/application perspective (focused on tasks) and the hypertext/information perspective (focused on information), which maps to different elements on each plane like functional specifications vs. content requirements and interaction design vs. information architecture. Ultimately, all elements on all planes must work together cohesively, with content being the most crucial factor for most Web sites.
🧠 Quick Revision Questions
- What are the four key areas of understanding required for strategic usability?
- What four media does the Web incorporate as a "super medium"?
- List the five planes of the conceptual framework for user experience, from top to bottom.
- What is the difference between the software side (tasks) and the hypertext side (information) of the Web duality?
- On the skeleton plane, what are the three components that must work together, and how do interface design and navigation design differ?