Welcome to the latest installment of my series on writing as user experience design. I’m discussing key principles of user-centered design with the aim of applying these ideas to writers’ contributions UX and UI.

Let’s talk about testing. As UX writers, we’re well aware that we work in a space obsessed with evaluating usability. And we love it! We’re in this for the users, after all. We try to understand their every value, want, and need so we can craft perfect pieces of UX copy and microcopy.

Yet our niche of the UX world is still new. While UX designers and researchers have a marvelous array of methods at their disposal, I’m often left wondering how UX writers can test our words. How do we narrow in on our part of the design process? What testing methods are available to us right now?

I don’t just mean readability scores. Flesch-Kincaid Index and similar metrics of readability can be a great baseline for user-friendly text, but they aren’t perfect. Nor can the algorithms powering these scores tell if your words are striking the right tone. As anyone writing in a particular brand voice can attest, every word needs to reflect the persona encoded by the internal style guide. Flesch-Kincaid can’t help you with that.

(I promise I’m not trying to be too harsh against readability scores. I use Hemingway all the time to keep my writing concise!)

Turns out, we aren’t starting from scratch when it comes to testing content and microcopy. Here are a few methods from around the web to help you start building testing into your process!

A note on participants: When testing comprehension, it’s important to recruit participants representative of your actual users. Proxy users aren’t enough. Running cloze tests on stakeholders from the marketing department or surveying engineers walking down the hallway won’t work here. These insiders bring company-specific assumptions and knowledge and can’t provide an accurate measure of comprehensibility.

Cloze testing

Evaluate users’ comprehension with a cloze test, sometimes called a cloze deletion test. Based on Gestalt theory of mental self-organization (“cloze” is related to “closure,” or the mechanism by which individuals understand an incomplete object as complete), this method was first proposed by W. L. Taylor in 1974.

Cloze tests are great for figuring out if your intended audience actually understands what you’ve written.

How to run a cloze test:

1. Select a piece of text and replace every Nth word with blanks. A higher N value makes the test easier. In typical tests, N = 6. Of course, UX writers working with microcopy will need to have a much smaller N!

2. Ask each participant to read the altered text and fill in the blanks with their best guesses about the missing words.

3. Calculate the score: the percentage of correctly guessed words. Synonyms and misspellings are allowed (in fact, if a synonym recurs, maybe that’s the word you’ve been looking for). If your participants score 60% or higher on average, your text is reasonably comprehensible for this user profile.

Highlighter testing

How to run highlighter testing

1. Select a piece of text and determine what you’re testing for. Code a set of highlighters according to these factors. In the linked example, participants highlighted words that made them feel more confident about the service in green and words that made them feel less confident about the service in red.
2. Provide participants with a printout of the text, the coded highlighters, and clear instructions about marking the page. If you’re testing for more than one value pair, avoid confusion by providing the user with a clean printout and new instructions for each pair.
3. Calculate the results. The colors provide an easily assessed visual indicator of what text works and what might need rewriting. If you’re a metrics person, you can also calculate statistics based on counts of color-coded lines.

Comprehension survey

Think back to your high school English class. Do you remember having weekly quizzes to make sure you were reading and understanding the assigned books? Your teacher was on to something.

Comprehension surveys are basically quizzes that test whether a user understands a piece of content. You can test long pieces, short strings, and everything in between.

How to run a comprehension survey

1. Determine the text you’d like to test. You can present it all at once, line-by-line, or a mix of both.
2. Develop questions. A List Apart has great tips for writing high quality questions, including stating questions in positive form, providing only one correct or clearly best answer in a multiple choice list, and providing incorrect but plausible distractor answers. You might want to add an option for “I don’t know,” as random guessing isn’t helpful here.
3. Provide your participant with the content and questionnaire. This isn’t the SAT, so keep things short — twenty minutes is more than enough time to gauge your user’s comprehension.
4. Calculate the scores. If you use a survey platform like SurveyMonkey or through a UX testing tool such as Optimal Workshop, you’ll get a handy score report automatically!

I hope these testing methods are useful for you. If you’re a UX writer or content writer and have other ideas about measuring the effectiveness of your words, let the world know in the comments!

Testing for UX writers: know when your words are working was originally published in UX Collective on Medium, where people are continuing the conversation by highlighting and responding to this story.

Your writing toolbox is overflowing.

Don’t get me wrong — it’s a great time to be a writer. The world is full of stories to tell, and it’s easier than ever to find an audience thanks to online publishing platforms like Medium.

And the tools! You’ve got tools to assist with every step of the writing process. You can organize notes with Evernote, draft material with Google Docs or Microsoft Office, conquer wordiness with Hemingway, and check grammar with Grammarly. Maybe, in an effort to streamline, you use a program that purports to do it all (looking at you, Scrivener!). There’s an app or program or widget for everything!

It can be overwhelming. I don’t know about you, but most of my writing takes place on a MacBook Air with no fewer than five apps and programs. Some play well with others and some are standalone, requiring repeated copy-pastes to move words around. And I mean around. Writing is rarely that linear trajectory we learned in school. When was the last time you started with an idea, built an outline, wrote three increasingly good drafts, and turned it in? Tenth grade? But with all of these tools trying to fit us into a straight and narrow writing system, frustration is inevitable.

Tool overload is why I decided to declutter my writing toolbox this spring. Sure, writing is a messy process. But wouldn’t it be great for that mess to look nice and be easy to navigate? Rather than switching between tools all the time, I want a program that lets me organize notes and photos in the same space that I draft material. I want to be able to reorganize the entire damn thing once I’ve written my way into an idea that’s actually worth talking about. Dealing with wordiness and grammar can come later. It’s that early phase of creativity that I want to manage in one space.

Welcome to the world of nonlinear writing.

In my quest for new tool to replace at least a few others, I came across an Australian startup focused on designing software that facilitates creativity. Milanote was originally intended for designers, but the team recently introduced several tools especially for us writers: longform mode, resizable notes, and a text clipper.

With Milanote, you can collect and visualize all the moving pieces that go into your writing. It’s simple: add notes, photos, and links to a board, where you can manipulate them with ease. You can upload DOCs and PDFs, if you’ve already got a project going. From there, you can drag these elements all over the board, adding lines, arrows, and colorful labels at will. You can create sub-boards for sections and chapters and easily compare drafts side-by-side. But wait, there’s more!

View your stock photography options right next to your backlink list. Pretend you’re Miranda Priestly preparing a spread by moving your titles, subheads, paragraphs, and images around. Reorganize the entire space to see how your thinking changes and break through writer’s block. And once you’ve cranked out a good draft, you can convert the note (or the entire board) to a more linear format: PDF, DOC, Markup, or Plain Text.

Writing on Milanote captures the feeling of sitting at those gorgeous stock photo desktops we all imagine ourselves working at. You know the ones: sleek laptop, blank pad of paper, cappuccino, and a tiny potted succulent? While you can still cram boards with inspiration and drafts, this unobtrusive UI won’t interfere with your creative process.

The future of writing?

At this point, Milanote is great for writers who deal in short and medium-length content. It’s not perfect. Printing a longform note requires you to first convert it to a linear format. And while the streamlined set of tools is a huge benefit — no distracting feature overload (looking at you again, Scrivener and Evernote) — the developers might consider adding a few more. Being able to see your word count while typing would be nice, as would additional formatting options and the ability to include images in longform notes. You also still have to manually move your text to Grammarly and Hemingway. But given that Milanote’s founders want to bring the way we write up to date with what software can do, I wouldn’t be surprised if those integrations are in the works.

Don’t just add Milanote to your already overstuffed writing toolkit. Milanote is the toolkit, or at least the giant container in the bottom half. Collect, create, organize, visualize, and draft all in one place. Like me, you still might reach for other tools once in a while, but creating that essential first draft will not longer be a hassle.

*****

Have you tried Milanote yet? Share what you’ve learned with other Writing Cooperative readers in the comments!

At The Writing Cooperative, our mission is to help each other write better. We’ve teamed up with ProWritingAid to do just that. Try it for free!

Spring Cleaning for Writers was originally published in The Writing Cooperative on Medium, where people are continuing the conversation by highlighting and responding to this story.

Following World War II, science and technology assumed an elevated position in American society. Technical innovations were expected bring unprecedented benefits to humankind. Jonas Salk had cured polio, televisions and radios brought immediate news and entertainment to directly to the home, and many believed atomic energy would soon power the world.

But a dark cloud hovered over this age of techno-enthusiasm. Just as fear of technological displacement had permeated the industrial era, public concerns in the postwar years included worries about mass unemployment caused by the new electronic computers. Would humans lose their jobs to faster, more efficient machines?

​Enter the Scientific American, a popular science magazine whose writers ignited enthusiasm for scientific and technological innovation while assuaging readers’ worries about what the future might hold. Science writing had been a staple of American journalism ever since the American Association for the Advancement of Science introduced the genre in the 1920s. Science journalists addressed society’s concerns in ways meant to sustain public support for continued technical development.

​When Dennis Flanagan and Donald Murray took over the Scientific American in 1948, they sought to “increase the immediacy, timeliness, and authority of the magazine” by publishing articles written primarily by subject matter experts: expository writing. Their efforts paid off. The magazine’s readership — it was and continues to be aimed at an educated base — grew steadily from 40,000 in 1947.

​Expository articles on computers published in the Scientific American between 1949 and 1955 reveal that readers found computers elusive and intimidating. The fears were not hyperbolic, but rather thoughtful reservations about how computers might impact society. These hesitant concerns reflected the magazine’s readership demographics: “scientists, engineers, and others looking for detailed articles about specialties outside their own.” In addressing these worries, writers characterized computers as analogs of human physiology and extensions of human capability.

During this five-year period, Scientific American authors tried to make computers more approachable by establishing a narrative about how computers could improve society. Of course, writers’ individual experiences and historical circumstances such as the Cold War influenced articles about computing. Set those themes aside for now though and instead focus on focusing how these writers understood cybernetics, computerized assistance, and the melding of man and machine.

​Writers employed two main analogies to describe computers: the human brain and nervous system and “inorganic organism.” Through these analogies writers made predictions about computing’s future. They imaginatively and critically addressed cybernetic possibility, in which in which computers would replace humans in certain menial and intellectual tasks while remaining subordinate to humans in general. Despite any capability or advantage machines may have held over the human brain, they were expected to always be safely under control of human programmers.

Giant brains and Mechano-biological organisms

Freight trains, scratch pads, and books — these were just a few comparisons between computers and other objects Scientific American writers made between 1949 and 1955. But of all the analogies made to help readers understand the computer, that of the human body or living being was most common. Computers, both analogue and digital, enormous and compact, were explained as mechanically analogical to human physiology and biological processes. Within this analogy, computers were either electronic, mechanical brains or inorganic organisms.

Most frequently, computers were described as similar to human brains. This comparison reflected a preoccupation with whether computers could think. According to professional science writer Harry M. Davis, “the use of ‘memory’ as a technical term of the computer trade has bolstered their anthropomorphic analogy to “brains.’” Davis considered the replacement of human intellectual and communications labor by computers to be nothing short of revolutionary. “The first phase of the Industrial Revolution meant the mechanization, then the electrification, of brawn. The new revolution means the mechanization and electrification of the brains,” he wrote in 1949.

​The brain analogy was more than an easy etymological transition from memory. According to Dartmouth mathematics professor John G. Kemeny, the famous mathematician John von Neumann made a “detailed comparison of human and mechanical brain in a series of lectures at Princeton University.” The “mechanical brains,” as Edmund C. Berkeley, author of Giant Brains, or Machines that Think, called them, were “endowed with the spark of automatic activity…the capacity to pay attention to and respond to a series of stimuli.” Just as a human brain gathered and processed information to make sense of the world, the computer brain received input in the form of a programmer’s commands and processed the information, either digitally or by analog means, to perform calculations.

Kemeny extended the brain metaphor to include the rest of the nervous system. One early computer, the Turing machine, “in fact resemble[d] a model of the human nervous system, which can be thought of as having a dial with many various positions and combining many simple acts to accomplish the enormous number of tasks a human being is capable of.” The dial, representing the brain, and the turning of the dial, representing the transmission of signals through neurological pathways, demonstrated the logical simplicity of Turing’s computer. Upon receiving and interpreting external stimulation, the brain sent tiny signals dictating the appropriate response through the nervous system. Similarly, the dial in Turing’s computer, upon receiving a command, spun to issue a series of responding signals to produce output.

​Kemeny continued the analogy by describing digital computing in relation to the nervous system, in which the brain interprets the binary firing or dormancy of a neuron as a signal requiring some type of action. Digital computers, which relied on the equivalent binary of on of off electrical signals, represented the “most complex logical machine” mimicking human neurology. Mathematician John von Neumann’s conceptual “universal machine” imagined a computer possessing a virtually complete electronic and mechanical nervous system. When a human programmer entered a program into the universal machine, the “neurons and transmission cells” would be “either quiescent or they [would] send out an impulse if properly stimulated.”

​Only one article contested the computer-brain analogy. In “Computer Memories,” Louis N. Ridenour, the first Chief Scientist of the United States Air Force in 1950, declared computer memory to be less like the human brain and more like the tools used by a “human computer” preparing a tax return. The computer’s inner memory was like a “scratch pad” used to store the data and instructions in current use by a computer. The intermediate step of memory was an “analogue to a human computer’s notebooks and files of documents,” while the third category corresponded “to books and similar large repositories of the knowledge of mankind.” Ridenour’s criticism of the brain-computer analogy stood alone; both the magazine writers and the authors of two letters responding to articles about computing seemed to prefer imagining the computer as an active, automatic, perhaps even “thinking” piece of equipment.

​The computer as organism was a less common analogy than that of the computer-brain or computer-nervous system, but its use shows that Scientific American writers considered computers to be not merely brains, but organisms. Lawrence P. Lessing, an editor of the Scientific American from 1953 to 1955, described computers as “impressive monsters” that “have proved harder to tame and put to work than we first thought.” Even if computers did not literally resemble humans or other organisms, writers still employed biological terms to describe computers’ components, processes, and potential for development.

​More than one journalist envisioned an evolutionary model of computer development, in which machines descended from less technically advanced ancestors toward future states of progressive improvement. Lessing adopted the taxonomic term “genus” to describe different generations of computers. “Big machines” were a “different genus from the smaller electronic computers or data processing machines that evolved out of them in some profusion after 1945.” Machines of different types shared the same types of “organs,” he suggested, furthering the biological nature of the system. Lessing’s 1954 article did not describe how the organs were connected to create an organism, but set the stage for Kemeny’s 1955 article that reaffirmed and enhanced the biological analogy.

​Kemeny made a more extensive comparison between computers and organisms when he reviewed von Neumann’s “universal machine” concept. Such a computing machine would be “remarkably human,” capable of learning, evolving, and even reproducing on its own. Kemeny sought to redefine the concept of living in order to show the possibilities for computing; the theoretical universal machine was “not alive,” but was still able to “create a new organism like itself out of simple parts contained in the environment.” If being alive simply meant the ability to reproduce, rather than possess organic material that was susceptible to change, growth, and decay depending on the ingestion of environmental factors, then computers could be considered living organisms.

​Von Neumann’s machine, comparable to a “higher order animal” would possess three kinds of parts to merge the “brains” and “brawn” necessary for reproduction. Neurons “similar to those… in the central nervous system,” provided the “logical control.” “Transmission cells” carried the messages from the brain to the “muscles” whose “primary use is, of course, the changing of an inert cell into a machine part.” Not only did the universal machine share neurons, cells, and organs with its human designers, but it also might imitate their behavior. Machines could “get into conflict with each other — imitating even in this their human designers” and, if left to reproduce in an environment of limited inert components, might even resort to killing each other. The universal machines, if given the command by their operators, might even undergo evolutionary processes. “If one might design the tails [of coded instruction] in such a way that in every cycle a small number of random changes occurred” without compromising the reproduction instruction, the machine would pass the “mutation” on.

​Writers understood the computers of the 1950s as “brain machines,” complete with neurons, and as organisms possessing brains, cells, organs, and even the potential for ears. Though they recognized that computers were not perfectly analogous to humans, the comparison enabled them to consider whether computers could “think” in the same way as their human users. The concept of the machine as an evolving, perhaps even reproducing, entity allowed for the possibility of a world in which humans and machines collaborated in the realms of labor, engineering, intellectual work, and even home management. Though Scientific American writers did not use the term “cybernetic” when describing the prospects of the human-computer relationship, they suspected that computers would become invaluable companions to humans.

Computer beings

Scientific American writers celebrated the rapid advances in computing made since World War II. They predicted that computers would assume control of many menial and tedious tasks that required humans at the time. Some even hoped that advances in computer speed and memory would allow computers to become intellectual companions to their human designers. While predicting the future benefits of advances in computing, writers also acknowledged the dark side of technology, including concern that computers would become dominant over human controllers. To alleviate these fears, writers always expressed that computers were subordinate to the human programmers who ultimately controlled them.

​Magazine writers celebrated the real and potential advantages of computers, particularly the speed and efficiency of digital machines. Digital computers were free from the “play” and “noise” that plagued even the most advanced analog computers. Ridenour suggested in 1952 that the Radio Corporation of America’s Typhoon missile simulator was likely “the most complicated analogue device ever built, and very possibly the most complicated that it will ever be rewarding to build.” The nature of digital computer programming enabled digital “brain machines” to operate logically and quickly.

​Although the human brain was more efficient electronically and chemically, it was also the “slowest” of computing machines, which possessed “advantage of high-speed operation, freedom from errors, and freedom from laziness.” If programmed correctly, future digital calculators and computers would “make all sorts of quick decisions that now require an alert and hard pressed human being.” Digital machines were “beginning to operate at levels of speed, temperature, atomic radiation and complexity that make automatic control [of tasks then requiring human operators] imperative… and the results are certain to be dramatic.”

​Scientific American writers also predicted that computers would supplement, if not replace, human labor in in both physical and intellectual occupations. Computers were already applied to tasks such as managing seat reservations for major airlines, collating flight schedules for the Civil Aeronautics Authority, managing inventory for the mail-order house John Plain & Co., and controlling human traffic flow at the Pennsylvania Railroad’s New York terminal. “Audrey” the Automatic Digit Recognizer machine developed by Bell Telephone, showcased the possibility of computers capable of hearing and listening.

Although Audrey could only “hear” ten numbers and sixteen English sounds spoken by one of her designers, Edward E. David celebrated Audrey as bringing a “talking, and listening, robot” closer to reality. Writers considered the new computers to be profitable, efficient alternatives to the expensive and error-prone human workers they replaced. Kemeny described the “electric eyes” of the New York terminal as “vastly faster than any doorman,” while the tally clerks John Plain employed to record orders during the holiday season could not keep up with their records without making errors.

​The writers did not mention potential job loss created by computers in these circumstances. Labor issues, particularly for jobs considered tedious or menial, were evidently not a concern for the writers. Beyond simple jobs, digital computers could also handle “elaborate engineering computations.” The computer could “take the place of several human brains” and had already replaced human calculators in aircraft manufacturing and missile piloting. Digital computers’ reliability compared favorably to that of the human brain in many writers’ eyes, meaning “problems that would not be practical for the human brain may be submitted to the machines.” In factory settings, computers could be used to calculate the “best way to distribute available manpower, funds, equipment, and so forth, to maximize a particular effort or to minimize cost.”

​The computer posed an opportunity to transfer control from human operators to steady, reliable machinery. Rather than requiring a human to operate “simple specialized control mechanisms,” the application of computers to automatic control in manufacturing, transportation, and even commercial industries would enable machine supervision of a whole job. Compact computers, perhaps successors to the small “Simple Simon” machine, might someday be found in average homes. These “little robots” would help their owners manage their household accounts and assist children with their homework.

Ultimately, computers might even become the intellectual partners of their makers. According to Turing, who one optimistic letter-writer quoted, “We may hope that machines will eventually compete with man in all purely intellectual fields.” Chess-playing computers already faced off against human opponents and could perform simple, word-by-word translations. These simple accomplishments and hopes for the future reflect the vision of a digital utopia. Computers would be programmed to handle tedious or complicated jobs for humans, assist them in managing their homes, and even provide opportunities for recreation and mental stimulation.

​Despite the apparent advantages of computers over humans in performing certain tasks, human operators held ultimate control over the machines. Interestingly, some of the faults assigned to computers were peculiarly anthropomorphic in nature. Computers, although faster and more reliable than human brains, were susceptible to “temperaments and moods” which led to incorrect answers; they were also “inconsistent in their mistakes.” The chess-playing computers occasionally made occasional bad moves based on same flawed reasoning. “This, of course, is precisely what human players do: no one plays a perfect game,” recognized Shannon.

Other shortcomings of computers emanated from the actual mechanics and programming of the machine. Davis quoted on expert: “the more I deal with these machines… the more impressed I am with how dumb they are. They do nothing creative. They can only follow instructions, which must be reduced to the simplest terms. If the instructions are wrong, the machines go wrong.” Computers could do only what they had been programmed to do; their logic worked by trial and error, “but the trials are trials that the program designer ordered the machine to make… the machine makes decisions, but these decisions were envisaged and provided for at the time of design.”

No matter how automatic or thoughtful machines appeared to be, human operators still controlled a computer’s functions. As advances occurred, machines might be given more routine responsibilities in factories, businesses, and, someday, homes, but the “human supervisor [would] still be vital to proper operation” and the “provision for human veto” of machine action would continue to be built in. “Though they replace other kinds of human mental effort,” predicted Davis, “the mathematical machines will never replace the mathematician.”

​Could a computer even be designed that did not require a human operator beyond the initial programming stage? Central to this question was the possibility of machine thought. “Could computers really think?” asked Davis in 1949. Davis was far from alone in asking this question; both magazine writers and letter writers posed the question and attempted to answer it through chess-playing scenarios, the conceptual possibilities advanced by von Neumann and Turing, and explanations of the actual electronics and mechanics of computing machines. At the heart of the argument against machine autonomy was the idea that the creative aspect of human thought gave humans the superior position.

​Berkeley and Shannon considered chess to be the best test of machine intellect. Both writers granted that machine brains were capable of logical reasoning. The chess-playing computer could be built to calculate the best move in any given situation. If thinking were regarded as a “property of external actions,” such as logical moves on a chessboard, then the machine could be considered a thinking entity. Adopting a psychological definition, Shannon defined thought as “essentially characterized by the following steps: various possible solutions of a problem are tried out…without being carried out physically; the best solution is selected by a mental evaluation of the results of these trials; and the solution found in this way is then acted upon.” Substituting the word “machine” for “mental” in this description, Shannon argued, rendered an exact definition of how computers operated.

​Only humans, however, employed truly flexible chess strategies marked by imagination. If a game of chess was a painting, only a human chess master could draw an outline of the strategy he would follow and follow it while making minor alterations here and there. Machines would “always make the same move in the same position” and could not learn from their mistakes. Either of these functions would require intervention from the programmer. The computer, at least as it existed in 1950 through 1955, lacked the creative capability to alter its own instructions, learn from its mistakes, or act without an initial human command.

​As early as the 1950s, computers were believed to extend human capability even though mechanical and programming limits rendered them subordinate to their human designers. Computers could be efficiently used to automate and manage industrial, transportation, and business tasks in addition to solving complicated equations for mathematicians and engineers. Some writers hoped that computers might eventually be applied to household duties, although such talk assumed computer designers could adequately miniaturize the machines.

​Despite the economical advantages of enlisting computers rather than humans for certain jobs, the machines still relied on human control. Computers could accomplish only what their coded instructions allowed, and were not yet capable of learning from mistakes and revising their programming to improve operations. Yet many authors expected that advances in computing would endow computers with these creative capabilities. Kemeny remarked that there was “no conclusive evidence for an essential [and permanent] gap between man and machine” and that “for every human activity we can conceive of a mechanical counterpart.”

​Turing, quoted by letter writer Samuel Ross, predicted that by the end of the twentieth century, “the use of words and general educated opinion will have altered so much that one will be able to speak of machines thinking without expecting to be contradicted.” If computers could not think for themselves in 1955, it was only a matter of time before advances in computer science endowed machines with real

As contributors to the Scientific American explored the human-machine divide, they determined that computers mimicked their human designers both physically and operationally. Computers were brains composed of neurons or mechanical organisms made up of brains, cells, and organs. Despite being limited by their lack of creativity, computers could compete with humans in a variety of industrial, business, and engineering jobs and face off against human opponents in chess. A few writers expressed hope that in the future, miniature computers would find a place in the average American household.

​Even though Scientific American writers did not use the term “cybernetic” when describing the prospects of the human-computer relationship, their bodily analogies and descriptions of how computers could be applied to human tasks indicated the potential for computers to extend human capabilities.

​The hopes and concerns of writers were, in some cases, tentative and would be resolved or forgotten within a few years. Expository writing about computers tapered off over the next five year period as developments in nuclear knowledge and extraterrestrial rocketry drew the public gaze elsewhere. By 1960, cybernetics pioneer Oliver Selfridge and cognitive psychologist Ulric Neisser published an article in the magazine declaring the question of whether machines could think to be an “old chestnut,” the answer to which was a definitive yes. New questions awaited computer scientists, for there was “not much doubt that [computers] can think, but they still cannot perceive.”​​

Interested in reading more? Articles from the Scientific American archives are available for a small fee.

Edward E. David, “Ears for Computers,” Scientific American 192 (1955): 92–98.

Harry M. Davis, “Mathematical Machines,” Scientific American 180 (1949): 29–39.

Charles A. Du Pont, “Letter,” Scientific American 182 (1950): 2.

John G. Kemeny, “Man Viewed as a Machine,” Scientific American 192 (1955): 58–67.

Lawrence P. Lessing, “Computers in Business,” Scientific American 190 (1954): 21–25.

Louis N. Ridenour, “Computer Memories,” Scientific American 192 (1955): 92–100 and “The Role of the Computer,” Scientific American 187 (1952): 116–130.

Samuel A. Ross, “Letter,” Scientific American 193 (1955): 2–6.

Oliver G. Selfridge and Ulric Neisser, “Pattern Recognition by Machine,” Scientific American 203 (1960): 60–68.

Claude E. Shannon, “A Chess-Playing Machine.” Scientific American 182 (1950): 48–50.

Thinking About Word Choice in UX Writing

Tl;dr: Consider the relationships words-as-objects have with readers

Welcome to the latest installment of my five-part series on writing as a user experience design. I’m exploring how user-centered design might apply to writing.

Today I’m talking about the affordances of words. I know, I know: words are signifiers. They communicate where action occurs. But I want to flip the script and ask what might we learn by considering words as designed objects with which human reader-users have relationships? What are the affordances of words?

Relationships 101

First, for you writers who don’t deal with design, let me clarify what affordances are. Don Norman, who first included the idea affordances in human-centered design, defines affordances “the relationships between the properties of an object and the capabilities of an agent that determines just how the object could possibly be used.”

Consider my favorite pen, the Paper Mate Flair. This pen is narrow and 5.5 inches long, making it easy and comfortable to hold. It emits ink, enabling me to write, or draw, or create an ink blot. As a leftie I especially like the felt tip, which lets me write without smearing ink across the page like I do with ballpoint pens and pencils. The pen’s plastic body means it’s never cold when I pick it up. One downside is that this pen doesn’t afford me is the ability to tuck it behind my ear when I need to type: it’s a little too wide.

These are just the intended affordances. These same properties allow other actions, depending on my creative capacity. Some of these unintended affordances are useful: I can use the pen to prop my phone up at a low angle. Other are ridiculous, but doable: the pen narrow enough to jam into the spout of my teapot, although it’s too long to fit all the way into the teapot. Why would I do that? I don’t know — what matters in this context is that I could.

Now let’s consider words. Words are artificial, designed for and used by humans. While philosophers have argued over what words are for aeons, I think we can agree that words have properties. Some properties include:

• Word length
• Letter combinations
• Etymology
• Language
• Organization (syntax, grammar)
• Meaning/semantics

Meanwhile, humans have capabilities that enable interaction with what would otherwise be weird marks on paper or blackboard or screen:

• Reading level/ability
• Language proficiency
• Accessibility (such as vision and processing differences)
• Interpretive creativity

Combine these human capabilities with word properties and you end up with infinite possible relationships between reader and text. Consider two warnings:

Assuming you can read English at an 8th grade level and are able to see the words on this page, you probably reacted differently to these similar warnings. A relatively vague statement, “caution” puts you on edge without prescribing an action. Learning you’re being warned about falling objects might prompt you to look around, but you also might interpret it as something to keep in mind. “Look out” is more urgent. It demands immediate awareness of your situation and prescribes a specific action for you to undertake.

You have different relationships with these words: in one case, general unease, for which you might take action; in the other, an immediate response to a clear threat. The word properties and your reading capability interact to afford these unique relationships.

Context matters, too. If you’re driving on a mountain road and you see a sign instructing you to look out for falling objects, you’ll probably assume rocks could tumble onto the road. But how do you react if you’re picnicking in the middle of an empty field and a sign tells you to look out for falling objects? Do you worry about birds or meteors or UFOs dropping out of the sky? Do you dismiss the sign as not making any sense after looking around? Or maybe you know the field is next to a golf course, entailing the risk of a stray golf balls. Context shapes your interaction with the words.

So why does it matter that words, considered as objects, might offer affordances? Despite marketers and tech titans who argue that content is king to justify low-grade filler for websites, the best writing engages in a conversation with readers. When considering the words you write, it might help to consider how individual words and phrases afford particular converational relationships with your audience.

Applied Affordances: Making Commitments in E-commerce

Many e-commerce websites post an “add to cart” or “add to bag” button (known as a “call to action”) somewhere on product pages.

At first glance, these are specific, action-oriented words intended to guide the customer into a transaction. But do these phrases also have an unintended affordance of letting shoppers add items to their carts without really pushing them to purchase those items? These words are noncommittal: customers can add shirts and jeans to their Madewell cart all day without ever feeling like they need to make a purchase. Maybe this is convenient for casual shoppers, but surely it’s not the effect businesses want.

A few companies seem to be wrangling noncommittal customers. Google’s online store requires customers to a click a button that says “Buy” to order a new phone. The next page, where customers choose the phone’s specs, also ends in a “Buy” button. These buttons don’t give shoppers the opportunity to question whether want they to make this purchase: they’re on a direct path to a transaction.

Amazon, ever customer-oriented, offers even more clever variation. Prime members can add an item to my cart or “Buy Now.” This second option is genius: it demands immediate consumer action. Buying now doesn’t even require a visit to a confirmation screen. Just click the button and the purchase is complete! These words afford a sense of immediate wish(list) fulfillment and efficiency. By the time a customer questions whether they need a rose gold desk lamp, she’s already committed to the purchase and moved on to looking at matching bookends.

Even if words and phrases are properly categorized as signifiers, I still think writers, especially those involved in product design and copywriting, can benefit from sometimes considering the affordances of their writing as a designed object. Stuck writing a sentence or confused about why users keep misinterpreting a call to action? Try to list all the affordances a word might have for a particular audience or user. Let this process be a strange and unusual analytical intervention in your writing toolbox and give yourself the chance to redesign to accomplish your intention.

What do you think? Can people and words exist in a distributive affordance relationship?

Check back in two weeks for part three of this series, where I’ll look at testing!

o you have ideas about user experience, design, and writing you’d like me to explore? E-mail me at annietadams@gmail.com.

Learn more about me.

Why Writers Should Apply User Experience Design Principles to Their Sentences

TL;DR: Write to clarify, not obscure. Give users time and mental bandwidth to appreciate and apply your message, not struggle over your prose.

Welcome to the first of my five-part series on writing as a user experience activity. Over the next few weeks, I’ll consider how key principles in user-centered design might be applied to text. I first became interested in relating UX concepts to writing as a Ph.D. candidate studying the history of technology. Academic writing is notorious (paywall) for confusing even the trained scholars to whom it’s directed. In my quest to write accessible, engaging stories, I realized it would be silly, even counterproductive, to ignore the incredible body of research addressing how individuals interact with products. After all, my writing is my product. My readers are my users.

Let’s talk about why writers should consider their readers as users of texts. Texts are products — they’re artificial and constructed for human interaction. They have many possible purposes: people read for leisure, to collect information, to be convinced (or at least to validate their existing convictions). Similarly, writers write to entertain, to inform, and to persuade.

But just at designers should not design for themselves, writers should write for their readers. When you write, do you use unusual words or turns of phrase to impress yourself, or do your choices clarify ideas for your readers? Consider this: do you utilize sesquipedalian terminology to optimize self-satisfaction or do you want your reader to know what you’re saying (and not hate you)?

In Several Short Sentences About Writing, Verlyn Klinkenborg challenges the notion that writers should experience a sense of “flow” as they produce text:

“Flow is something the reader experiences, not the writer.”

Writers work slowly and painstakingly through iterations (beta versions, if you want). We test these drafts (prototypes) on sample audiences comprised of our students, our partners, and our colleagues. We solicit their feedback to improve each round of development until publication (the writer’s product release). At this final stage, it’s the reader for whom the product must appear effortless. It’s the reader whose attention must be engaged, who has to be convinced to continue reading in order to resolve whatever information gap or leisure requirement drove them to read in the first place.

Building user-friendly “flow” into your writing is important for all writers, not just those who design novels and narratives. Maybe you write product descriptions. You want to persuade your reader to buy, so write for them. Engage their attention by emphasizing benefits using language they’ll relate to, rather than just describing features in tech-speak and industry jargon meant to make your product look sophisticated. As the UXR firm Nielsen Norman Group points out,

“Users shouldn’t have to interpret content; they should be able to easily understand it.”

The same goes for writing products. The goal here isn’t necessarily to help your readers avoid thinking. If you’ve written an essay on the complexity of modern feminism, you want to inspire consideration and reflection. Your readers probably want to learn something. Don’t exhaust their mental bandwidth by making them struggle to figure out your message. Help them understand the text so they have time to proceed: with preparing their dinner, with installing their dashboard phone mount, with discussing how your brilliance compares to your rival’s.

What can you do to understand your readers as users? First, you need to know your audience. Who exactly are you writing for? Next, empathize with them. What problems or challenges might keep them from using or appreciating your text? Are they pressed for time? Already dealing with a stack of reading material? Do they just want to make dinner in under fifteen minutes? If you have several potential audiences, consider them all and list the concerns each group has.

Remember, even identifying user needs is iterative. Your audience might not know exactly what challenges they face. Don’t just take the first answer you hear (or think of, if you’re brainstorming) and run with it. Ask follow-up questions, read between the lines, and revisit the subject throughout your design process to make sure you’ve stayed on track.

You can’t design your writing to meet particular user needs without understanding why those needs exist. Once you’ve established your user’s needs, you can write directly with them rather than at them. Bring your ideas and words to them; it’s an easier sell than making them come to you.

If you’re a writer, how do you understand your readers’ needs? Readers, what are your favorite pieces of user-friendly writing?

Check back the first week of March for part two of this series, where I’ll explore the affordances of words.

Do you have ideas about user experience, design, and writing you’d like me to explore? E-mail me at annietadams@gmail.com. Learn more about me here.