In 2016, the artist Deborah Roberts created “Pluralism Series,” a series of prints that, borrowing the aesthetic of a Microsoft Word document, lists the names of Black people. Name after name has that jarring squiggly red line underneath it, as if it was a misspelling.
I know this squiggly red line all too well. As a Black, Nigerian woman, design and technology often fail to acknowledge my middle name “Ogorchukwu” or my last name “Iyamah.” Roberts’ piece made me think of the many ways in which my name has been “othered” by design. Sometimes it’s autocorrect changing my name to the nearest Anglo name; sometimes it’s a form’s input field telling me my middle name has too many characters, and sometimes it’s the notification that the name in my email is simply invalid.
One of the main barriers to designing equitable realities that dignify communities of color is something I call the white default. The white default speaks to the ways in which society teaches us that to be white is to be “the norm.” Designers often use the white default to make assumptions about how everyone shows up in the world, which ends up contributing to practices that exclude. This can look like facial recognition that is unable to detect people with dark skin, “nude” products such as Band-Aids that only cater to white skin, speech-to-text software that mistranscribes Black speakers nearly twice as often as white speakers, or medical illustrations that only display white skin.
In an effort to get a better sense of how design and technology have impacted people’s experiences with their names, I created a community survey titled, “Say My Name, Say My Name.” Over three days, I received 257 responses from Black, Indigenous, and people of color across the diaspora. I also spoke to experts who are passionate about ensuring the inclusion of people’s names in technology.
The results were telling:
- 87% of participants responded that spellcheck implies that their name is incorrect.
- 80% of participants responded that autocorrect changes their name to another name.
- 50% of participants responded that speech-to-text software (dictation) autocorrects their name.
- 23% of participants responded that form fields make an assumption that they have a middle name.
- 23% of participants responded that they are notified that their name is incorrect when creating an account.
- 21% of participants responded that form fields don’t allow them to use accents, hyphens, apostrophes or symbols in their name.
- 21% of participants responded that their name exceeds character limits on the form.
As one participant wrote: “From kindergarten to fifth grade, my parents shortened my name so that it would be easier for people and software to spell and pronounce correctly.” Anurag Agarwalla, an engineering manager at Uber, explained how many “white default” design assumptions occur even before people interact with digital interfaces. “The keyboards on which we type are optimized for English, so at the input level there is a natural disadvantage,” he says.
While some participants in the survey responded that they add their name to the dictionary of the software they are using or simply give up on whatever it is they were trying to submit, 38% of participants responded that they shortened their name, 30% responded that they used a different name, and 22% responded that they changed the spelling of their name. “My name has history,” one survey participant responded, “and connects me to my ancestors who had it as a last name so it’s frustrating when technology undermines its importance.”
People have to constantly engage in what I call digital assimilation, which speaks to the ways in which Black, Indigenous, and people of color are urged to whiten ourselves in order to mitigate the racist barriers we face in the digital world. People often view assimilation as something that is voluntary, but when digital products default whiteness, assimilation is always covertly or overtly forced.
People often view assimilation as something that is voluntary, but when digital products default whiteness, assimilation is always covertly or overtly forced.
Digital assimilation happens constantly, from changing the way people pronounce words in their native tongues so that Siri can understand, to using specific lighting and styling so that Zoom virtual backgrounds work as intended (even with these changes, it still doesn’t always identify Black folks). Through each of these interactions, we are told that our ability to survive in this society is linked with our ability to abandon our cultural markers—clothing, mannerisms, voices, language, accents, appearance, and names. As Joseph Akoni, a product manager at Google said, “When someone hears someone mispronounce their name without making an effort to say it correctly, it’s almost as if the person is hearing that their background or their culture doesn’t matter; as if that person must Americanize themselves and deal with it.”
Thien Joseph Dang a UX strategist who wrote an article about name mispronunciations in tech, has thought about what it would take to design a more inclusive future for people with non-anglo names and identities. He says it starts with shifting the design process away from a euro-centric mindset by trying challenging defaults and centering non-white folks in design process. “Including these voices in the design process, through diverse teams and also listening to our diverse users, will help identify where products are restrictive and not fit-for-purpose,” he says.
There are some individuals who are already creating products, features, and resources to do this. Joseph Akoni, for example, created The P-Zero PM, a website that provides aspiring product managers with tips and tools. And before joining Google, Akoni worked at Linkedin on a name pronunciation feature that allows people to add an audio recording of their name pronunciation to their Linkedin profile so that others can listen to it. This feature plays an important role in creating an inclusive workforce as it enables people to be mindful about pronouncing each other’s names correctly during calls, interviews, and meetings.
Or consider Richard Ishida, an internationalization lead at World Wide Web Consortium (WC3), who coordinated the MultilingualWeb initiative which highlights standards and best practices that support the creation of multilingual web-based information. Ishida, who has a background in translation, computational linguistics, translation tools, interpreting, and software engineering, also developed the WC3 Internationalization Checker, which performs tests on web pages to determine if they work well across countries. He writes culturally inclusive informational articles such as “Personal Names Around the World” which provides engineers with tips on how to design forms, databases, and ontologies that caters to different types of names. This work is important as it serves to sensitize engineers to names around the world and push them to think beyond current constructs, both literally and figuratively.
Some companies are built around these ideas, like Namecoach, a service that creates accurate audio name pronunciations and integrates them into the tools that schools and businesses use such as Blackboard, Canvas, and Salesforce. In a blog post, founder Praveen Shanbhag shared that Namecoach started after he, his immediate family, and their relatives who had traveled from afar went to his sister’s graduation. “When it came time for her to cross the stage, her name was butchered,” he said. “Marring the big moment we had all come to see.” It occurred to Shanbhag that the problem could be solved if people could just hear how to say a name on demand. The name pronunciations at Namecoach can either be generated by users or drawn from a database of accurate audio name pronunciations. The database highlights the most likely pronunciations based on factors such as ethnicity, nationality, location, and gender.
When we design beyond the white default, we create experiences that allow people to show up authentically.
The benefits of designing inclusively are priceless. When we design beyond the white default, we create experiences that allow people to show up authentically, we reach larger markets, we spend less energy having to fix our reputation and products post-launch, and we ensure that people aren’t paying for a lesser product. This last issue is something that is often not talked about. Rashmi Dyal-Chand, a law professor who wrote the research paper “Auto-Correcting for Whiteness” highlights how when we do not design exclusively, we are producing economic inequalities between consumers. “The most direct economic losses are those we experience as consumers: we are paying for a bad product, one that disparately affects people on the basis of race and ethnicity (and many other differences).”
As researchers, product managers, designers, data scientists, and engineers in the technology industry, we have a responsibility to create products, services, and experiences that empower people to feel like they can show up as their whole selves.
Sometimes it starts with recognizing that people’s names don’t fit within a defined amount of space.
Sometimes it starts with recognizing that people’s names can contain accents, hyphens, and symbols.
Sometimes it starts by saying a name — and saying it right.
