Basha
The Challenge
People looking to learn less common languages face challenges in finding resources and tools to aid their learning process, unlike those learning more common languages like Spanish, English, Mandarin, and Hindi. These challenges include sourcing their own translators and the inability to take advantage of automated, user-friendly tools. Often, individuals seeking to learn less common languages are 2nd or 3rd generation immigrants looking to reconnect with their family's culture.
The Solution
Basha is a free tool that empowers individuals to learn less common languages by providing spaced-repetition learning with downloadable Anki decks and recruiting community members for textual and audio translations. The solution was to create a user-friendly marketing site that highlighted the importance of cultural preservation and community engagement, while the app would allow individuals to learn less common languages, promote cultural heritage, and encourage community involvement, all while remaining free to use.
Research Method: General Inquiry
I knew a little about the young second-generation immigrant experience due to prior experience working as an ESL teacher. As you’ll read on the Basha marketing site, this provided insight into shortcomings of the schooling system that lead to ESL students losing their native language. Basha was built in accordance with someone who intended to use the app as a learner, himself; because he and his family members fit the profile of the target user, he was able to answer any questions regarding the audience. There are also many online communities that consist of the same attributes, which we referenced to avoid confirmation bias.
I also read some research papers regarding how spaced-repetition learning works in practice and had the tech lead walk me through the different workarounds to avoid paying hosting fees - I then provided guidance on how to balance this restriction with usability.
Research Synthesis: User Personas
Basha will have two distinct user profiles:
The Learner: This user is primary in the sense that the app is built around meeting their goals; they drive the engagement with their translator, doing the initial set-up and giving them the materials necessary to complete the translations.
The Translator: This user only begins using Basha after the learner sets up the portal. Their goal is to support the learner.
Basha users must be able to:
Access an individualized learning portal by running a server instance.
Translators must be able to access the learner’s portal by clicking a personalized link.
Upload decks as text files and download completed decks as Anki files. C/R/U/D all apply on the learner’s end, the translator can only read and update decks.
Translator must be able to quickly cycle through a deck’s editable phrases, type in latin and not latin characters, and record audio per phrase.
Locate help guides and community decks through the learning portal.
User Flow
Visual Design: Style Tile
Our intended primary audience, Indian-Americans, drove the visual design aspect of Basha. The symbolism of color varies by culture; for example, white is associated with purity in most European natures but East-Asian cultures associate it with death and misfortune. You can also see cultural differences as it relates to color by looking at flags (the color palette that a nation chooses to represent their values through the color that symbolizes said value) - for example, look at how many western nations use the red, blue, and white combination in contrast to the rest of the world. Since we are targeting one demographic group, it was important to choose visual design elements that would align well with that culture.
Logo: Consists of the sanskrit characters that spell out Basha in Marathi, layered in a condensed square. To give depth, the characters are colored with the Basha brand yellow-turqoise gradient and have an inner shadow.
Colors: Orange: After researching color symbolism in different Indian cultures, orange consistently stands out as a very powerful color . Aside from being featured in the Indian flag, it represents the ‘quest for light’ in Hinduism. All secondary colors were selected for how they paired with/contrasted the tangerine orange in addition to symbolism in Indian cultures - which you can read about here.
WireFrames
Survey
After an observation of a small user group of translators, we conducted a survey broaden our view to the general public. Our first method tested a smaller group of users who were already familiar with Basha and committed to using it. Our second method introduced Basha to a larger prospective audience to measure the likelihood of organic adoption. Another contrast is that while our first group consisted of translators of one racial and age demographic, this new group were of unknown demographic backgrounds and mainly interested in being learners.
We wanted to share Basha with the general public for feedback. We posted a general introduction and survey link to the app on a major language-learning community on Reddit. We received many positive comments and ‘upvotes’, however only two people completed the survey (which was actually more than we were expecting since they weren’t being compensated for taking it.). Though this obviously is too small of a pool to draw any conclusions, I think it is still worth sharing to highlight insights and demonstrate my method in conducting survey research.
Aside from wanting to gather opinions on the ease, usefulness, and design of Basha, I also added questions to test comprehension/if they read through the marketing site. I’ll expand on these insights in the Iteration and Conclusion section.
Iteration and Conclusion
The survey results suggest that we need to make a few key changes to improve the impact of Basha in the next version. Specifically, we need to add more emphasis on what Basha is not, as some participants seemed to misunderstand its intended use. To address this, we could do a better job of clarifying that Basha is not meant to assist learners with popular languages and is solely intended for use cases where other resources don't exist.
Additionally, we plan to add practice files and build a library of decks to increase user buy-in and provide a preview of the functionality before engaging a translator. In-app messaging between the learner and translator would also help speed up workflows.
Finally, we should remove the Render section of the start-up flow to minimize the time investment required before users can start using Basha. By implementing these changes, we can ensure that Basha is more effective and user-friendly in the next version.
