Polyglot: Project Documentation

The initial motivation for our project centered around creating an engaging language learning experience.  The following report outlines that process and finishes with our reflections on the experience.

Initial Brainstorm

The first few meetings started with discussing the scope of the project, what we wanted our platform to have, and some initial sites to start researching what was on the market. After some discussion we came up with a few of the following criteria as well as some initial research to see what other language applications were on the market.

Scope

  • Young adults and adults

  • ESL

  • Mobile

  • Level(s)

  • Skills to practice

 

We want to build a mobile platform with the following features:

  • using mobile technology to scale learning

  • integrated/immersive learning

  • flexibility of learning (personalization)

  • student-centred and/or students-generated content

  • allow for sustained engagement

  • creates an authentic learning environment

  • encourage fluency but also accuracy

  • Incorporates a safe, fun, and inspiring online community

 

Some ideas to start with:

  • Fill in the blanks (speech bubbles)

  • Use random cards (with a word and picture) to complete a sentence

  • User Generated Content (e.g. students taking photos/video and then learning based on them perhaps using a crowd-source model) for JIT learning

 

Sites to research:

  • EnglishCentral

  • Memrise

  • LiveMocha

  • DuoLingo

 

User Profiles

After coming up with some initial ideas, we started to think about what kind of users would use our product.  All of us have been language learners, and two of us are actively involved in the language learning and teaching process.  Again we brainstormed some ideas for possible usage scenarios based on our experiences and dove deeper into what the user needs might be, where they were located, and how they might use the product.  Listed below are the usage scenarios and the profiles we came up with.

User scenario 1:

Background

Two university students, Taro and Adam, have agreed to do conversation exchanges via Skype.  Taro wants to learn English conversation and Adam wants to improve his Japanese skills.  Taro is a student at University X in Tokyo, Adam a student at University Y in Boston.

Taro has the basic knowledge of English but his oral skills (i.e., speaking and listening) are weak (compared to his reading and writing skills).  He wants to improve his oral skills via interacting with a native speaker of English directly.

He is also interested in learning vocabularies/expressions specific to his field (biotech) as he plans to do an internship next summer at one biotech company in Cambridge.

Adam has taken two years of Japanese at University Y  in Boston.  He can have basic conversations in Japanese.  He is interested in improving his Japanese overall skills.  Also, he is interested in learning Japanese customs and culture.  He plans to do a homestay program in Japan next year.

User Scenario

Taro and Adam collect what they want to learn via mobile devices (e.g., photos, video’s).  They can gather their learning materials in real life or from the web sites.  They meet once a week via Skype.  The meeting consists of two sessions: (i) Japanese session where they both speak in Japanese and discuss what Adam wants to learn and (ii) English session where they both speak in English and discuss what Taro wants to learn.

User Scenario 2:

 

User: Krishna Prakash, Office Boy

Background

Krishna was born and grew up Belathur, a village two hours away from Bangalore. He studied at the local government run school until grade six after which he helped his father run the family cashew farm. At the age of fifteen he moved to the city because of the opportunities to earn more money there. For three years he worked in a steel plant doing unskilled labor, before finding a less strenuous job waiting tables at a roadside restaurant. Rakesh Murthy, a sales rep for a software distributor would often eat at the restaurant and struck up a rapport with Krishna. When the peon at Rakesh’s company left, Rakesh recommended Krishna for the job.

Krishna’s duties involve doing odd jobs around the office where around twelve people work:

  • making tea and coffee

  • opening the office in the morning for the cleaning lady

  • delivering the newspaper to the manager’s desk

  • washing dishes

  • fetching lunch from a nearby restaurant 2-3 days a week

  • fetching cigarettes and Starbucks coffee for the staff

  • sweeping the office every night and locking up

  • buying office supplies

 

He earns Rs. 4,000 (around $50) per month and lives in a tiny room in a slum two hours away from the office. He spends four hours on his daily commute by bus.

 

In contrast, the people who work in the office earn between $1,200 and $5,400 per month.

 

Krishna only knows very few words in English – like ‘water’, ‘tea’, ‘coffee’, ‘cigarettes’ etc. He only speaks the local language – Kannada. Half of the people in the office speak Kannada so complicated tasks can be explained to him. Most of the time though, he doesn’t need much more knowledge of English to do his job apart from the basic words he knows.

 

Key Goals

Krishna has a friend Hari who was more fortunate than him. Hari stayed in school till grade 12 and had some basic English skills. His parents were able to pay for him to study English for 1 year after school and now he works in an office as an junior telemarketer earning a salary of Rs. 8,000 (double that of Krishna). Krishna would dearly love to learn English so that he can get a better job. His daily contact with the people in the office are a constant reminder of how his life could be better if only his English were better.

 

Usage Scenario

Krishna saved for one year to buy a smartphone. He paid a whole month’s salary for it and it is his most prized possession. He can listen to music on it and play games while he is commuting to and from work.

 

 

User Scenario 3:

 

User: Yuka

Background:

Yuka is a Japanese girl from a middle class family in Japan. Her parents take her education very seriously but since she is the eldest of 2 other kids, the parents are cautious about their spending. They pay for the private school, but they cannot afford to send her an English preparation course such as TOEFL prep, which will cost the family around $10K a year.  Yuka has always wanted to live abroad, master English, and become an architect while she has no idea how to even get started.  She faces the following problems when learning English:

  • unqualified English teachers at school

  • TOEFL exam questions that she does not understand how to approach

  • high cost of a test preparation school

  • her local community and family members do not have any information on the topic

The diagram below outlines other problems that Yuka faces:

 

Key Goals

An online platform that Yuka can learn and practice how to become “a near native speaker” and a scholar in English language and cultural context. By engaging in various projects offered at this site, she will not only explore her interests but also learn to speak, read, and write in English naturally. Within the 3 years of heavy engagement, she only spent 6 months in TOEFL/SAT prep while being able to get good enough score for her to acquire scholarships (offered domestically) for a liberal arts college.

 

Use Scenario

She can access the site with her iPhone during her commuting time to school (1.5 hours each day), free time, and any other time she has data access to.  She makes friends (who are partner school kids, college volunteers, or other high school/middle school kids learning English from all over the world) there, collaborate on projects, and really enjoys the gamified and truly engaging system that allows her creativity flow.

 

User Scenario 4:

User: Clare

Background

Clare is the Director of Operations at a multinational company in Costa Rica that has just opened operations there.  She was sent to Costa Rica from the US to jump start operations there and get local staff and management up to speed on the latest projects.  She may be in the country for a year or two.  Her company has put her in accommodation in an upscale area in San Jose where most expats typically live.  She has almost everything she needs near the house and a local driver takes her around. At work, she speaks English as all staff members must have at least a 6.5 to 7.0 on the IELTS or passed the FCE, CAE, or CPE (Cambridge Proficiency exams). Her job is very demanding and she works long hours.

Key goals

Clare would really like to learn Spanish given that she might be living in Costa Rica for the next two years.  She studied for a couple of years in high school, but then never continued after that. She doesn’t always like living in an expat bubble and would love to explore the city and the rest of the country more, but feels that this might be difficult with her very limited Spanish. She also feels that it would really help develop relations with local staff and allow her to learn more about Costa Rican culture and customs. Her company is willing to pay for private language instruction, but it is difficult for her to have regularly scheduled classes given her busy schedule. Thus, she would like to have the flexibility to do some independent study on her own.

Usage scenario  

Because of the demands of her job, she has a desktop computer, laptop, tablet, and smartphone. She works on her computer all day, and usually uses her tablet at home for entertainment and surfing the internet.  Often, she also brings work home with her and does that via the laptop.  She also carries a smartphone around with her and uses that for texting, e-mail, some simple games, managing her schedule, and doing some simple work tasks. She generally has some free time in the evenings, tries to take an hour lunch about 3 times a week to get away from the office, and tries to have one day to herself on the weekend. She would like practice her Spanish via her tablet or mobile phone as those devices are more portable and she can study while being driven around the city, travelling, or just sitting on the couch at home. Initially, she would love to speak Spanish in restaurants, ask directions, go shopping for clothes at the mall, shop and bargain at the local market, and have a simple, but extended conversation with locals to find out about more about them and their country.

 

Design Phases and Decisions

Iteration 1:

Once we came up with user personas, we tried brainstorming ideas again, but had difficulties during this stage identifying what our next steps were.  We tried coming up with potential ideas for what what the final product might look like and came up with the following:

  • plugin model – pull in apps, plugins or a toolkit based on your language needs

  • feature driven

  • example document you can take a picture of, upload, put comments and ask questions, crowd-sourced answers

  • plugin model – pull in apps, plugins or a toolkit based on your language needs

  • feature driven

However, we were not sure where to go from there.  It seemed as if we were going in circles.  Therefore, we tried picking a user and trying to imagine how a user might use a particular application and came up with the first, rough design for the user Clare.  

How Clare uses technology in her daily life:

How it might work:



The idea focused on vocabulary learning and incorporated the concepts of capture, communicate, and creating new content.  We decided that vocabulary learning was a good area for focus and met with Pattie to get feedback.  Based on our meeting with Pattie, we changed slightly and came up with slightly more focused “how might we..” questions.  These are listed below:

 

“How Might We” list

  1. How might we actually make this platform about language learning and not just another social media network?

  2. How might learners actually know they are improving?

  3. How might learners actually improve their level?

  4. How might we provide enough scaffolding to support learners but also still allow them to be self-directed in their learning?

  5. How might we engage users and provide a structure or platform that allows the content produced to be intrinsically motivating?

  6. How might we create a community of learners feel to the platform where learning is symmetrical rather than one-directional and unequal?

  7. How might we get expert feedback into the process to help validate learners’ experiences and attempts?

  8. How might we take into account non-roman script languages?

  9. How might we design the platform so that learners can migrate towards what interests them?

Iteration 2:

During this iteration, we focused on the following issues:

 

Target audience: what level of language learners are we going to target for?  How does the design should differ depending on each level?

  • 3 levels: beginner, intermediate, advanced

  • beginners – focus on word-level features (pronunciation, spelling, maybe simple sentences, etc.)

  • intermediate – longer sentences and some dialogue around the picture and context

  • advanced – full discussions about picture, context, culture, etc.

 

Just In-Time learning: How can we incorporate the concept of just in-time learning?

  • “learning on the go” – users can capture content via mobile device (photo’s or video’s)

 

Learning Progress: How can we let users know/ensure that they are learning and advancing their language skills?

  • flashcards to keep a record of that content

  • allowing learner’s to induce grammar and vocabulary rules by pulling in authentic examples from the web and from peer feedback

 

Content Categorization: Can we categorize/cluster the contents that users post, so that we can utilize these data/metadata later on?

  • categorizing content so that participants can organize as well as access content by their interests using hashtags.

Example: I can organize my flashcards by hashtagging them with #food and I can look and discuss other people’s cards on #dance

 

Personalization and Beyond Vocabulary: We incorporated this concept into “Phrases” feature.

  • some kind of space for learners to actually create, produce, or use the language they are learning

  • some kind of feature that allows learners to indicate on the picture/video/audio/text  what it is they need they help with

 

Correct/Incorrect & Peer Feedback:

  • allowing users to indicate by correctness and incorrectness of a response as well as by whether they like it or not

  • peer and machine feedback

Example: users can post particular sections of something they have written or said and ask for feedback from peers as well as utilize features like spell check, grammar check, or even some translation

  • focus on English language only or at least one other

  • users can build their reputation and also move up a level

 

During this iteration, we came up with the following questions:

  1. What’s relationship will the users have with each other?

    1. symmetric p2p – sometimes you are the mentor and sometimes you are the mentee

    2. native speakers to non-native speakers

      1. What will attract native speakers to the platform if they are not learning a language?

    3. advanced speakers to less advanced speakers

      1. Is it okay if their feedback may not be accurate or correct from a native speaker’s standpoint?

      2. Begs the question..are native speakers always going to be correct about their own language and be able to give meaningful feedback to a learner?

  2. What is the motivation for users to help each other?

    1. gamified aspects – building a reputation that other users can see by the quality and/or quantity of the responses you give

    2. you are learning a language and want others’ help, you help others in the hope that they will return the favor

    3. allowing learners to ask more advanced or native speakers for help or feedback maybe via private message or direct reply

  3. Should a learner’s reputation be the only determiner for advancement to the next level?

  4. How might we actually make this platform about language learning and not just another social media network?

  5. How might learners actually know they are improving?

  6. How might learners actually improve their level?

  7. How might we provide enough scaffolding to support learners but also still allow them to be self-directed in their learning?

  8. How might we engage users and provide a structure or platform that allows the content produced to be intrinsically motivating?

  9. How might we create a “community of learners” feel to the platform where learning is symmetrical rather than one-directional and unequal?

  10. How might we get expert feedback into the process to help validate learners’ experiences and attempts?

 

Iteration 3:

During this interaction, we narrowed down the scope of our project and started exploring the design in concrete terms.

Here are what we decided:

User

  • beginner

  • both native speaker (NS) and non-native speaker (NNS)

  • motivated/active learner

  • learners and helpers

 

Platform

  • Will be multilingual – focus on Spanish and English for symmetric learning and to incentivize native speakers to join

  • Will have the option for users to both “help others” and “learn”

  • learners can circle or annotate the parts of the image (similar to iannotate/skitch app)

  • allowing users to indicate by correctness and incorrectness of a response as well as by whether they like it or not

  • Help Others

    • will include items that need to translated

    • voting on “correct” answers

    • Language hangout – conversation exchange

  • Learn

    • take picture and post content

    • create flashcard

    • review flashcards

    • use others’ flashcards (like Quizlet)

    • play games with the flashcard content

    • read others’ posts

    • language hangout – conversation exchange

    • ability to post to a discussion forum or private message users if they have more general questions

 

Questions that haven’t been answered well yet(some of these were also from the previous meeting with Pattie)

  • How will learners know they are improving?

  • How will learners actually USE the language they are learning? – most of what we have now is just isolated input without any practice of it in a communicative context.

For example, if a user learns the phrase “give up”, how would they use it in a conversation?

Takako and I suggested having a language hangout where learners could have real-time talk with other learners.  Perhaps we could expand on this idea or adapt it. For example, should it be face to face chat, written chat, both? We should think about learners who might initially be shy about communicating face to face.

  • Could we somehow create conversations around interests? Even for beginners?

  • Can we actually test this using a prototype of some sort, similar to the facebook prototype from one of the groups?

  • How might we provide enough scaffolding to support learners but also still allow them to be self-directed in their learning?

  • How might we engage users and provide a structure or platform that allows the content produced to be intrinsically motivating?

  • In what ways do we want to use gamification?

 

Iteration 4

Deliverables and the Workflow for the class presentation:

1. Profile/Login Page/User input

  1. Language to learn

  2. Native language

  3. Country of Origin

  4. Self-determined level

  5. Interests?

2. Photo Capture

3. User Input from learner

a. Text label in native language

b. Voice label

c. Tag

d. Phrases

4. User input from helper

a. Text label in native language

b. Voice label

c. Tag

d. Phrases

5. Voting/Correction

6. Practice

a. Flashcard review

b. cover information review

c. games

d. Conversation exchange

7. Dashboard

a. Review process (what needs to be labelled?)

b. What needs to be commented/corrected?

c. Your progress

d. Categories/Interests?

  • Two users (both languages)

  • 1st time login workflow

  • 5th time login workflow

  • 100th time login workflow

Final Design Mockups

Profile


Learning

Teaching

 

Practice

 

Dashboard

 

Learning happens best when it is a byproduct of having fun

Activities

  • Starting a new business
  • Launching a new product
  • Creating robots for fun

Environments

  • Workplace
  • Home
  • Lab

Interactions

  • Market feedback
  • Teams
  • Purchasing parts from a hobby store

Objects

  • Raspberry Pi
  • Laptop
  • Lego

Users

  • Children
  • Robotics Experts
  • Employees

Interpretations:

  • People learning by doing
  • Learning happens everywhere, and is a lifelong activity
  • Each interaction is an opportunity for learning

Insight: Learning happens best when it is a byproduct of having fun i.e. when it is not the goal in itself, but happens accidentally in the process of achieving another goal.