Connected Devices Weekly Update/2016-07-19

Vision

• Vaani aims to bring a voice to the Internet of Things (IoT) using open, Mozilla-backed technologies.

Assumptions

• People want to use voice to create lists
• People need a new solution to making shopping lists
• People want a device in their kitchen to interact with
• People would like their shopping list shared with others
• People would like their shopping list on their phone
• People need a solution for remembering items while they are near the store

MVP being tested this week

• People are interested in being reminded of tasks while they are near that store
  • We expect 1% response to a Facebook advertisement https://geolocalshopping.wordpress.com/
• Multiple parts on a survey
  • Do people want to use voice to shop
  • Are brand names important?
  • Do people need quantity with their order?

Initial Interesting Results

• Ranked in order of importance to a shopping list
  • Quantity
  • Brand Names
  • Store Names
  • Tasks
  • Other

Milestones

• Vaani internal testing - July 1 - July 15 - Done
• Nascent Objects Shipping to Berlin - ETA July 23 - On Target
• Legal review - July 13th - Delayed
• Find Users to Test With + Sign Legal Doc for Data Collection - July 15th - Delayed
  • Delays to Legal are known risks of adding external testers and are acceptable within the project relm

An always-on voice-enabled smart screen device to keep track of the activities, events, and to-do lists of family members. Initial focus for the prototype is on Reminders only.

Vision

• Increase the amount of stress free, uninterrupted quality time that families spend with each other

Assumptions

• Busy families are looking for an alternative solution other than what they currently use (memory, post-its, black boards, calendar apps etc.) to manage & organize their schedule
• Busy families are looking for an alternative solution that is inclusive of the entire family's’ needs and not just for an individual responsible for coordinating & maintaining the family schedule
• Families are looking for a solution that can serve as a hands free alternative when in need
• VOICE is a friction free way to create reminders, lists, coordinate schedules.
• Busy families would require an alternative way to create & manage reminders, lists & schedules using a SMART DISPLAY that would further simplify their management.
• Busy families would require access to these sets of functionalities whether they are at home & away
• Busy families would require access to the functionality as they are moving about in their homes

Hypothesis

• By creating a voice enabled solution to create reminders, lists, coordinate schedules, we will reduce the time taken to perform these tasks by 20%.
• By offering a solution that provides an alternative way to create and manage reminders, lists & schedules using a smart display, families would be able to accomplish their tasks 20% more times than before.

Prototype MVP

• Voice enabled solution to create & access reminders
• Smart Display to view & delete reminders
• Voice & Visual feedback
• Notifications to multiple devices
• Remote access to retrieve/create reminders

Study

• Focus will be on reminders, lists, and event and activity coordination of family members and those outside the family
• Study will include questions, exercises on reminders and design, and interaction with a prototype
• Recruitment
  • Preferably select 5-7 of the same families of friends and neighbors interviewed for the initial study given that they have the context of the study.
• Goals
  • Get a deeper insight on what types of reminders, lists, schedules people are looking to manage
  • Get an understanding of the speech patterns: Determine 2-3 speech patterns (different ways of instructing the robot) to set reminders, to-do lists, schedule events both within the family & outside the families
  • Gather feedback on the voice interaction to create reminders
  • Gain a better insight into other ways that these families foresee interacting with this device
  • Walk through and get a high level feedback on the paper prototype. Or might hold a separate session with recruited users for determining the right interface.

• Passed Gate 0 on June 1, 2016.
• Engagement with an external creative agency for an 8 week program started on July 11, 2016 to identify user needs and propose product solutions.
• Project scope
  • Screens as core touch point to be explored
  • Centered in the home
  • Look to deliver on human value beyond the utilitarian
• This week
  • Planning the primary research and begin recruiting on that front
  • Planning the Subject Matter Experts (SME) interviews and reaching out to subject matter experts
  • Secondary research that will broadly look at the current state of the industry, key players, technology and related trends

• Last Friday (July 15th) we presented in the Checkin #5 (slides and video recording) the insights from the User Research #2 and the decisions made about what we'll be doing next.
• We didn't validate our primary hypotheses "girls would want ambient communication that did not distract them from other things"
• We neither proved nor disproved our secondary hypothesis "this device would improve communications with their friends." But saw some evidence in their desire to know about the status and availability of close friends

Decisions

• Focus more narrowly on improving communications between close friends by helping girls know the status and availability of their close friends.
  • Same target audience (12-15 year old girls)
  • Uncover other “desires to know information about my friends” that could be supported by an ambient device
  • Focus on small group of friends

This week we're taking some time to design our next experiment using some of the tools we learned in Lean Startup training last week.

• Make the pieces we are testing smaller, with the purpose of learning faster
• Meanwhile, we are exploring the technology options for always-on/ubiquitous and smaller/wearable devices and different interaction models.
  • Bluetooth LE modules (nRF51822), prototyping BLE wearable
  • Simple status/availability web service
  • Smartphone app to connect to BLE device & client for status web service (Cordova, Bluetooth Web API?)
  • Secure logging/telemetry from embedded code

• What's done
  • In the previous sprint, SensorWeb team had pretty much done the preparation for the 1st experiment (https://trello.com/c/fdvHl50c).
  • The 8th sprint report https://goo.gl/j4lZB6
• What's next
  • Follow up measurement, analytics on the poster experiment (target done July29)

(read only)

• 2 user interviews carried out so far
• Delay in shipping prototype to Canada and a technical problem has delayed the other user interviews
• Update to the board to be shifted by two weeks

• Leap of Faith Assumptions (LOFA) we are addressing:
  • We assume that people want recipe ideas that they can make with their existing inventory
  • We assume that people want recipes based off the inventory of their fridge
  • We are assuming that image recogniton can accurately recognize a user's inventory
• Hypotheses
  • If we create a Facebook ad for Smart Kitchen, then we will get a 2% click rate
  • If we create a "kickstarter type" type video, then we will get 4% likes (4 likes every 100 views)
  • If we give a user an image recognition device for identifying their inventory, then we will be able to recognize 20% of items with our existing training to date.
• MVPs for Experiments
  • Facebook ad
  • Video
  • Image recognition Device
• Status Update on Implementation of Experiments
  • Created Smart Kitchen Landing Page
  • Creating 2-5 prototypes of the Smart Kitchen image recognition device
  • Working with Research to make sure our language is correct for the Facebook ad
• Google spreadsheet of our process and experiment details are here
• Trial agreement for Image recognition device ready today.

Broadcasting helpful information (e.g. schedule, maps) to conference attendees will help alleviate their anxiety and Broadcasting personal tidbits to other conference attendees will make networking easier User Study

• We already have the User Findings from the experiments in MozLondon: User Study #2 Findings and Themes and Data Analysis. The key learnings from the User Study were:
  • More barriers to people interacting with and broadcasting their own content than expected.
  • People didn’t stay engaged with the app - they forgot about it or it was too overwhelming.
  • People seemed more interested in locating people they knew than meeting new people.
  • Running many experiments at the same time made it harder to uncover insights.
• Regarding to the hypotheses we were not able to validate them or invalidate them
  • Although attendees found useful info (as sched) we think that it could have been more useful if it were more widely implemented in all the hotels, besides, we don't think we lowered anyone's anxiety.
  • Regarding the second hypothesis, we did not create any network connections (in general), the barriers play a huge part for those that were broadcasting that combined with the fact that it was only the cd group and we already know almost everyone.
• We feel that continuing to focus on conferences/events, and being able to control the “ecosystem”, we will obtain clearer insights that can then be used to improve that platform (View source and Mozfest)

Development

• The team has just landed basic "notifications" feature in iOS and Android to engage users with the app.
• This week we will continue Improving how notifications work (e.g. by keeping history of beacons discovered, so users are not notified about them), testing and bug fixing of this feature and some visual updates.

Other

• Working together with Legal and Marketing teams to see the possible options to upload the app to Apple Store and Google Play to reduce the installation barrier
• Making a plan with Viewsource team to prepare the experiments in that event
• Logistics: checking and testing different beacons and providers to get reliable and cheaper beacons
• Last check in last Friday (July 15th): slides and video recording

• Setting up a template for the technical research to be done as part of the CD strategy
  • This looks into the technical aspects of other companies to see market opportunities
  • This initiative will be started with the team early next week. The template is currently being reviewed and revised

• New Open Innovation Outreach Team
  • Michael, Rina, Natalia, Rosana, Katharina
  • Met last week for initial planning
  • Scope and priorities still being figured out
  • Will have much overlap with CD participation
• Dates set for initial Vaani hack events in Brazil (Rio)
• Mozilla IoT class at University of Jordan
  • Syllabus submitted, waiting on University approval
• IoT HackHyderabad, approved, budgeted, ready to go
  • https://reps.mozilla.org/e/iot-hackday-hyderabad-1/