V(isually)I(mpaired)nput
As defined in our Alpha release project plan, below are the deliverables we have completed:
- Documentation of Brainstorming and Sprint Divergence (Google Drive)
- Initial Brainstormed Workflow (Google Drive)
- Characteristics of Individuals Involved in Testing (see "Target Users" in Usability Testing in Google Drive)
- Usability Testing Procedure and Results (Google Drive)
- Feature Discovery (Google Drive)
- "Hello World" full screen keyboard (GitHub)
As defined in our Beta release project plan, below are the deliverables we have completed. Additional and more detailed information can be found here.
- Algorithm logic
- The
Valuesinterface and subclasses primarily handle this logic - Swiping/gesture setup
- Gestures registered in
KeyboardViewController.swiftinclude: swipe left, swipe right, swipe up, swipe down, hold, two finger swipe right, three finger swipe right, double tap, two finger hold, two finger tap, and pinch - Gesture processing in
KeyboardViewController.swiftand primarily different mode files (i.e.InputMode.swift) - Layout of keyboard
- Different UIView elements make up the visual interface within
KeyboardViewController.swift - Controller development and input processing
KeyboardViewController.swifthandles this in conjunction with different mode files (i.e.InputMode.swift)- Voice prompting
- Pervasive use throughout the project of AVFoundation's
AVSpeechSynthesizer - Keyboard activation
- Initialization and setup upon triggered activation is found within
viewDidLoad()ofKeyboardViewController.swift - "Happy path" testing: see TESTING.md
We also went beyond these specified items in our project plan in completing the following:
- VoiceOver Integration: UI setup to switch between VoiceOver and VInput gestures and speech announcements
- Tutorial and Training Modes: informative and interactive guide and steps to using VInput
- Multiple Alphabets: users can two finger swipe between lowercase, uppercase, basic emoji and numeric alphabets
- Fault Tolerance: resiliency and correction against crashes, errors, faults by reloading in memory the last word where the user left off (allowing the user to continue where they left off)
- Ensuring Correct Input: correctly places the input cursor at the rightmost location in the text field to prevent accidental deletion and correct appending of additional characters
- CoreData Implementation: as a user types right now, the words they type are stored on-device for later use in developing prediction features
- Code Quality: we spent significant time in the design of our code such that it is readable, maintainable, and extendable (i.e. inheritance in different input modes and alphabet values)