A web application to share your voice on a topic.

User Stories we want to address with this application

User Hammad

Hammad is an employee of a large organization

Use Case 1

Hammad wants to be able to submit his concerns to leaders in his organization with confidence that they will be read and addressed.

Use Case 2

Hammad wants his concerns to be visible to all members of this organization. If some one shares his concern, he wants them to be able to +1 him. He feels this will make his concern be taken more seriously.

Use Case 3

Hammad thinks that leaders of his organization can dodge hard questions using techniques they learned in media training. He wants a real answer.

User Emilie

Emilie is a leader of a large organization

Use Case 1

Emilie wants to hear the concerns of all employees so she can address them timely. She wants to do that in an inclusive but time efficient manner.

Use Case 2

Emilie wants to know what concerns are more prevalent so she can prioritize them.

Use Case 3

When Emilie addresses a concern, she wants it to reach everyone. She find that often she has to address the same concern multiple times.

Tech Notes

This is where I dump my thoughts on tech implementation.

Authentication

I want make it extremely easy to integrate this app with an arbitrary authentication system. There are so many popular authentication standards out there; it impossible to support all or choose some. So to meet my goal of easy integration I have decided to keep authentication completely out of this app.

This app should be deployed behind a proxy that handles authentication and just passes the username of the authenticated user to this app. The authenticated username is passed as a header (default name 'X-Forwarded-User').

This app only accepts requests from localhost. This will ensure that no external user can bypass the proxy and connect directly to the app with a fake header.

Vocabulary

Topic: Conversation subject. Voice: Think of it as a comment.

creator: Username of the creator of a topic slug: a url friendly identifier for a topic. Together, creator and slug uniquely identify any topic.

DB Scheme

Table: Topics
	creator string | partition key
	slug string | sort key
	title string
	timestamp string
	lastupdated string

Table: Voices
	topicIdentifier string | partition key   (topicIdentifier = creator + slug)
	voiceItentidier string | sort key  (voiceItentidier = speaker's username + timestamp)
	
	timestamp string 
	username  string 
	text string
	upvotedBy arrayOfString | username
	heardBy arrayOfString | username
	responseVoiceIdentifiers arrayOfStrings | {voiceItentidier}
	type string | primary or response
	archived bool 

URL Scheme

/						homepage explaining the tool + list top N topics by lastupdated + create new button
/creator 				list all topics of user by last updated + create new topic
/creator/slug    		list all voices by creation date + add voice

Api

Create Topic  | creator + slug
Get Topic  |  creator + slug

Add Voice | topicIdentifier + speaker's username
Delete Voice  | topicIdentifier + voiceIdentifier

Mark Read  | topicIdentifier + voiceIdentifier + username

Add up-vote  |  topicIdentifier + voiceIdentifier + username
Remove up-vote  |  topicIdentifier + voiceIdentifier + username

Add response voice  topicIdentifier + voiceIdentifier + speaker's username
Remove response voice  topicIdentifier + voiceIdentifier

Task list

• Setup Node/React env

• Create database tables

• Integrate with a material-ui module. (choose http://www.material-ui.com/)

• Setup redux

• Basic topic feature

  • add topic database controller
  • get topic database controller
  • get topics database controller
  • add topic react communicator
  • get topic react communicator
  • / page. See topics
  • / page. Add topic modal
  • / page.
  • /creator page. See topics

• Basic Voices feature

  • add voice database controller

  • get voices database controller

  • add voice react communicator

  • get voices react communicator

  • /creator/slug page. See topic info

  • /creator/slug page. See voices

  • /creator/slug page. Add voice

  • support pagination on getVoices. dynamodb paginates results at 1MB i.e about 4k to 8k voices

  • Delete Voice feature

  • delete voice database controller

  • delete voice react communicator

  • delete voice UI

• Upvote Voices feature

  • upvote database controller
  • upvote react communicator
  • upvote UI

• Voice Heard Feature

  • mark as heard database controller
  • mark as heard react communicator
  • heard UI
  • heard auto detector based on scroll

• Auto username feature

  • openID integration so we can get username automatically. (See Tech Notes)
  • readonly support for those without ID

• Response voice feature

  • add response voice database controller
  • add response voice react communicator
  • add response voice UI

• auto load changes made by other users

  • add voice
  • delete voice
  • add response voice
  • delete response voice
  • edit topic ?

• Create react database communicator to talk to node database controller

• Sorting Voice feature

  • do we need to do something on the backend for optimization?
  • sorting UI

DynamoDB Setup

Edit dynamoDB configurations in '.env' file so they points to a valid DB.

NOTE: For a better development experience you can setup a local dynamoDB instance. See: https://aws.amazon.com/blogs/aws/dynamodb-local-for-desktop-development/

Once the '.env' file points to a valid dynamoDB, create required tables and initialize table values:

node ./server/databaseHandlers/initializer

Dev Setup

Prerequisite: DynamoDB Setup. See section above.

npm install
npm run start