# SL+ specifications
###### tags: `sl+` `dsl+` te`st`and `a21` `km` `v4v`

## Project Iterations

| Reference | Title | Iteration 1 (due end Jul.20) | Iteration 2 (due end Sep.20) | Iteration 3 (due end Nov.20) | Iteration 4 (due end Dec.20) | 2021 ---> |
| ---------------------- | --------------------- | ---------------------------- | ------------------------------ | ---------------------------- | ---------------------------- | ----------------------------- |
| **Curation (Squidex)** | | | | | | |
| 1.1 | Sets | dashboard only? | All data input (dsl+,st) | | | |
| 1.2 | Questions | selected good examples | Create all instances (dsl+,st) | Mastery Probe schemas & data | ...continued | |
| 1.3 | Picture Books | configure schema and content | | Create a proper Schema | | |
| **Monitor / Admin** | | | | | | |
| 2.1 | Users/Groups/Projects | basic association | Users into local Groups | | | |
| 2.2 (sl+,dsl+,st) | Activity Monitor | progress | | | Device status | |
| 2.3 (sl+,dsl+,st) | Participants | collect data, store | | Create for a User manually | | Add to another User |
| 2.4 (st) | Review | | Overlay in-app | | | |
| 2.5 | Mastery Probe | | | Mastery Review, data LRS | ...continued | |
| 2.6 | Sequence Control | | | | | Configurable Sequences |
| **App** | | | | | | |
| 3.0.1 | Web App | Web App | | | | |
| 3.0.2 | Mobile App | Mobile App | | | | |
| 3.0.3 | Canvas LTI | | | | | Canvas LTI |
| 3.1 | Login | FEIDE | | IDPorten, UI, User list, PIN | | |
| 3.1.1 | Consent | | | | | New participants from Consent |
| 3.2 | Organisation | dashboard only? | All nav / level layouts | UI | | |
| 3.2.1 | Avatar Creator | | | UI - edit name and avatar | | |
| 3.2.2 | Group Selector | | | | | UI - multi-select |
| 3.2.3 | Week List | | | UI | | |
| 3.2.4 | Day List | | | UI | | |
| 3.2.5 | Task List | | | UI | | |
| 3.2.6 | Include Participants | | | | | Teacher adds Participants |
| 3.3 | Question Sequencer | hard-coded | Sequence State Machine | Mastery sequencing | ...continued | |
| 3.4 | Question Types | see A.4. | 24 types (dsl+), (ST) #? | | | |
| 3.5 | Picture Books | re-code from old app | | UI, narrative assistance | | |
| A.1 | Audio & Video | | Record Audio / Video locally | | | |
| A.2.1 | Answers & Tracking | basic framework | Answers & Tracking -> Server | Answers & Trackings to LRS | | |
| A.2.2 | Progress Tracking | | Sync with server | | | |
| A.3 | Set Templates | Dash (st) | Week, Days, Tasks (dsl+) | | | |
| A.3.1 | | | | | | |
| A.4 | Question Templates | selected examples | 24 Question Templates (dsl+) | Mastery Probe Templates | ...continued | |
| A.5 | Server API | (from VIVA) | | | | |
| A.6 | Device API | (from VIVA) | | | | |
| A.7 | Animations | | Picture books (dsl+) | Animation book overlay | | |
| A.8 | Question Preview | | Preview | | | |
| A.9 | Caching Data | | | | Local media cache | |
| **Server** | | | | | | |
| S.1 | API | essential routes, GraphQL | API | Create Participants | | |
| S.2 | Database | essential models | Database | Set up LRS | | |
| S.3 | TSD Connector | | TSD Connector | TSD & LRS connections | | |
| **CMS** | | | | | | |
| C.1 | CMS Itself | file storage | | | | |
| C.2 | Set Data | schema, graphQL, API | Question Type Data | | | |
| C.3 | Question Data | schema, graphQL, API | Set Type Data | | | |

## Previous exmaples

**DSLPlus**
http://intermedia-prod02.uio.no/ispapp/part1
Use any code to connect, note there are extra buttons here to allow 'cheating' 😊
Some picture book pages include delays to allow storytelling before an animation is available, which prevents moving to the next page (look for the pointing hand)
Apple App Store: https://apps.apple.com/no/app/dslplus-i/id1453384808
Login: "apple": Passord: "app1etesting"

**Kaptein Morf**
https://engagelab.uio.no/morfologi
Click the padlock icon. login: '0000ZZ' password: 'demo'
May be helpful to use browser Dev Tools to set window to iPad size
Animation in KM was made with Adobe Animate
Apple App Store: https://apps.apple.com/us/app/id1316085218

## teSTand - Specifics

* (as SLPlus) To run on tablets or larger screens
* Tests are run with examiner and child (1-on-1)

## Q&A

#### LOGIN
- [x] Best alternative for non-educational login?
* FEIDE not commonly used in kindergarten
* IDPorten (BankID)
- Favoured second option!
* Persistent login
- Can we prevent users sharing their account with others...?

#### GROUPS
- [ ] How do Groups work?
* kindergarten is an overall group
* children can be in groups of 1+
* Groups don't always consist of the same people

- [ ] How do we record progress and answers in groups?
* Suggestion: Log against each Participant, and include IDs of other members in log
* Suggestion: Next task unlocks if at least one member of the group has completed current task

- [ ] How does the same User perform with a 2nd or 3rd Group of Participants?

#### PARTICIPANTS
- [ ] How to reliably connect a Participant (student) to a second User (adult)?
* PIN number?
* Just allow teacher / parent to correspond using ID
* Teacher and parent will know the child's identity anyway

- [ ] How can we transfer a Participant's work from one device to another?
* this is difficult, especially with possibly Red data

- [ ] Who creates Participants? Teacher/Parent or Researcher?

#### MASTERY
* mastery probes will be individual (not group)
* these are like 'task types' but in a new category

- [x] Can we affect the screen's sensitivity to touch?
* preferrence is a 'confim choice' dialog

- [ ] How should random items be selected for display?

- [x] What kinds of statistical / progress information should be available?
* mastery test results
* ability to control progress based on data

## 1. Content Editor / Curation

### 1.1 Curate Layout, Sets

(Layout)
* The structure of a new intervention is configured in a 'Layout' schema. This determines the names of other Schemas to retrieve (the Sets) for presentation
* Sets are configured with 'names' that present appropriate terminology to its users

(Navigation levels)
* An intervention consists of one or more Sets
* Creation of 'Sets' allows hierarchical organisation
* Add, Remove or Edit Sets here
* A Set contains a description
* Includes a section for tabular-formatted information `adapt21`
* A Set can be a container holding one or more Questions
* A Set can hold a list of another Set ('sub-Set')
* A Set cannot hold both Questions and a sub-Set
* Controls. Ability to control whether data is collected for the Set
* Sets can be given custom additional fields e.g. location, password, type, shuffle etc.
* A Set can contain an audio recording
* Questions can be presented by 'order',
or as decided by its corresponding UI Template (-> _A.4_)
* Custom fields can be allocated custom lists as select options `i2`
* Scenes are selectable from a preset collection `i3`
* Scenes can be seen as the 'story books' from DSL+, or the 'animations' in Kaptein Morf
* Scanes can be interactive

### Notes

* The lowest Set (that holds Questions) is referred to here as the **Question Set**

Set {
   title
   setName
   name
   description
      tables (data curated & presented in tabular form)
   control {
       tracking <- Tracking enabled (audio/video recording, events)
       cycleOn [question | set]  <- One Question per screen (e.g. DSLPlus) OR One Set per screen (e.g. ADAPT21)
   }
   scenes {   <-  Low Priority
       in: []
       out: []
       random: []
   }
   ...custom fields... (CMS-standard types, flat)
   
   holds [question | set]    <- If this is a Set holding Questions, it is the lowest Set level

   [if holds == questions]
     type [warmup | test | ?]
     ...custom fields...
     questions {}

   [else if holds == subSet]
     subSet [Empty | task ? Empty : Set]
}

---
**Further work needed:
Can CMS users also define the 'select' options or are they pre-defined?**

---
### 1.2 Curate Sets with Questions

(Sequenced challenges)
* A Question is the smallest unit
* There can be multiple Questions in a Set
* New Question types are created by EngageLab
* A Question will contain a data structure that models its particular requirements e.g:
* Question text
* Substructure e.g. categories of multiple choice questions
* Media (image, sound, video)
* A special Question type can represent a replica of another Question by reference `i3`

Question {
    reference (unique) (use id?)
    type  <- Which template to use
   ...custom fields... (based on Question design - all data types)
}

### Notes

* Different Question types can be created to suit all projects

### 1.3 Curate Picture Books
`dsl+`
DSLPlus requires picture books that present 'overlay' animations that can also be configured in the CMS
These need also to bre specially presented inside the app.

* Squidex: Convert Picture book schema from JSON to a more usable structure

## 2. Trial Editor / RCT Administration

Section 2 requires an 'Administration' user interface running outside the CMS

### 2.1 Assign Users, Participants and Sets (interventions)

* Begin with FEIDE2 (OAuth) connection option
* Apply for IDPorten
* Connecting users:
* Assign Users to as Set (intervention) inside it
* Assign 1 Participant to a User to begin with
* Allow also for 'Groups' of participants, later
* Do users connect themselves to a project?
* PIN codes to join?
* Or do Admins connect users to their project?
* Admins must be able to find user details
* Create Groups to contain Users
* Groups will be assigned to interventions
* (Consent for a user to be checked before beginning) `i3`
* (Use Dataporten to obtain school groups) `i3`

User {
    userID
    role [user | monitor | admin]
    name
    groups {
        ownedGroups [ groupId ]
        myGroup
    }
    participants []
    tokens {}
    consent {}
}

Participant {
    participantID
    data {}  <--- Data about a particpant allowed to be stored based on consent
    progress {
        setId: {
            lastActive
            ...custom fields... (e.g. surveyData, introSeen)
            completions: {
                taskId: [ questionId ]
            }
        }
    }
}

### 2.2 Activity Monitor

* Authorised access only based on User 'role'
* Give some Users rights to access progress details etc.
* Possible to unlock further Sets in an intervention for a User
* Possible to see a User's general results as feedback to the Teacher / Researcher
- number of repetitions attempted
- time spent (dosage) on each task and on entire session
- correct vs incorrect answers
* View progress in regard to intervention layout (-> *1.2. 1.3*)
* Possible to see and adjust the User's progress

### 2.3 Participants

CMS
* Each field can be marked 'private'
* Private data is not sent back to server

`st` Front End
* Add a participant to record test data against
* Approved admins can modify how many children (Participants) are listed on the User's profile
* Template determines which data types are collected (name, school, fellesnummer, gender etc.)

`dsl+` Front End
* Select a participant to work on, based on IDs generated at admin-side on the user's profile
* Confim the selection
* Can also select 'self'
* Rename a subject stored at front end only - names not sent to server

### 2.4 Review

In-app overlay that allows an observer to review Participant results

`st`
* Statistics are accumulated over time
* An 'Administrator' can see all data (on tablet)
* Can review answers
* Change / Correct answers
* Task answers can be reviewed
* A Task should be able to highlight the given answer in this mode
* Task answers can be corrected, and 'marked' as in/correct
* Note-taking box

### 2.5 Mastery Probes

Front End
* Retrieve Mastery data from LRS
* Control progress ☕︎☕︎☕︎
* See an evaluation of Tracking data
* Activate or block entry to next week
* Mark Participant as 'controlled' or not
* General statistical / status data (device status etc.)

Stage 1) Baseline

STUDENT        Week 0: Mastery[Tasks]
               ^                    |
RESEARCHER     '-----unbalanced-----'----approved ----> Stage 2
                Evaluate Basleine results (from Mastery tasks)

 
Stage 2) Intervention: Weekly Probes

STUDENT        Week W: Picture Book + Days[Tasks] + Mastery[Tasks] 
               ^                                                 |
RESEARCHER     '--- W = W + 1 --- approved ---------------------- 
               '--- W = W, D=0 -- rejected ----------------------
                Evaluate Fidelity Data (from regular tasks) + Probe results (from Mastery tasks)

#### Monitor Mastery view UI

Data needed (from LRS or TSD)

    Tracking: {
        itemId (mastery question ID)
        type ('mastery', 'question', 'set' or 'book') (use to filter DB search)
        duration (fidelity)
        start (fidelity)
        data: {
            participantID (use to filter DB search)
            weekID
            weekIndex (for overall progress)
            weekWord (for easy reference)
            masteryMode (baseline or probe) (use to filter DB search)
            correct (fidelity & mastery)
            round (repetition number) (highest is latest??)
        }
    }

Participants[] listing
* Progress overview (total weeks). Data: Highest `Tracking.data.weekIndex` for all Participants
-> select a Participant

Participant view
* General details
* Sub-views: Baseline or Intervention
* Progress (each week): Quantity of each `Tracking.data.weekIndex` for selected Participant

Baseline view
* Stacked bar chart: Mastery Trackings: Correct, Incorrect
* Data: {Tracking[]}

Probe view
* Stacked bar chart: Correct, Incorrect
* Fidelity table:

1. Probe Training session. Only can do them once. If correct moves on.
- but more difficult when child is asked something, what their answer is.. therefore:
2. Allow teacher/parent to evaluate correct answer
- preferable that Teacher does the evaluation
- supply correct / incorrect buttons for teacher/parent on these task types

How to master?
* Need three consecutive days showing mastery of the tasks contained
* PROBES - test items
Vocab intervention, picture book , tasks 1-8, PROBE

Probe: Different types:
1. Show a picture: Instruction: what is this?
2. Show two pictures: Instruction: this is "A", that is ("?")
3. Show four pictures: Instruction: Choose correct image from four

### 2.6 Sequence Control

#### 2.6.1 Sets Sequenced in an Intervention

* Linear order
* or Randomised on first entry

#### 2.6.2 Questions Sequenced in a Set

---
**Further work needed: How shoud sequencing work? How to define with flexibility?**

Questions can be sequenced by output formula or linear

formula {
    a: {
        threshold: 0.7
    }
    b: {
        calculation: nextQuestion = currentQuestion => calc(currentQuestion, resultData)
    }
}

Question {
    sequencing {
        type: linear | formula | custom
        calculation: null | 'a' | { threshold: 0.65 }
        
    }
}

---

#### Sequence Details

Sequence is very important - we need to be flexible how this works for each project

##### `km` Sequence
1 Intervention ( 3 Episodes ( many Sessions ( many Tasks ( Warmup | Test) ) ) )
Linear progression
* Sessions in order
* Warmup Tasks in order
* Test Tasks in order

##### `sl+` `DSL+` Sequence
1 Intervention ( many Weeks ( 3 Days ( many Tasks | Consolidation Tasks ) ) )
Linear progression
* Weeks in order
* Days in order
* Tasks in order (can repeat)

##### `adapt21` Sequence

1 Intervention ( many Tasks ( many Units ( many Questions ) ) )
Linear progression
* Tasks in order
* Units in order
* Questions **any** order (can not repeat)

#### 'Pre-intervention condition' `sl+` `adapt21`
* An activity that must be completed to a satisfactory level before proceeding
* Satisfied by number of successful completions or number of correct items
* Draws from Question and Question Set completion markers

## 3. App

* Standardise each section as a component
* Always build mobile-first, responsive design

* Which platforms / devices?
* How to introduce styling and project-level customisation?

#### 3.0.1 Web App

* First version to function in a web browser

#### 3.0.2 Mobile App

* App released to App Store under UiO account (iOS and Android)
* Caching of media function `i3`

#### 3.0.3 Canvas LTI (Web App)

* Run within Canvas and use login from Canvas. `v4v`

### 3.1 Login

`dsl+`
* ID-Porten
- ID-Porten for parents & others outside education sector
- after first successful login, sae token, ask for a PIN add User to list of 'recent users'
- allow a recent user to log in with PIN (+ stored token)

### 3.1.1 Consent

* Check logged in user with Consent Portal
* Search for a valid consent for this user
* Create number of Participants matching those on consent form

### 3.2 Organisation / Dashboard

* Display basic user profile
* Display user progress
* Progress should be a standardised data structure, with customisable template (_User - 2.1_)
* Consist of a hierarchy of Sets: pages representing levels of organisation
* e.g. Dashboard `adapt21` or Weeks, Days, Tasks `dsl`, Ship, Map `KM`
* Display Set data (available or not)
* Descriptive HTML text, images, used in a layout customised to this project's style
* e.g. Alien on Earth `KM`
* Set should be standardised data structure, with customisable template (_Sets - 1.1_)
* Set data (_C.3_) is matched with Set Template (_A.3_)
* On load, check the current status for this user
* User profile
* Participants
* Set access
* Progress
-> Relies on (*A.3* Set Templates)

UI
* Transition motion between levels
* Basic
* ↓ Advanced (old and new pages together) ☕︎☕︎☕︎
* Profile + 'quiet' mode activation (see Right Panel)
* Progress activation (see Left Panel)

#### 3.2.1 Avatar creator
Participant creation / selection

* Select a new (unused) participant?
* Enter name
* ↓ Create profile ☕︎☕︎☕︎

#### 3.2.2 Group selector
Select participants taking part in a group
* Sliding grid of Participant avatars with checkboxes ☕︎☕︎
* Mastery status of participants?

#### 3.2.3 Week List
* Vertical responsive scrolling item layout ☕︎☕︎
* Left & Right hand side panels
* Components:
* Week cards
* Main image card
* Progress / details segment ☕︎☕︎
* Progress
* Mini-icon
* Status icon (reward / locked / inProgress / checked / done)
* Left Panel
* Progress bar
* Mini-icons
* Right Panel
* Name
* Actions (help, logout)
* Help (separate page completely?)
* Video?
* Settings

#### 3.2.4 Day List
* Sliding card layout ☕︎☕︎
* Components:
* Day cards (background, details)
* Progress
* Status icon (reward / locked / inProgress / checked / done)

#### 3.2.5 Task List
* Clockwise Responsive item layout ☕︎☕︎☕︎
* Components:
* Task Circle / Book icon
* Status icon (reward / locked / inProgress / checked / done)

#### 3.2.6 Include Participants
* Teacher can add participants to it's profile
* ? HOW IS THIS TO BE DONE ?

### Sequences (Question or Set)

`i1`
* Sequence is hard-coded in app for each project
* Progress stores 'completion' in the Participant's profile
* Completion of one step in the sequence produces a Tracking output
* Completion of one step in the sequence contributes to Set Progress (stored in Participant)
* Questions must produce compatible outputs used to measure completion success prior to next step

`i2`
* A sequencing function or 'state machine' is used to control Question Set completion, Result reporting, Progress update, possible animations, and transition to next Question Set (or finish)
* This machine can be supplied a sequence type assigned in the CMS (*2.3*)
* When each Set holding Questions is chosen, it will be ordered based on the Sequence (*2.3*), or possibly a non-linear progression (next Question set determined by result of the current one)
* Question may populate from a previous question's answer `adapt21`
* Task produces an Answer output that **matches LTI standard** (*!*)

-> Relies on (*2.3* Sequence Control & *2.5* Mastery Probes)

### 3.3.1 Mastery Sequences

---
#### The concept of Mastery - sequencing in DSL+
`dsl+`

**Baseline**
- Baseline(shown as week 0) is locked at the beginning
- A user has to finish Baseline before they can continue to the intervention
- Baseline contains a sequence of 4 expressive items, 4 receptive items and 4 indepht items.
- The items are randomly chosen from pool of predefined items
- In each item type, the items has to be from the first week and 2 other lessons.

**Probe**
- Probes are shown at the end of each day
- The probes will be checked by a researcher and approved if the answers are correct
- Probes contains a sequence of 4 expressive items, 4 receptive items and 4 indepht items.
- The items are randomly chosen from pool of predefined items
- In each item type, the items has to be from the current week and 1 other lessons.

Week X: Day X Interventions -> Probe at the end of the day complete -> unlocks Day X + 1, and at the same time lock Day X(previous Day inaccessible) -> repeats until everyday are completed
Week X+1(next week): Remains locked until Week X-1(previous week) Probes results are approved 

**How to achieve mastery?**
* Need to satisfy three consecutive days showing mastery of the tasks contained
* Need to satisfy Probe - test items
* Researcher unlocks the next week when the current week Probes are completed and approved

**Probe Types**
1. Expressive Types
- Show an image(for nouns) or 2 images(for adjectives and singular/plural) or 1 video(for verbs)
- Instruction: Adapted to each word class(An audio file).
- Examples for Instruction: What is this?, What is he doing?
- Content: Random items from that can be from mountain, planks, fast, friends and loud(max 12 items per lesson).
- Record: The teacher/parent registers the task as correct or incorrect
The ipad records audio for fidelity?

2. Receptive Types
3. Vocabulary indepht items

5. Show a picture: Instruction: what is this?
6. Show two pictures: Instruction: this is "A", that is ("?")
7. Show four pictures: Instruction: Choose correct image from four

**Sequence**
* This sequence determines when a child has achieved 'mastery' of the week's work

  ,------------------------------------- W = W + 1 ------------------- pass ----------------------------------------,
  ˅                                                                                                                 |
Baseline Probe W ---> Week W [Picture Book + Days [Tasks]] ---> Week Probe ---> Day Score Threshold + Probe Pass ---'
                      ^                                                                          |                 |
                      '------------------------------------------------------------------ fail --'  <------ fail --'

**Week 0 - Pre-Intervention Probe**
* Will be structured into days, same shape as the main intervention
* Requires a large set of probe items, each day will show perhaps 10
* Randomised items

-> This can be built as another 'week'

### 3.4 Questions
* Lowest Set contains one or more Question types
* One container of Questions is presented on screen at any one time (presentation based on template)
* The Questions are cycled if only one Question per screen
* Or the parent Set is cycled if multiple Questions per screen
* Questions data (_C.2_) is matched with Questions Template (_A.4_)
* A link inside the CMS Question creation area leads to a demonstration as it would appear to the client in the App

-> Relies on (*A.4* Question Templates)

#### Generic Format
The possibilities for presenting internal content of a Question are extremely variable, so Question data structure may take many forms. However there should be well-defined system compatible inputs and outputs for a Question, including 'state' type data that can be used to decide which Question / Set to present next, in possible cases of non-linear progression. This data may only be needed client-side. For example:

resultData {
  score: 0.7   <- Level achieved
}

questionSetData {
    output {
        threshold: 0.5 <- Minimum pass level
    }
}

-> The student is allowed to move to the next Question

resultData {
    evaluation {
      score: 0.7
      direction: 3
    }
    questionSet {
        completed: true
        questions {
            id: { completed: true }
        }
    }
}

questionSetData {
    input {
        path: ['QuestionA', 'QuestionB'] <- Path already taken
    }
    output {
        threshold: 0.5
        paths: ['QuestionC', 'QuestionD', 'QuestionE', 'QuestionF']    <- Next directions
    }
}

-> The student is allowed to move to the next question, and is sent to questionF

#### Review & Early Exit
If the user returns to a Question Set, it should be possible to
* Reload and display the previous answer
* If completed, prevent answers being changed (read only)
* If not, allow the user to continue answering

These mechanics will have to be taken care of inside the Question Set

#### Question Template ('boilerplate')
A carefully designed Question _template_ would be a useful way to structure functions needed by every question and speed the production of new Questions

### 3.5 Picture Books
`dsl+`
Re-write Angular code from previous app to run and present picture books including animation overlays.

Create a panel that displays narrative help
* Text to be read for the current page
* Checklist of things to do on this page

* Responsive layout ☕︎☕︎
* Sliding / Changing pages with swipe gesture ☕︎☕︎☕︎
* Components:
* 'Tappable overlay' awareness - how to display the trigger for interactions?
* Refine overlay components

### A.1 Record Audio & Video, Specialist Questions

* Reusable components should be build to enable recording of video and audio (web + mobile)
* Data should be saved to device storage (mobile)
* Audio or Video files should have a reference included in the Answer or Tracking data
* Aduio or Video should be transferred to secure storage as part of the Answer or Tracking submission
* These tools should enhance existing Cordova code established in the lab to reduce development time
* Other components may be needed (e.g. Drawing)

`st`
* Child (User) input is **oral only**
* Requires Audio recording across Task or entire Task set
* Allow possible override of recording in Review? (*4.3*)
* Up to 800 children will use the app
* Up to 20 tests
* Each test 20 - 50 items (10 seconds per item)

---
**Further work needed:**
* Define a collection of input components to support?
* teSTand Visualisations?

---

### A.2.1 Answers and Tracking Data

* Results need to be saved locally until able to be posted to the Server
* Results are separate from Progress
* Results data tells us how the user answered the questions
* Tracking is separate from Results!
* Tracking data tells us what actions the user performed on screen
* Data must be forwarded on to TSD or LRS, this is for the Server to decide absed on data type
* Answers & Tracking DATA to LRS
* Tracking FILES to TSD
* Results may contain data that is sensitive, remove local copy after successful post
* This has been a major point of contention in cases where data was not correctly posted. **Thorough testing is required**
* (Results data will need to be retrieved if the software is asked to review a previously completed Task)
* The 'review' feature may not be available if data is deemed sensitive

### A.2.2 Progress Tracking

* Each time a Question or Set is completed, not this as progress on the user's Model
* Progress should be able to recover the User's position in the project after reloading the app
* Progress data should be displayable visually

### A.3 Set Templates

* HTML views that can present a customised view of a Set level
* Area to display introduction page form CMS (HTML)

### A.4 Question Templates
* These are components each custom made to represent a specific Question type
* Data to populate the component (image, sound, text) is loaded from the CMS
* They may be simple or complex, and possibly include other components such as video player
* **Important not to use external libraries unless absolutely necessary**
* Questions should be designed responsive mobile-first
* _possible exceptions for mobile telephone_

`i1`
* Create a selection of good examples `sl+`

`i2`
* Recreate the 24 task types as they functioned previously (DSL+ Vanderbilt Uni version) `dsl+`

`i3`
* Create 'Mastery tasks' `dsl+`

___
**Define a reusable sub-components for this**
e.g. button types, flip cards, etc.
___

### A.5 Server API

* Secure connection
* Transfer login, user details, CMS data, results

### A.6 Device API

* App uses Cordova functions to attach to device for access to
* File storage
* Microphone
* Camera
* Some of this functionality can be taken from other EngageLab projects

### A.7 Animations
* Present the picture books as they were previously `dsl+`
* A Sequence can trigger 'intro' and 'outro' play scenes surrounding Tasks
* A Sequence can include the option to randomly display an animation or 'cut scene'
* An animation can be used as a 'character' or 'avatar' who speaks instructions and appears at times throughout the app
* Present animations in a module that can display as full screen, or as overlay / sub-screen

### A.8 Question Preview
* See a preview of a Question's appearence in the app directly from the CMS.
* Use this format in Squidex > Schemas > (schema) > More > Preview URLs:
`https://[app.host]/#/sample/slplus/dslplus/${schemaName}/${id}`

* slplus -- project name (determines project queried for data)
* dslplus -- project type (determines templates used to display data)
* ${schemaName} -- interpolated from schema data **do not change this**
* ${id} -- interpolated from schema data **do not change this**

### A.9 Local Data
* Allow login to app without internet access
* Save models to device (App, User, Participant, Answer, Tracking)

Cache:
* Use the Squidex API to get a listing of all assets
* Write a script that downloads all assets to the local build
* Ensure the app finds these assets locally instead of online, if available

## 4. Integration

### SERVER

### S.1 API

* Receives, processes and returns calls for data from
* CMS
* Trial Editor
* App
* Monitor
* FEIDE
* Consent Server
* TSD Storage
* LTI Server
* SL+ Database
---
**Further work needed: Define Routes**

---

### S.2 Database

Models:

User
Group
Intervention
Set
Task
Question
Result

### S.3 TSD Connector

* Send data to TSD using the API
* JSON structured files
* Data files (M4A audio, MP4 video)
* Video and audio goes to TSD, not to be stored on our server

### CMS

### C.1 Content Management System
* CMS will be open source
* Self hosted
* Suitable Query language (GraphQL?)
* Attach media with asset preview
* Localisation (multi-language) support
* Draft / Publish
* Versioning
* Webhooks
* Custom plugins / content editors
* Blocks as HTML?

Most likely candidate at this point is squidex.io/comparison

#### CMS Comparison (our needs)

                        Sanity           Strapi                   Squidex
    Rich Text                       
      - Front           ✓                with ShowdownJS(*1)      with CKE Editor
      - Back            ✓                markdown                 html
    Tables              ✓                with ShowdownJS(*1)      with CKE Editor
    Custom Plugins      ✓                ✓                        editors & integrations

(*1) ShowdownJS provides useful functions, but has 110 open issues.. https://github.com/showdownjs/showdown/issues

(*2) CKE Editor 5 can be added as an 'editor' to the CMS.
Editors are a HTML file that need to be hosted somewhere, see:
https://github.com/Squidex/squidex-samples/blob/master/editors/cke-simple.html

### C.2 Set Type Data

Stored data for each instance of each Question Type
* Based on the representative data structure created to model this
* Curated using forms provided by the CMS
* Create an appropriate GraphQL query to obtain this data
* **Input real data for all projects**

### C.3 Question Type Data

Stored data for each instance of each Set and Task
* Curated using forms provided by the CMS
* Create an appropriate GraphQL query to obtain this data
* **Input real data for all projects**

-----------------------------------
# DSLPlus - Specific Specifications

### C.2, C.3
* Toggle to turn off recording
* Toggle to randomise
* Getting old data into the new system?