Skip to main content

Question Library Roadmap for Evalium

This document outlines the planned features and functions of the Question Library in Evalium, from the MVP baseline through phased improvements. Each feature set includes the rationale for why these decisions will deliver maximum value to SMB admins and power users.

MVP (Weeks 1–4): Core Productivity​

features​

  1. Inline Question Editor with Live Preview

    • Function: Expand a row or card to reveal an embedded form for creating or editing MCQs, complete with real-time rendering of stem, choices, attachments, and metadata.
    • Why: Eliminates context-switching between list and edit pages, accelerating authoring flow and reducing errors.

  2. Basic Filtering & Sorting

    • Function: Full-text search, tag filtering, and date-sorting on the question list. Users can bookmark (“save”) filters for one-click reuse.
    • Why: Even a simple library can grow quickly; basic faceting ensures admins can find and manage questions without manual scanning.

  3. Core Bulk Operations

    • Function: Shift/Ctrl+click multi-select, then apply bulk Tag, Delete, or Clone actions. A snackbar provides an immediate Undo option.
    • Why: SMBs often import legacy content or need rapid curation. Bulk ops eliminate repetitive clicks and support rapid library cleanup.

Phase 2 (Months 2–3): Flexibility & Scale​

features​

  1. Multi‑View Library (Table / Gallery / Kanban)

    • Function: Toggle between a data grid, a visual card gallery (with metadata badges), and a Kanban board grouped by review status or folder/collection.
    • Why: Different admins prefer different mental models—list for precision, gallery for visual browsing, Kanban for workflow management.

  2. Advanced Bulk Controls

    • Function: “Select all matching” after applying filters; contextual action bar surfaces only relevant bulk tasks; multi-level Undo messaging (e.g., “12 questions archived – Undo”).
    • Why: As libraries grow, admins must perform complex operations across hundreds of items; advanced bulk controls maintain speed at scale.

  3. Saved Views & Question Templates

    • Function: Define and name common filter combinations (e.g., “Medium-Difficulty JavaScript”) and create reusable question templates for patterns like True/False or Likert.
    • Why: Saves time setting up views and reduces manual re-entry of common question formats.

Phase 3 (Months 4–5): Power‑User & Keyboard‑First​

features​

  1. Global Command Palette

    • Function: Invoke via Ctrl/Cmd+K to fuzzy-search actions (Create MCQ, Switch View, Bulk-Tag) or jump directly to any question.
    • Why: Empowers advanced admins to complete tasks without mouse, boosting efficiency for high-volume authors.

  2. Slash‑Commands & Custom Shortcuts

    • Function: Within inline editor, type /tag, /image, or /template for instant insertion; define shortcuts like E for edit, C for comment; numeric keys to switch views.
    • Why: Minimizes friction, lowers keystroke count, and caters to keyboard-centric workflows.

  3. Full‑Screen Question Builder

    • Function: One-click expand to a dedicated, shareable URL for in-depth editing (long stimuli, media, complex formatting).
    • Why: Supports edge cases that don’t fit the inline form, without bloating the primary UI for the 80% use-case.

Phase 4 (Months 6+): Collaboration & Governance​

features​

  1. In‑Context Commenting & Review Threads

    • Function: Highlight any part of a question to add threaded comments, @-mention teammates, and resolve or reassign review items.
    • Why: Facilitates SME feedback and compliance checks directly in the authoring flow, reducing reliance on external communication channels.

  2. Versioning & Branching

    • Function: Automatic version history per question with diff views; create draft branches (e.g., language variants, difficulty variants) and merge changes.
    • Why: Provides auditability, supports translation workflows, and enables safe experimentation.

  3. Collections & Folder‑Like Organization

    • Function: Create named Collections (virtual folders) by saving any combination of filters or dragging questions into them; display alongside Evaluations in the sidebar.
    • Why: Balances the flexibility of faceted search with the familiarity of folder structures, without rigid hierarchy limitations.

  4. Item‑Level Analytics & Quality Flags

    • Function: Surface basic psychometric metrics (difficulty index, discrimination index, pass rates) alongside questions; automatically flag outliers (too easy/hard, low discrimination).
    • Why: Empowers data-driven improvements to question pools, leading to fairer and more reliable assessments.

Rationale Behind the Phased Approach​

  1. Rapid Value in MVP: By first addressing the biggest pain points—context-switching and manual grunt work—we deliver immediate productivity gains without complex backend work.
  2. Building on a Stable Foundation: Phases 2–3 layer in flexibility (views, templates) and power-user tooling (keyboard shortcuts, command palette) atop the same data model and UI components, maximizing reuse.
  3. Scaling for Teams & Compliance: Phase 4 adds collaboration, governance, and insights, which require additional infrastructure (comment storage, version control, analytics pipelines) but unlock true enterprise-grade capabilities.
  4. Low Risk, High Reward: Each phase is independently valuable—admins benefit at every step, and the architecture remains modular and maintainable throughout.