Skip to main content

​
Overview

  • Project name: Corpora
  • Type: Full-stack application with dual-stack architecture
  • Status: MVP (Minimum Viable Product)
  • Primary goal: Provide a secure local development environment and admin interface for managing scholarly text corpora — manuscripts, biblical texts, Quranic texts, lexical data, and more.

​
Purpose and vision

  • Core mission: Create an accessible platform for scholars, researchers, and language enthusiasts to work with, analyze, and manage large text corpora in a secure, offline-first sandbox environment.
  • Value proposition: Combines local-first development infrastructure with a native macOS management interface for seamless corpus curation and research.
  • Key philosophy: A safe, isolated sandbox for experimenting with text data without affecting production systems.

​
Architecture

​
Stack 1: TypeScript/Deno local development sandbox

A local, safe development environment mirroring Supabase functionality — database storage, authentication, and edge functions mapped locally. Responsibilities:
  • Abstract Supabase interactions
  • Provide a local authentication sandbox
  • Enable offline edge function development and testing
  • Serve as a development proxy for database operations
The MVP phase focuses on core connectivity and basic CRUD operations.

​
Stack 2: macOS SwiftUI admin application

A native admin interface for corpus management and data curation, built with Swift and SwiftUI. Core MVP features: The MVP focuses on admin operations; it is not (yet) a user-facing research interface.

​
Target users

Primary audience:
  • Scholars and academic researchers working with manuscripts
  • Linguists analyzing lexicons and language data
  • Biblical and religious studies researchers
  • Manuscript enthusiasts and paleography students
  • Language learners interested in text fabric research
Secondary audience:
  • Researchers preparing publications or thesis work
  • Digital humanities projects requiring text corpora
  • Theological and historical research initiatives
Use cases:
  • Offline scholarly research on sacred texts
  • Corpus preparation for NLP or computational linguistics
  • Manuscript digitization and cataloging
  • Lexical analysis and linguistic research

​
Scope

​
In scope (MVP)

  • Local Supabase sandbox environment (TypeScript/Deno)
  • macOS SwiftUI admin application
  • Core CRUD operations for corpora
  • Basic authentication and access control
  • File upload and ingestion pipeline
  • Metadata management (title, source, language, etc.)

​
Out of scope (future phases)

  • Web-based UI for corpus access
  • Multi-platform support (Windows, Linux, iOS)
  • Advanced research analytics or visualizations
  • Collaborative/multi-user features
  • Public sharing or distribution features
  • AI-powered analysis tools

​
Success criteria (MVP)

  • Local Supabase sandbox runs without errors
  • macOS app successfully connects to the local sandbox
  • Users can upload PDF/text files to the sandbox
  • Users can view, edit, and delete corpus entries
  • Metadata (title, language, source) is persisted and retrievable
  • App is performant with 100+ corpus entries
  • All operations work offline — no internet required

​
Timeline and phasing

1

Phase 1 (MVP)

Local Deno sandbox + basic macOS app: sandbox scaffolding and database schema, SwiftUI UI and file upload, then integration, testing, and polish.
2

Phase 2

Advanced editing and annotation tools.
3

Phase 3

Research and analysis dashboard.
4

Phase 4

Web interface or collaborative features.

​
Development notes

  • Language priorities: TypeScript for full-stack consistency, Swift for native performance
  • Offline-first: All critical features work without network connectivity
  • Data safety: Local encryption option for sensitive manuscripts (future)
  • Extensibility: Designed for future corpus types and metadata schemas
  • UX priority: The admin app should be intuitive for non-technical scholars