Epics for Pecha Server and API

πŸš€ WG ΰ½¦ΰΎ‘ΰ½ΊΰΌ‹ΰ½šΰ½“ΰΌ πŸ’β€β™‚οΈ Pecha API WG
1

Epics for Pecha Server and API

Here is a set of high-level Epics to guide the development of the Pecha Server and API, aligned with the Product Requirements Document (PRD) and tailored to the distinct needs of the WeBuddhist.com team, the Pecha AI Studio team, and external users.

Epic 1: Foundational API Infrastructure & Data Modeling

Description: Establish a scalable, secure, and robust backend infrastructure using Firebase and Google Cloud Storage. This involves defining and implementing the core data models based on the OpenPecha Format (OPF) to ensure all texts, annotations, and metadata are stored consistently and efficiently. This epic is the bedrock for all other functionalities.

Goal: To create a stable and secure foundation that can support data ingestion, storage, and delivery for all user groups.

Potential User Stories/Features:

  • Set up and configure the Firebase project (Authentication, Firestore, Cloud Functions).

  • Establish Google Cloud Storage buckets with appropriate permissions for data storage.

  • Implement the OPF data schema in Firestore for texts and annotation layers.

  • Develop a secure authentication and authorization system for different user roles.

  • Create a CI/CD pipeline for automated testing and deployment of the backend.

Epic 2: Data Ingestion & Standardization Pipeline

Description: Develop and integrate the Pecha Toolkit parsers with the backend to create a seamless, automated pipeline. This pipeline will be responsible for ingesting, standardizing, and storing Buddhist texts from various sources (e.g., BDRC, user-submitted DOCX files) into the OPF format.

Goal: To reliably populate the database with standardized, high-quality data from diverse sources, making it ready for consumption.

Potential User Stories/Features:

  • Integrate the Toolkit’s OCR and DOCX parsers with a cloud function for automated processing.

  • Build a workflow to process the complete BDRC dataset and ingest it into storage.

  • Implement robust versioning for all ingested pechas and their annotations.

  • Develop validation checks to ensure data integrity during the ingestion process.

Epic 3: Public Read API for Data Consumption (WeBuddhist & External Users)

Description: As a developer at WeBuddhist.com or an external academic institution, I need a well-documented, reliable, and performant REST API. This API should allow me to easily retrieve Buddhist texts, translations, commentaries, and their associated annotations, so that I can display them in my web/mobile application or use them for research and publication.

Goal: To provide fast, open, and easy-to-use endpoints for consuming Pecha data, enabling wide-scale access and integration.

Potential User Stories/Features:

  • Develop GET endpoints to retrieve a full Pecha (text and all its annotations).

  • Develop GET endpoints to fetch specific layers of annotation (e.g., only translations, only commentaries).

  • Implement filtering and pagination to efficiently browse lists of available texts.

  • Create comprehensive, public-facing API documentation (e.g., using Swagger/OpenAPI).

  • Optimize API response times for a smooth user experience on client applications.

Epic 4: Contribution & Enrichment API (Pecha AI Studio Team)

Description: As a developer on the Pecha AI Studio team, I need a secure and well-defined API to programmatically load texts and then create, update, or delete annotations. This will empower scholars and translators using our editor suite to enrich the textual data by adding translations, corrections, and other valuable insights.

Goal: To enable the Pecha AI Studio team to build powerful editing tools that contribute high-quality annotations back into the ecosystem.

Potential User Stories/Features:

  • Develop secure POST/PUT/DELETE endpoints to manage annotation layers on a Pecha.

  • Implement secure, token-based authentication specifically for write/edit operations.

  • Ensure rigorous data validation for all incoming annotation data to maintain quality.

  • Maintain a complete and accessible revision history for all changes made via the API.

  • Develop an endpoint to allow authorized users to upload a new Pecha.

Epic 5: Advanced Search & Discovery Capabilities

Description: As a researcher or application developer, I need advanced search capabilities to find specific verses, terms, or texts across the entire dataset. This will enable in-depth analysis and allow for the creation of powerful discovery features for end-users, fulfilling the roadmap goal of integrating advanced search and RAG features.

Goal: To move beyond simple data retrieval and provide powerful tools for scholarly inquiry and data discovery.

Potential User Stories/Features:

  • Integrate a dedicated search service (e.g., Algolia, Typesense, or Elasticsearch).

  • Implement full-text search across all Pechas and their translations.

  • Allow faceted search based on metadata (e.g., author, source, language).

  • Develop a public API endpoint for submitting complex search queries.

  • Begin research and prototyping for Retrieval-Augmented Generation (RAG) features.

Epic 6: Multi-Modal Data Support (Audio & Image)

Description: Expand the API and data models to support non-textual data. This will allow users to store, link, and retrieve audio and image formats associated with the texts, thereby preserving and sharing Buddhist knowledge in its diverse forms as outlined in the project roadmap.

Goal: To enrich the dataset by incorporating audio and visual materials, providing a more comprehensive resource for practitioners and researchers.

Potential User Stories/Features:

  • Extend the OPF data model to include references to audio and image files.

  • Develop API endpoints for uploading and retrieving audio/image files from Cloud Storage.

  • Implement functionality to link audio/image annotations to specific text spans (e.g., an audio clip of a specific verse being recited).

πŸ•‰οΈ Homepage