Digital gardens and publishing

This project functions as a curated software directory and developer resource index, providing a centralized platform for discovering and evaluating high-quality open-source repositories. It serves as an aggregator that monitors trending software and educational resources, organizing them by technical domain and programming language to assist developers in identifying tools for their specific technical challenges. The directory distinguishes itself through a community-driven curation workflow, where repository lists are validated and updated based on collective developer consensus. This information is maintained within structured, human-readable markdown files, ensuring that the evolution of the index remains transparent through version-controlled change tracking. To maintain accuracy, the platform utilizes automated data aggregation to fetch real-time metadata and contributor statistics directly from remote service endpoints. Beyond simple discovery, the platform supports analytical research into open-source ecosystems. Users can evaluate the maintenance status and reliability of projects by researching contributor activity, analyzing technical stacks, and monitoring project lifecycles through versioned releases and changelogs. These capabilities facilitate informed decisions regarding software integration and dependency management.

MkDocs is a static site generator designed specifically for creating project documentation. It functions as a command-line utility that transforms structured Markdown files into professional, searchable websites. By utilizing a centralized configuration file, it manages site metadata, navigation hierarchies, and build settings to ensure consistent output across documentation projects. The platform distinguishes itself through a highly extensible architecture that separates content from presentation. Users can apply visual themes to control the site's appearance, while a plugin-based build system allows for custom hooks to intercept and modify the generation process. This flexibility enables developers to integrate third-party Markdown extensions, inject custom assets, and automate tasks like link validation or search indexing. The tool provides a comprehensive environment for documentation maintenance, featuring a live-reload development server that offers immediate visual feedback during the editing process. It supports complex site structures, including nested navigation and directory-based URL formatting, and facilitates deployment by generating static files ready for hosting on any web server or platform. The project is distributed as a Python-based package, providing a standard command-line interface to initialize new projects, manage dependencies, and execute the full documentation build lifecycle.

This project is an enterprise knowledge platform designed for teams to create, manage, and publish structured documentation. Built on a high-performance runtime, it provides a centralized environment where contributors can author content using markdown, HTML, or a visual editor. The system is engineered to handle collaborative workflows, ensuring that technical and non-technical users alike can maintain documentation with consistent rendering and version control. What distinguishes this platform is its focus on secure, scalable, and synchronized content management. It features granular path-based access controls and pluggable authentication middleware that integrates with enterprise identity providers, social logins, and multi-factor security layers. To ensure data persistence and collaborative safety, the platform supports direct synchronization with version control repositories and automated backups to cloud object storage. The system offers a comprehensive suite of tools for managing organizational knowledge, including multilingual localization, media asset management, and detailed content version history. Administrators can further tailor the platform through custom visual themes, site-wide style injections, and flexible search configurations that support both built-in engines and external indexing services. The architecture is designed to scale across diverse infrastructure, from low-power hardware to high-performance cloud environments.

Markdownload is a browser extension that functions as a markdown web clipper, converting webpages and selected text into clean markdown files for offline storage and archiving. It operates as a content extractor that isolates the main document from the page while removing navigation elements and advertisements. The tool includes a template generator for injecting dynamic front-matter and metadata into documents via user-defined placeholders. It also serves as a local media downloader that saves remote images to the filesystem and updates links to reference those local files. Additionally, it acts as an integration tool to transfer captured web data and metadata directly into Obsidian vaults using custom URI schemes. The extension supports capturing content from all open browser tabs simultaneously and clipping specific highlighted text. Users can customize markdown styling for links and images, organize downloaded files into specific subfolders, and export media as formatted embeds or hyperlinks to the system clipboard.

Pumpkin-book is an open-source educational textbook that provides annotated study materials and mathematical derivations for foundational machine learning concepts. It functions as a technical documentation archive, breaking down dense academic literature into accessible, plain-language notes designed to support self-paced learning. The project distinguishes itself through a collaborative knowledge curation model, where the curriculum is managed via a version-controlled system. This workflow relies on community-driven updates and peer review to refine explanations and ensure the accuracy of the technical content. By utilizing markdown-based authoring and static site generation, the repository transforms structured source files into a navigable, chapter-based reading experience that mirrors the logical progression of advanced computational theory. The content is organized into a hierarchical directory structure that guides learners through specific technical topics using curated examples. The entire curriculum is maintained as a public repository, allowing contributors to track changes and participate in the ongoing development of the educational material.

This is my own personal website, built with Astro and Svelte. It also holds my own personal blog.

ArchiveBox is a self-hosted archiving tool designed for personal digital preservation and research data management. It functions as an automated web preservation engine that monitors URL inputs from bookmarks, browser history, or manual entries to capture and store permanent, offline copies of web content. By utilizing headless browser automation, the system renders dynamic web pages to ensure that captured snapshots, PDFs, and media assets remain accurate and accessible even if the original source disappears. The project distinguishes itself through a modular extractor pipeline and a task-queue-based processing model, which allow it to handle long-running ingestion jobs reliably and at scale. It organizes all captured data into a predictable, file-system-based directory structure, ensuring that archives remain portable and accessible without the need for a dedicated database engine. This architecture supports the generation of static, self-contained archives that can be hosted on any standard web server. To maintain high fidelity across diverse web environments, the system includes configuration-driven dependency management that coordinates the necessary browser binaries and command-line tools. The platform provides a comprehensive suite of command-line interfaces, configuration options, and core modules to support operational management and integration. Detailed documentation is available to guide users through installation, dependency maintenance, and the security considerations of managing archived web content.

This project is a community-driven directory that aggregates and categorizes high-quality technical resources, tools, and learning materials. It functions as a centralized knowledge management repository, designed to help developers navigate the software development landscape by providing structured access to curated lists and external project references. The directory relies on a collaborative, peer-reviewed workflow where external contributors submit and maintain links through a version-controlled system. This community-maintained approach ensures that the information remains current and reflects collective intelligence, allowing users to explore technical ecosystems and discover specialized solutions across various domains and programming languages. The content is organized using structured text files that serve as both the primary source of truth and the final presentation layer. These files are managed through a decentralized revision system, which tracks the historical evolution of the documentation and supports concurrent contributions from the community.