Aptlantis is a long-horizon archival, mirroring, and metadata-first infrastructure project focused on preserving software ecosystems, documentation, and technical artifacts in forms that remain inspectable, reproducible, and useful over time.

At its core:

The internet is more fragile than people think.

The goal is not just to host files —
it is to preserve structure, context, provenance, and access in ways that remain useful to:

• humans
• crawlers
• AI systems
• future tooling

Aptlantis is a system composed of:

• static-first archival websites
• software ecosystem mirrors
• metadata generation pipelines
• schema-driven page generation
• hashing, signing, and verification tooling
• knowledge ingestion and indexing systems
• desktop and command-line tooling for structured workflows

The project is built around durability:

• ship data often
• deploy code rarely
• keep formats inspectable
• make systems understandable
• avoid unnecessary runtime complexity

• Preserve first
• Structure everything
• Make metadata a first-class artifact
• Prefer static output over runtime dependence
• Keep pages crawlable, inspectable, and long-lived
• Favor transparent pipelines over black-box systems
• Design for future readers, not just current users

Aptlantis explores and archives software ecosystems and registries, including:

• package registries
• language documentation
• release archives
• metadata snapshots
• long-term reproducible mirror structures

Current ecosystem efforts include:

• Rust / crates.io
• Python
• Node.js
• RubyGems
• Flatpak / Flathub
• and related technical archives

Aptlantis treats datasets, manifests, schemas, and indexes as primary outputs, not implementation details.

This includes:

• machine-readable archive manifests
• JSON / JSONL / TOML / Parquet pipelines
• dataset segmentation and packaging
• archive release metadata
• static metadata surfaces for downstream tooling

Long-term archives are only useful if they can be verified.

Aptlantis includes work around:

• multi-hash verification pipelines
• detached signing workflows
• reproducible release packaging
• publication-oriented integrity manifests

This includes:

• AAMHS — Aptlantis Archive Multi-Hash Standard

AAMHS defines how archive artifacts are hashed, described, and signed for publication and long-term validation.

Aptlantis websites follow an archival-first model:

• fully static or mostly-static output
• minimal JavaScript
• graceful degradation
• crawler-friendly HTML
• embedded metadata and schema
• long-lived, inspectable formats

The goal is to build sites that remain useful for:

• crawlers
• archives
• AI systems
• researchers
• future reconstruction efforts

Aptlantis includes internal systems for:

• document ingestion
• indexing and embeddings
• metadata extraction
• command generation
• schema transformation
• archive introspection
• structured editor workflows

These systems support both:

• building the archive
• navigating the archive

Aptlantis is intentionally polyglot — tools are chosen based on fit, not preference.

• Go — archival tooling, manifests, hashing, mirroring
• Python — data pipelines, metadata generation, indexing
• Rust — performance-critical tooling and validation
• TypeScript / JavaScript — dashboards, tooling, interfaces
• C# / .NET — Windows-native tools and desktop systems

• MongoDB
• DuckDB
• Docker
• Caddy
• Cloudflare
• HTMX
• Tailwind CSS
• D3.js
• WSL
• GitHub Actions

Aptlantis is developed primarily on Windows, with Linux, Docker, and WSL used where appropriate.

Common tools include:

• JetBrains IDEs (GoLand, PyCharm, RustRover, Rider, CLion)
• Visual Studio / MSVC
• Windows Terminal / PowerShell
• WSL
• Docker & Docker Compose
• GitHub
• Caddy
• Cloudflare Tunnels
• local-first scripts that eventually stop being one-off

This repository acts as a shared foundation across the Aptlantis ecosystem.

It may contain:

• shared documentation
• architecture notes
• release metadata
• templates
• manifests
• standards drafts
• infrastructure notes
• cross-project configuration
• publication workflows

This is less a single application and more a working surface for a larger preservation system.

Unless otherwise noted, projects and materials are released under the MIT License.

Open reuse, inspection, remixing, and archival redistribution are encouraged.

“Build for the version of the internet that survives.”

Or simply:

Preserve first. Optimize later.