OUTLAND 🌲 — Less Scroll. More Soul.

This is a submission for the Hacktoberfest Open-Source AI Challenge — Week 1: Touch Grass.

What I Built

OUTLAND is a local AI-powered outdoor adventure guide that turns spare moments into real-world discoveries.

We spend countless hours scrolling through the internet, sometimes overlooking the beauty of the world around us. OUTLAND encourages a different approach: tell it how much time you have, your current mood, where you want to go, and an optional interest, and it will generate a practical outdoor mission tailored to you.

Want to spend 15 minutes observing birds at a nearby park? Or take a short walk around your neighbourhood and discover something new? OUTLAND helps turn that intention into action.

Its philosophy is simple: less scroll, more soul. The screen should be the shortest part of the experience.

Demo

OUTLAND currently runs locally on your computer. Follow the setup instructions in the repository to try it yourself.

Repository: github.com/Txxjasss/Outland

OUTLAND in action 🌲 — a locally generated outdoor mission tailored to the user's available time, mood, destination, and interests. Powered by Qwen2.5 3B through Ollama.

Code

GitHub Repository: OUTLAND

The project is built with vanilla HTML, CSS, and JavaScript, without a frontend framework or build step.

The repository includes the application source code, setup instructions, and an MIT license.

How I Built It

OUTLAND uses Qwen2.5 3B through Ollama as its local AI engine.

Users can customise their missions based on:

• Available time: 5, 15, 30, or 60 minutes
• Current vibe: Curious, calm, playful, creative, or needing a reset
• Destination: A nearby park or garden, neighbourhood, campus, backyard, or another safe outdoor place
• Optional interest: Birds, architecture, photography, or another topic they want to explore

The app combines these preferences into a prompt and sends it to Ollama's local HTTP API. The model generates a personalised outdoor mission with actionable steps and safety guidance.

I also implemented a deterministic fallback mechanism. If AI generation fails, times out, or returns an invalid response, OUTLAND displays a predefined mission instead.

The interface clearly distinguishes LOCAL AI MISSION from OFFLINE MISSION, ensuring that fallback content is never misrepresented as AI-generated.

Tech Stack

• HTML5, CSS3, and vanilla JavaScript
• Ollama for local AI inference
• Qwen2.5 3B as the configured model
• Python's built-in HTTP server for local development

Why Does Open Innovation Matter?

For OUTLAND, open innovation is more than a technical decision — it supports the project's core purpose.

Privacy: When local inference succeeds, prompts can be processed on the user's own computer rather than sent to a hosted AI provider. The application does not require user accounts or a cloud database.

Control and flexibility: Using an open-weight model through Ollama gives users a way to run inference locally and experiment with compatible models instead of depending entirely on a proprietary, hosted API.

Accessibility and resilience: Local inference reduces dependence on paid cloud inference services, although it still requires compatible hardware and an installed model. The deterministic fallback system also keeps the prototype useful when AI generation is unavailable.

Most importantly, the open approach allows people to inspect, run, learn from, and build upon the project. OUTLAND demonstrates how AI can be used for something beyond keeping people online: encouraging them to step outside, pay attention to their surroundings, and discover something new.

The goal isn't to make people spend more time using an app. It's to help them spend less time on their screens and more time experiencing the world around them.

Less scroll. More soul.