$ cd /projects/urGlass
2026-02-24

← back

urGlass — Open Source Smart Glasses

GitHub: private


PRE — Idea · Setup · Build

Goal: Build open source smart glasses with a plugin framework.

A HUD for everyone. Not Google's $1,500 beta program for Silicon Valley insiders. Your Glass. ur Glass.

The idea had been in my head for years. The vision: a tiny OLED screen and a mirror element that projects text onto real glass lenses — a miniature heads-up display, like the HUD you know from cars but mounted on your face. "Incoming call." "Google Maps: turn left in 200m." "WhatsApp: Mom says hi."

But the real innovation wasn't the hardware. It was the software architecture. I wanted to build an open source framework on GitHub where anyone could write their own extensions. A connectivity layer — a standard protocol — so that today you get phone notifications, and tomorrow some guy codes a plugin for his smart doorbell that does "ding dong" on your glasses.

Basically: MCP before MCP existed. A universal connectivity standard for wearable displays. In 2020.

(For the uninitiated: MCP is Anthropic's Model Context Protocol, published in 2024, which does exactly this — a standard for connecting AI models to external tools and data sources. Same idea, different domain. I was four years early and in the wrong industry.)

The deeper motivation: less phones, more reality enhancement. Stop pulling a glass rectangle out of your pocket 200 times a day. Instead, overlay relevant information directly onto your field of vision. Enhance reality rather than replacing it with a 6-inch screen.

This is the same obsession that shows up across multiple projects:

  - urGlass: augment vision with information
  - Real-O-Tron: augment sensation with hardware
  - Realify: strip away illusions, reveal actual reality
  - ChordKiller: augment audio beyond what commercial DACs deliver

Different inputs, different outputs, same drive: make reality better than the default settings.

Stack:
  - Arduino (Adafruit ecosystem — Flora, Bluefruit LE, NeoPixels)
  - SSD1306 OLED display (128x64 pixels)
  - BLE (Bluetooth Low Energy) for phone-to-glasses communication
  - iOS / Swift (Adafruit Bluefruit app as foundation)
  - Mini mirror/projector optics for HUD effect

POST — Learnings · Afterthoughts · Timeline

What happened:

Started with the projection technology. Got the Adafruit PCB, the OLED, the BLE module. Experimented with iOS-to-BLE communication — that part was genuinely cool. Built a proof of concept that could receive text over Bluetooth and display it on the tiny OLED. The test message was a simulated incoming call:

  "Eingehender Anruf
   -----------------
   André Richter
   Status: Klingelt..."

It worked. Text from phone to glasses via BLE. The basic pipeline was proven.

Then I gave it up. Not because the tech didn't work — it did. But building physical hardware products is a different beast than building software. The optics (getting the OLED image to project cleanly onto glass), the form factor (making it look like glasses, not a science experiment), the power management (batteries small enough to fit into an arm piece) — these are industrial design problems, not software problems. And I'm a software guy who occasionally does hardware, not the other way around.

The Adafruit PCB is still here. Somewhere. In a drawer. With the other dreams that needed a mechanical engineer.

Learnings:
  - The idea was right. The timing was right. Meta released
    their Ray-Ban smart glasses in 2023. The market validated
    the concept. I was three years early with a prototype and
    no supply chain.
  - Open source hardware frameworks need a community, not just
    a vision. One person can code a plugin system. One person
    cannot manufacture lenses, design hinges, and figure out
    where to put a 100mAh battery.
  - BLE communication between iOS and Arduino is surprisingly
    elegant. The Adafruit Bluefruit ecosystem is genuinely well
    designed. Learned a lot about embedded BLE in one day.
  - The "reality enhancement" pattern in my work isn't random.
    It's a consistent thread: augment what's real, don't replace
    it. Glasses, not goggles. HUD, not VR. Sound quality, not
    sound effects. This is apparently who I am.

Timeline:
  - 2020-08-07: One day. Four commits. Arduino + OLED + BLE +
    iOS proof of concept. Incoming call notification displayed
    on a 128x64 pixel screen via Bluetooth. Then: drawer.

Status: Hibernating. The PCB is in a drawer. The idea is in the
  zeitgeist. Meta, Apple, and Google are spending billions on what
  I prototyped in a day with an Adafruit board and spite.