February 9, 2026
The Notebook Cage Match
I made an open-source juypter alternative
Open‑source Jupyter rival drops — cheers, gripes, and a spelling war
TLDR: A developer launched MoreCompute, a free local tool for running and converting Python notebooks compatible with Jupyter. The comments mix excitement, dissatisfaction with Jupyter, a request for richer writing formats, and a brief flare‑up over a mistaken “Jupyter is closed” claim that the community quickly corrected.
One indie dev just dropped MoreCompute, a free, open‑source tool to run Python “notebooks” locally in your browser. It opens on your machine (no cloud), and it can convert back and forth with Jupyter’s standard file format so your old work still opens in tools like Google Colab. That’s the news — but the comments? That’s where the fireworks are.
Early hype arrived fast: one user said they’ve been hunting for a Jupyter alternative and are “excited to see where this goes.” Another asked the builder why they felt the need to make a rival at all, confessing they’re “kind of unhappy” with Jupyter too — laying out the central vibe: curiosity mixed with pent‑up frustration.
Then the thread took a spicy turn when someone asked if Jupyter was closed source. Cue record scratch. Others (politely) pointed out that Jupyter is open‑source; still, the moment lit up the chat like a reality‑TV cutaway. Meanwhile, a surprise subplot: a plea for a version that uses AsciiDoc instead of Markdown because “writing anything complicated” in Markdown can be painful. Notebooks, but make them book‑book.
Also making the rounds: jokes about the title’s “juypter” typo — because it’s not a launch without a little chaos. Verdict from the crowd: promising new toy, clear pain points, and plenty of feature wish‑lists already forming.
Key Points
- •MoreCompute is a local interactive Python notebook environment using .py files with # %% cell markers.
- •It supports conversion between .ipynb and .py, extracting dependencies from !pip install and adding UV metadata.
- •Prerequisites are Node.js v20 and Python 3.12; installation via uv (recommended) or pip is provided.
- •Development setups include Devcontainer (Docker Desktop + VS Code/Cursor), Docker-only, and native (pyenv, nvm, venv, npm).
- •Configurable environment variables set default ports: backend API at 3141 and frontend UI at 2718; the project is MIT-licensed.