Hi HN, over the last couple of weeks I created a reactive Python notebook that you can run in your browser without relying on the server-side code, everything is a static file.
I wanted to build an experimentation platform that you can bring everywhere the web goes, i.e. a URL that you can share with anyone to start playing with a Python environment. It doesn’t matter whether it’s loaded on desktop or mobile.
My story, simply put, is frustration with Jupyter notebooks. I wanted something much easier to work with, i.e. load a small dataset, run a function and visualize results. An environment where I can manipulate code & data and immediately see feedback.
I was motivated to create an environment that:
* runs a reactive Python, cells are not executed top-to-bottom, but topologically; therefore there’s no hidden state, i.e. whenever you change a variable, all cells that depend on that variable will be automatically recalculated
* runs entirely in the browser: a) there’s zero provisioning & setup b) rich visualization libraries
* stores notebooks as plaintext files, as little metadata as possible, I do want my research to be versioned in the most native format
* empowers the experiments and exploration
* ultimately, has a (subjectively) friendly UI
It’s an early alpha version, but I am happy to answer any questions. Thank you!
While this page is not buried deep in the archive, I’d like to take the opportunity to say thanks for everyone who’s taken the time to share your feedback. Thank you for those who tried the application so far, it’s rough around the edges, but I find this inspiring! Happy to chat over if you find it’s useful & interested.
4 comments
[ 2.9 ms ] story [ 19.2 ms ] threadI wanted to build an experimentation platform that you can bring everywhere the web goes, i.e. a URL that you can share with anyone to start playing with a Python environment. It doesn’t matter whether it’s loaded on desktop or mobile.
My story, simply put, is frustration with Jupyter notebooks. I wanted something much easier to work with, i.e. load a small dataset, run a function and visualize results. An environment where I can manipulate code & data and immediately see feedback.
I was motivated to create an environment that:
* runs a reactive Python, cells are not executed top-to-bottom, but topologically; therefore there’s no hidden state, i.e. whenever you change a variable, all cells that depend on that variable will be automatically recalculated
* runs entirely in the browser: a) there’s zero provisioning & setup b) rich visualization libraries
* stores notebooks as plaintext files, as little metadata as possible, I do want my research to be versioned in the most native format
* empowers the experiments and exploration
* ultimately, has a (subjectively) friendly UI
It’s an early alpha version, but I am happy to answer any questions. Thank you!