##Thoughts on a #fediverse based #data science platform
Say take an instance, put something like jupyterlite on it, and federates across instances. How'd that be?
@freemo what would you say?
🇳🇱
@arinbasu I'm not sure if I am following.
So I used jupyter notebooks before. It usually gives you a nice little web based IDE with your own environnment, I liked it. But there isnt much of a sharing/social aspect to it. You can of course post notebooks to anaconda, but generally your isolated in your own little workspace... Where is the federation element?
@freemo
Right. However, with something like jupyterlite
https://jupyterlite.readthedocs.io/en/latest/
It is possible to create a web based data science environment that can be served over the web. There is also observablehq that allows for dynamic visualisation.
The federated part is similar to this discussion that we have using text, except now live code fragments can be shared and evaluated along with visualisation over the federated instances.
It’s still fuzzy in my head but after working in qoto.org, these possibilities seem realisable, 🙂
@arinbasu One day an entier federated work flow would be interesting.. not sure if it is practical or how it would look though.
@freemo
I mighty like the idea!
@arinbasu Startt coding! Once you have something I can run on a server ill run it here on QOTO for ya :)
@freemo
😃 Love it!
@arinbasu I've been patiently waiting a long time for a federated github... It almost happened a few times but the projects ultimately never picked up steam... maybe you can pick up where they left off.
@freemo
I am sure you've seen this but sharing it here anyway for reference,
https://gitlab.com/federated-git-platform/whitepaper
@freemo
Yup. As a matter of fact, that would make sharing data visualisation and analyses a lot better with version control