Conversation

Notices

1. Embed this notice
   Ludovic Courtès (civodul@toot.aquilenet.fr)'s status on Monday, 04-Nov-2024 19:00:28 JST Ludovic Courtès

   Exposés du Groupe de travail Notebook du 4 octobre, organisé par @SReyCoyrehourcq & cie.
   https://www.canal-u.tv/chaines/gt-notebook/webinaire-20232024/gt-notebook-webinaire-7-jupyter-notebook-et-calcul-intensif

   J’y parle de Guix-Jupyter vers 1h30.

   #Jupyter #Guix #HPC

   • Embed this notice
     kakafarm (kakafarm@shitposter.world)'s status on Monday, 04-Nov-2024 23:49:45 JST kakafarm

     in reply to
     @civodul (i don't know french but (and sorry for the unprompted lamentful soliloquy)) Nooo.. noo.. Jupyter Notebooks are horrible. Why did we have to invent them? I've seen people write free software libraries using it and it's awful. AWFUL. It doesn't jive with git and diffs, and it's a pain ta work with! Why! ;______;
   • Embed this notice
     w96k (w96k@fosstodon.org)'s status on Monday, 04-Nov-2024 23:49:45 JST w96k

     in reply to

     • kakafarm

     @kakafarm they invent org-mode but worse.

     @civodul

   • Embed this notice
     Ludovic Courtès (civodul@toot.aquilenet.fr)'s status on Tuesday, 05-Nov-2024 01:59:02 JST Ludovic Courtès

     in reply to

     • kakafarm

     @kakafarm Well, I’m more on the side of Org, for many reasons.

     So yes, Jupyter “reinvents the wheel” to a large extent (the latest paper by @khinsen has nice thoughts on historical “programming systems” and incidental complexity in Jupyter), but it has become a go-to tool for many in reproducible research.

     Guix-Jupyter thus is sort of a gateway to reproducible deployment:
     https://hpc.guix.info/blog/2019/10/towards-reproducible-jupyter-notebooks/