Emacs

Fabric¹ is a collection of crowd-sourced prompts, exposed via a CLI tool. I used it for a while some time ago but never fully exploited it because I prefer Emacs.

Eshell buffers are an option, but I am principled in my tool usage and prefer to delegate longer-running CLI tasks to a combination of Alacritty and Tmux.

Maintaining my Emacs shell usage to ephemeral popups feels natural.

Gptel² is a versatile LLM client that integrates smoothly into my workflow (buffer/text manipulation and management) without disrupting my thought flow.

Alter your doom block in init.el:

(doom! 
    ...
    :lang 
    (python +lsp +pyright +conda)
    ...)

I recommend pyright, is snappier than pyls..

you’ll also need lsp enabled in your tools, I also have +peek

(doom! 
    ...
    :tools
    (lsp +peek)
    ...)

for more info, hit <space> h d m lsp<CR>

Alter your config.el (I use miniforge to get mamba and conda):

                                        ; Conda
(use-package! conda
  :config
  (setq conda-anaconda-home (expand-file-name  "~/mambaforge"))
  (conda-env-initialize-interactive-shells)
  (conda-env-initialize-eshell)
  (conda-env-autoactivate-mode t)
  (add-hook 'find-file-hook (lambda () (when (bound-and-true-p conda-project-env-path)
                                    (conda-env-activate-for-buffer)))))

Once that’s done, hit <space> h r r and your ready to go.

People have been responding to emacs and my beard.

For a while I thought I did not have a lot to say. Turns out a lot of other humans are quite interested in the much of the same things that I’m into: my hatred for the mouse, and the consequent love for efficient workflows being some of them.

I’ll give the people what they want: I get to enjoy esoteric tech and spread the word, growing a cult, while other humans have enlightened experiences (believe me I do talk like this in person) and continue propogating the techno-propaganda about enjoying computer science.

Emacs has been a trustworthy companion of mine for around 4+ years now. We’ve shared some highs and lows. I owe a major fraction of whatever conceptual depth I maintain in computer science to messing around with my tools.

Emacs is also the core of my first major youtube breakthrough.

I haven’t yet explored the emacs niche completely and will be investing more time into honing my repertoire and understanding what sort of mindset the tooling expects from you rather than just fitting emacs into my thoughts.

I recently shifted my operations to a thinkpad x13g4amd (Ryzen 7840U + 780M Radeon) and am using this transfer as an opportunity to realize and improve my work related habits.

GNU/Linux

All of my pipelines are based around Linux and the core GNU utils. I had to use windows for a while due to org-wide constraints but I’ve recently been given the signal to shift to linux.

I rarely used any windows native tools for my workflows and there was always a layer of linux emulation/virtualization (wsl, scoop, and the likes) upon which I conducted my usual operations. Given I use Debian at home, a non native linux experience was frustrating at best and depressing at worst. No more, do I have to confine myself.

I’m an AI Research Engineer and that involves messing around a little with python. I’ve spent the past 4 years perfecting a disciplined, minimal, but enabling setup.

This is a tour of my trusty set of tools, in the hopes that it will help you find the same CLI zen that I’ve been enjoying for a while now.

Tmux : towards an Eternal Shell

I start my work sessions off by ssh’ing into my remote compute cluster.

I’ve been building up my vimrc again because emacs’ tramp mode just wasn’t cutting it when it came to speed for my remote work environments.

I’ve been a vim user for around 4 years now and having read some books partially and sampling a lot of blogs and conference recordings over this span, I decided I should commit and formally invest into a definitive resource to get me upto speed and beyond.

This is an auxilliary post collating resources for the recent video I posted …

The Pipeline

1. All the ideas, resources that I want to process, any miscellaneous questions I have, are fed into the input-queue in the buffer
2. All the manipulation takes place in these buffers - they’re org-files and I use org-roam to maintain the connections
3. whenever a node set ripens and is worth sharing, I write a post or publish a video.
4. It can go both ways : I can force a set into maturity if I wish to publish something specific or I may chance upon a concept when observing connections.

Observations

I use org-roam-ui to visualize the buffer and check for linkages that might result in something useful. I also want to publish this graph (demonstrated in video) but there’s no explicit solution for that yet and I’m planning to build one myself with rust and webassembly as a compilation target.

This is the first in a series of blog posts that follow the educational common lisp series on my youtube channel as an auxilliary.

I’ll be summarizing the videos in these blogs and be using these to point to references and additional resources that further elaborate the matter.

[As of 0x213B : 2023-08-17 Thu]

I use emacs for a lot of my daily tasks and spend majority of my time in it. This is a review of some significant components of my init.el

;keyboard all the way
(menu-bar-mode -1)
(tool-bar-mode -1)
(scroll-bar-mode -1)

;I don't like distractions
(setq byte-compile-warnings '(cl-functions))
(setq ring-bell-function 'ignore)
(setq visible-bell t)

I chose the most recently engineered package-management solution when I began with emacs, haven’t switched since and don’t think I’ll need to. Migrating from vim, I preferred a configuration file rather than installations via melpa interface. Also, freezing and thawing is cool : exact reproducibility is guaranteed.