@TheGingerBill The main speed up comes from proficiency of using your editor in the first place (saving potentially many seconds per action), not from a small variance of efficiently doing something in editor A vs B.
@TheGingerBill Why would sublime vs. vim vs. emacs vs whatever keybinds differ in speed? It probably averages out to being a similar number of keypresses / finger travel, assuming that that everyone uses shorter combinations for more common actions.
@TheGingerBill@phyce_ Furthermore, most of us terminal users also use some form of a tiling window manager for structure/organization, either on desktop level (i3, sway, dwm, etc), or on terminal level (tmux, zellij, etc). TUIs tend to work better with these.
@TheGingerBill@phyce_ A collection of terminal based tools that work well together creates an integrated experience. Similar to a large (Integrated)DE. A collection of distinct GUI tools (file manager, code editor, git client, debugger, task manager) does not have as good of a flow, more friction.
I don't like the expression "working hard". First, it's not "working" in itself that matters, it's achieving outcomes. And second, work should not be hard. You should find work patterns that enable high-volume productivity without incurring a high psychological or physical cost.
Common misconception: Newton's method provides the "ideal" descent direction, and everything else is a cheap approximation.
Couldn't be further from the truth!
Newton is great when you're really close to a minimizer… and is a total heuristic everywhere else.
[1/4] In Euclidean geometry, parallel lines don’t meet.
In *projective* geometry, they intersect “at infinity”. If that sounds plausible, but a bit vague, it can be made precise.
But in *complex projective* geometry, *any two circles* intersect at (at least) the same 2 points.
[1/2] This is what (part of) the circle
x^2+y^2=1
looks like, if x and y are complex variables.
The imaginary part of x is shown by moving points in the solution above/below the real plane, while the imaginary part of y determines the hue, with green for zero.
It looks as if