FYI, AutoHotKey on Windows is a much better (my opinion) version of Karabiner, with it's own programming language. It's quite amazing what you can accomplish with AHK: it goes well beyond key mapping to process control, building quick utilities or quick mini GUI applications, compiling those into executables that you can share, etc;

See https://www.dcmembers.com/skrommel/downloads/ or https://www.autohotkey.com/docs/scripts/ for a larger collection.

For example, I would have never thought of AltEdge ("Sends Alt-Tab when the mouse is on the left edge of the screen and keeps cycling through the windows") or Barnacle ("A programmable toolbar that fits inside any window") or MonitOff ("Turns off the monitor at user defined idle times during the day") but it took 10 seconds to download the executable and try out one of these (AltEdge).

https://www.dcmembers.com/skrommel/download/altedge/

For me, Windows' programmability, especially AutoHotKey, has been a major reason to find Mac OS rather unattractive. Over the years, AutoHotKey has solved several minor-but-daily annoyances for me.

On OS X, you're looking for Hammerspoon.

It uses Lua for its scripting engine.

Not sure there's quite an equivalent for the executable feature, but it does support publishing and installing modules written by users, which are called 'spoons', IIRC.

> compiling those into executables

Ha, never knew you could do that. Pretty cool.

"C:/Program Files/AutoHotkey/Compiler/Ahk2Exe.exe" /in Cleaner.ahk /out Cleaner.exe /icon Cleaner.ico

That's it :-)

The one thing I have against learning in public is that it often seems like pandering to the lowest common denominator with generic learnings/advice.

I think the people who are inclined to be constantly learning are probably less inclined to focus on broadcasting their life to others.

That's not to say that there aren't people who learn in public and create valuable content in the process, they just seem rare.

I started doing this a couple of months ago[1], and the fact that my notes are public acts as something of a forcing function; I’ve found that I engage with technical content a lot better this way. I consider myself the primary consumer of these notes (I’m not looking to provide learnings/advice in particular), but making them public has been a step in the right direction for me.

[1]: https://timothyandrew.net/learning

Hey just wanted to say, after reading some of your posts, your subject matter is exactly the things I'm trying to learn, so I much appreciate your writing. I will be following your progress.

It's up to the person how deep he/she wants to go in learning. The idea behind wiki/notes is that you are the main user of it. You just open it up for everyone. Right now so many notes are done privately and that's a shame.

I also hope more people make an open learning repo [1] where they test out new technology in the open. It's been a nice productivity boost for me.

1: https://github.com/lpil/learning & https://github.com/nikitavoloboev/learning as examples

> Right now so many notes are done privately and that's a shame.

I have a notes SVN repository dating back about a decade with over 600 text files. I could split them into many more files as these notes aren't "atomic" as Zettelkasten folks say.

The main barrier to release at this point is that many of these notes contain private information, or have brutally honest assessments of certain things that I don't want to be made public. If I had written them with public consumption in mind from the start then I would have done things differently.

Another problem is that the quality of many of these notes is not at a level I'd be comfortable releasing. If I had intended to release these notes in the first place I would have spent some extra time to clean them up earlier on.

My current plan is to slowly ramp up a public repository once I revamp my personal website, but as with many plans, I keep pushing that into the future.

> it often seems like pandering to the lowest common denominator with generic learnings/advice.

I have a huge problem with this and I'm not sure how to fix it. When I start researching something, I don't write about it, because I don't feel like I know enough and I don't want to repeat the same "generic advice" that many people already presented. Then, when I reach my desired level of understanding, I start writing a blog post. I start with "fundamentals" I had to grasp before I could understand the topic. Then, after 6000 words, some diagrams, and several tens of hours put in, I start writing about the topic, but at that point it's almost guaranteed I'm already interested in something else. So, obviously, I leave the previous topic to rot in the "unpublished" folder forever.

It would, thinking realistically, take someone paying me for the extra time to finish one of these posts, which would become either a really looong article, or a short book/pdf. That won't happen, because most of my enquires about how to be paid for writing something like this lay in the same "unpublished" folder... That's on top of the topics being rather peculiar (most recent example: "Using Haxe's Lua target to script my (Awesome) Window Manager" - started from quick intro to OCaml, two fixes I made to the Haxe compiler (which is written in OCaml) to get Lua target really going, quick intro to Lua and LuaJIT, plus AwesomeWM architecture and class library... at which point I stopped, and switched to the next topic: "Wiring and writing a Home Security-like system with RaspberryPis and Erlang/Elixir", which will turn into another abandoned try at a write up.)

Basically, I start writing at the point I'm pleased with the solution to a problem (eg. the state of my WM, the monitoring around the house, or most recently a working Tizen Studio for native devel on non-Ubuntu Linux), which rids me of the problem, so then I have no motivation to write about it, or its solution anymore. I could go back, change the title, and generally edit whatever I have already scribbled, but that feels like work, and at that point I'm already learning about something else entirely...

This is why this:

> I think the people who are inclined to be constantly learning are probably less inclined to focus on broadcasting their life to others.

seems plausible to me, at least. Though whether my broadcast would be more valuable to others than the "generic advice" (or more specifically, shallow posts on some tech, in "3 things you didn't know about IEx"-like style) is doubtful.

I love Karabiner, though I only remapped a couple of keys and did the rest with HammerSpoon. I think I remapped the rock cmd to backspace and caps lock to hyper. The rest was HammerSpoon.

HammerSpoon: https://www.hammerspoon.org/

Do you have any specific configs/use cases for Hammerspoon that have jumped out as super useful?

For me, the two biggest improvements were:

1. App switching (hold F then use all keys under the right hand, each key bound to launch and/or focus a particular app)

2. Window management (hold W then use right hand as arrow keys to snap windows)

Paired with a keyboard that lets you remap held keys to modifier combinations (I use Ergodox) has been great.

There is one beyond those two that has been amazingly useful : I wrote a script to type out the contents of my clipboard. That's been very useful on more than one occasion.

Could you please elaborate how exactly does Karabiner yields that kind of multiplier effect?

Every key on my keyboard is custom hyper key.

My config: https://github.com/nikitavoloboev/dotfiles/blob/master/karab...

Article to explain how to achieve this: https://medium.com/@nikitavoloboev/karabiner-god-mode-7407a5...