Maing capslock useful

tap to esc, hold with hjkl for arrows

git clone https://github.com/rvaiya/keyd
cd keyd
make && sudo make install

sudo systemctl enable --now keyd

cat /etc/keyd/default.conf
[ids]
*

[main]
# Create a dedicated navigation layer called 'vim_nav'
# Tap = Esc, Hold = Activate 'vim_nav' layer
capslock = overload(vim_nav, esc)

[vim_nav]
# Map hjkl to arrow keys
h = left
j = down
k = up
l = right

# --- Optional but recommended ---
# Map other useful navigation keys to this layer for convenience
# Example: Caps + u/i = PageUp/PageDown
u = pageup
i = pagedown
# Example: Caps + n/m = Home/End
n = home
m = end

sudo keyd reload

OpenWorm

Would be fun to work on. OpenWorm.

Dont be the strongest ant

Move into another category. The strongest ant is not as strong as a human. Humans moved in another category when they formed groups and again when they formed societies. Terrorists are groups, nations are societies. The first can not cause real damage - compared to an army. Here is a funny example of category change: Epic final moments of Kung-Fu Panda 3

democratizing sharing of insights

It would be nice if all insights travel into the right brains at the right time. Existing solutions are not there yet. On youtube, you have to do your videos about one area only. This is the same problem with other follower-based platform as well. It would be strange to get a video in your feed about 3blue1brown on insights about cheesecake backing in his specific oven type. Blogs are ok, but when you have RSS, it is the same problem. Tiktok goes in the right direction. Content is king, nothing else, followers dont count. But due to insentive problems tiktok is aweful. Google might be good, but where to publish. In a dream world of a decentralized web with a search engine on your side it could work. LLMs, maybe with sources, are a very promising candidate for such a database. But there are currently no insentives to share your insights.

Motivating EVERYONE to share EVERYTHING slightly useful for SOMEONE and then bringing that to the right person at the right time is an important unsolved problem.

Is everything exclusively relational / defined by its context?

I had a conversaion about where taste comes from. We concluded from associations. I constructed an example where you had to choose between 2 things that were clearly distinct, but had NO existing associations. We could not come up with a preference.

But if positive preferences diffuse through the association-web in our mind, where is the source? Maybe a more fundamental algorithm like evloution and the drive to survive are the sources, diffusing through associations to other objects building our preferences

For example I like a nice shirt, because I associate it with wealth, associating it with resource abundance, which evolution programmed into me since I was a single cell is a good thing

Anyway, just speculations, I did 0 seconds of research.

which problems to work on?

Dimensions: Interesting, Hard, I can find joy. We assume valuable is given. Wook for not interesting, hard and where I am able to find joy in it.

• not interesting, not hard, no joy: obviously bad
• not interesting, not hard, joy: hobby
• not interesting, hard, no joy: why bother?
• not interesting, hard, joy: great! A good problem
• interesting, not hard, no joy: just use others work
• interesting, not hard, joy: why is i not solved already?
• interesting, hard, no joy: use others work and be happy about it
• interesting, hard, joy: would be nice, but the space is crowded

paste to stdin of a program

I am using wayland, thus I do

sudo apt install wl-clipboard
wl-paste
wl-paste | xargs echo
wl-paste | my-program
echo hi | wl-copy

and put in .zshrc

alias paste="wl-paste"
alias copy="wl-copy"

Linux is great.