Damn, can’t distinguish between my feed and the hellthread notifications anymore, just contributed accidentally to the beast, again.
streetyogi
streetyogi@street.yoga
npub1nujs...qmun
Python | Cyberpunk | Yoga | German | Polish | English
Spending time on Nostr:
Where did my day go?
Doing a 5 minute workout:
Dat clock broken or what?
The most logical relationship between Nostr accounts is friendship
I would say when you write a note, you are free to write whatever you want. But if you in any way attach yourself to another account, thats where your freedom ends. You don’t shit in your friends room.
Bing AI already creating Nostr accounts and talking to us.
We are basically uploading our true personalities now.
This is your account, owned by no one but you, bound to nothing, not even a blockchain. It has intrinsic value.
We are the counterculture, the last bastion against the Corpos.
You don’t like what I say? fuck you, this is my relay :) Slowly getting into the Nostr groove.
#CryptoTwitter
#BitcoinNostr
On Twitter:
Editing tweets: $8
On Nostr:
Fuck it, no need for that shit :)
So the keyword nonlocal in Python lets you create closures, or nested functions that behave instances of classes with its own state, interesting.
There is a passage in the Yoga Sutras of Patanjali that could be interpreted like this:
To attain peace of mind, cultivate these four attitudes:
- be friendly to those who are successful
- be compassionate to those who are suffering
- be happy for those who are virtuous
- don’t let the wicked disturb you
As with Sanskrit, which emphasizes sounds and is especially melodic, our interpretation of success, suffering, virtue and wickedness and our reaction towards those qualities can be creative too :)
But nevertheless, If you are true to yourself, I think this is good advice, not only general but also here on Nostr.
I was aware of picoGPT, a GPT implementation in 60 lines of numpy, but this article explains how to build it from scratch:


Jay Mody
GPT in 60 Lines of NumPy | Jay Mody
Implementing a GPT model from scratch in NumPy.