Notices by Wim 🈂️ (wimv12e@cybre.space)

Found this on Twitter ...

"We must put an end to #pleroma before it's too late."

I wrote a little guide about hacking #pleroma

https://wimvanderbauwhede.github.io/articles/hacking-pleroma/

I've been running #pleroma on my #raspberrypi3 for a while now.

About 2 weeks ago I started learning #elixir and #phoenix
Now I've hacked my local #pleroma instance to support bot services: parse an incoming message, take an action, post the result. Elixir is really easy!

Inspired by the Reddit "Place" I created a proof-of-concept of a collaborative canvas. If you like, you can give it a try:
follow @pixelbot
and post some pixels to it. The bot will post the updated canvas.

https://pynq.limited.systems/pixelbot/canvas.html

@deadsuperhero @kelbot I have it running on a Raspberry Pi and it uses about 100M. Mastodon also runs on the Pi but uses almost all memory (1GB)

If you want to contribute to #pleroma:

- Create an account on https://git.pleroma.social/
- Fork the pleroma repo: go to
https://git.pleroma.social/pleroma/pleroma and click on the "Fork" icon in the top bar
- Clone the repository
- Make your changes
- Commit and push your changes
- Create a merge request ("+" icon in top bar -> New merge request)

On a related note to my previous post, two of the top-5 largest instance are company owned (pawoo.net by Pixiv and friends.nico by Dwango, who owns niconico). And the admin for mstdn.jp works for Dwango. So from one angle, the success of Mastodon is already strongly influenced by commercial companies (without these three, instead of the much-quotes one million users, there would be only about 400 thousand). Looking at the bright sight, so far this has not ruined the fediverse.

Yay, I now have #pleroma running on a PYNQ-Z1 FPGA board. It has a 650MHz dual-core Arm Cortex-A9 with 512MB RAM. That is not enough memory to run #mastodon though.