Pjotr Prins

Public Key

npub12ddjdwkds3anfkjlppy0ey4f5f5r75tu0r0m083azxv5zwhwn3psetde26

NIP-05 Address

Profile Code

nprofile1qqs9xkexhtxcg7e5mf0ssj8uj256y6pl29783hahnc73rx2p8thfcscpz4mhxue69uhhyetvv9ujumn0wd68ytnwv46qzrthwden5te0dehhxtnvdakqz9rhwden5te0wfjkccte9ejxzmt4wvhxjmcfchs3d

Publishing to

relay.nostr.net

nos.lol

relay.damus.io

Author Public Key

npub12ddjdwkds3anfkjlppy0ey4f5f5r75tu0r0m083azxv5zwhwn3psetde26

Show more details

Last Notes

2026-07-09 18:30:01 UTC

- reply

@nprofile…wfat Angkor Wat

2026-06-13 16:42:24 UTC

- reply

@nprofile…wfat @nprofile…j497 @nprofile…dmr0 @nprofile…x7h0 Thanks Andrew for digging in. I also had the same reaction - and suspected LLM writing.

2026-05-28 06:09:44 UTC

- reply

@nprofile…wfat I wrote a program in zig two years ago. It is interesting. But I found its syntax not so natural. That is why I keep looking for something better. prescheme would have cut it if it had matured. guile -> wasm -> native could be interesting without GC. Right now I am not coding performance code, but when the time comes I would like something schemy without GC.

2026-05-22 16:19:54 UTC

- reply

@nprofile…dmr0 I appreciate the sentiment and I admire your heroic stance. But even Debian postponed after 'debate'. I admire Debian - they rarely get it wrong. But the point I am making is that we are not really having a debate. Maybe we should save it for guix days? Though I am not looking forward to that. Remembering RMS.
https://agent-wars.com/news/2026-03-14-debian-punts-on-ai-generated-code-policy-after-contentious-debate

2026-05-22 13:10:15 UTC

I think we should drop the GCD008 idea. The differences are mostly ideological and political. I think we should accept that (at this stage) we have opposing camps and a clearly divided but quiet community. Supporting one group over the other is a bad idea. We can afford to wait for the rest of the free software world to take position. No need to try to lead with a minor distro. We should be more humble.

2026-05-06 11:25:40 UTC

- reply

@nprofile…dmr0 if that argument holds then Guix should stop downloading 50Gb per machine per annum ;)

2025-08-16 17:49:50 UTC

- reply

@nprofile…jrff unfortunately it is the norm. Very hard to get accepted because of overwhelming submissions. Having it out there is the main thing :). You can get it published if you want it to be a peer reviewed paper. I don't doubt.

2025-07-17 10:50:24 UTC

- reply

@nprofile…jrff exciting progress. People can join on the ML https://lists.sr.ht/~abcdw/rde-discuss

2025-06-01 18:23:54 UTC

- reply

@nprofile…jrff Great! Did you blog somewhere about how you are using the guile REPL and manage stack traces? That would be awesome :)

2025-04-20 17:07:12 UTC

- reply

@nprofile…jmss @nprofile…gmfj I have used a single org todo almost forever(!) Having over 5000 todos gives a sense of peace because you can never achieve it :)

2025-02-28 16:16:17 UTC

- reply

@nprofile…33hd that is amazing news. Anything that runs apptainers will work?

2024-02-23 07:52:02 UTC

Guix, Mes and Hurd proposals can be submitted for GSoC!
https://lists.gnu.org/archive/html/guix-devel/2024-02/msg00280.html

2024-02-06 17:55:05 UTC

- reply

Amazing writeup on designing low-end hardware and making it run!

2023-12-01 12:16:04 UTC

- reply

@npub1gfd…7gg6 I have been mentioning Turbo Pascal in recent talks: why do I have to wait for a C++ or Rust compiler building today when we got immediate results in the 80s? I don't want to be waiting for a compiler prompt today. Zig is today's Turbo Pascal. It is a pleasantly fast and modern compiler. #zig

2023-11-28 17:09:41 UTC

Web iterface for finding packages, versions, also on external channels. For Zig: https://toys.whereis.みんな/?search=zig @zig @guix @npub15zj…ydka @npub1zpw…6m3s