paweladamczuk

- Navidrome

- Jellyfin

- Vaultwarden

- Immich

- pi agent. I use it through tmux and iSH on my iPhone

- librechat. nice UI but honestly I probably jsut use pi more often

- copyparty. to visually inspect whatever I create with pi, and to access my documents, to upload files I need on the server quickly and easily from any device

The back gesture works indeed. I didn't figure it out. Thank you!

Same here, I can't believe there is no way to go back. I have to s of things around my house that I can just add because I live here, but I don't remember everything, and once I hit a question that I cannot answer from memory, I am stuck.

The image of a shaman I had had in my head was heavily romanticised. The one in the tribe who transcends what is ordinary, the one who explores the world of spirits so that others don't have to, the courageous one, the sacrificial one, the ultimate poet, the ultimate cleric.

Not sure if this is specific to mestizo culture, but the shaman that the book describes looks to me much more like a superstitious, opportunistic charlatan than a clerical psychonaut.

Gordon Wasson did not reveal to the west the idea about ethneogens that Maria Sabina represented. He came to her with a preconceived goal of looking for god, which was not what she was using the mushrooms for.

Anyway, my point is that shamanism looks different through the lens of rigorous ethnography than it when conveyed to us by the influencial popcultural figures. I won't be able to explain it well in a comment, but I highly recommend the book, even if you only read fragments.

"Singing to Plants" by Stephan V. Beyer.

The description of the actual world of shamanism leaves me feeling massively disillusioned. McKenna et al. did us a disservice by shaping the western popcultural idea of what indigenous shamanism is.

I think the same. I appreciate the effort put into research, but the AI tone is unbearable to me.

I want to ask instead what stone you use and where/how did you learn to sharpen the knife?

This is really lovely. I think it's pretty obvious that it bases on sources like Internet Archive and the Library of Congress, but it does add value.

I just wonder if it can survive - is there a business model here? If not, it's just a tech demo and then it doesn't have value as the hosted version, it would have it if the whole processing/indexing pipeline was open sourced so that anyone can host it themselves and pay for their own infrastructure.

It increasingly seems to me like the only way to prevent value to be extracted from myself is to stop engaging with new tech altogether.

This is a frustration of mine as well, but I vaguely remember someone showing something on HN where the file could essentially save itself, I think they were using some file APIs for that... sadly I can't find the details anymore. But I keep running into the "HTML file opened locally can't update itself" issue repeatedly now that I build tools for myself with LLM agents.

They want to own the glass (the one in our pockets) so much.

Good thing they never claimed their feelings were proof of anything.

I remember one of the disputes an LLM helped me with. The human beings on the other end threatened me with mall security and I had to record them (legal where I live), write to the company's HQ and to the consumer ombudsman to get a refund honored. In the end they had to concede. I would have given up before LLMs: finding all the relevant case law and statutes cost 10 times as much effort back then.

I have a feeling that some people were sheltered by the system in their belief that they don't have to take responsibility for what they sign with their own name. I had zero issues with LLM writing so far, I think it's because I always pay lots of attention to what I sign with my own name, no matter how exactly it was written. People like the person you describe, ones willing to apply the tech mindlessly, get found out. Maybe it's ultimately for the better.

LLMs are a great equalizer. I had no idea how much of the world runs on ignorance of one's own rights and how much administrative burden is used adversarially. Now my learning is greatly accelerated. I hope it's the same for more people than not and we will end up with a more just society.

Funny, I had a similar task for an LLM recently but it succeeded, so I'll describe it:

I asked the chatbot (librechat, tavily-mcp connected, kimi-k3) if anybody tests dish soaps rigorously. It said yes. I asked it to find the consumer reports from Germany. It did so, full access is paywalled but it did see the top positions. The I asked it what term to put in evay.de if I want to order. It gave me a string that found the auctions for me. I also asked it what the actual difference is and it described what info the EU law forces onto the label and told me what to look for in the ingredient lists.

The process is still involved. I wish I could just do it all by saying "find me another dishsoap, this one sucks". But then would everybody saying this phrase except the same approach to the issue that I took?

How would the plausible deniability piece work? If you stream some bytes to someone else, you can't deny doing it. Unless you imagine this working over Tor, idk what the performance of Tor is these days but I never thought it could support such an application.

As if the people utilizing the judiciary system before LLMs fully understood the laws themselves.

Give me a break. These are the talking points of the elite scared that the playing field is being leveled by technology. I'm on the side of technology on this one.

I would like everybody to be completely responsible for the words they sign with their own name and present as their own. Regardless of how the words were produced.

> applicants to the Fair Work Commission will need to disclose AI use

I really don't like where this is going.

When technically correct PRs (actual fixes, harmless) get rejected, maybe it's a signal that the codebases are too big?

My big hope for LLM coding is that it will finally force us to focus on minimizing the amount of code that is out there instead of ever producing more.

Care to share your setup?

I use pi and librechat and I am really limited by the lack of proper web connector. Testing Tavily right now, Kagi is next.

I wonder why I see so few comments about it on HN, does everybody have this figured out already? Am I a dummy? Am I severely overstating how important of a frature web access and search is for LLM use?

Not him but I can as well throw in my setup.

For playback, I mainly use Navidrome.

For autotagging and importing (moving files), Beets.

For tagging edge cases, MusicBrainz Picard as the GUI and the Python tagging lib underneath it (mutagen iirc) with an LLM inside the pi cli agent for automation.

For format conversion and copying to portable devices, ffmpeg in custom bash scripts managed by an LLM.

I noticed the same. Housework consumes a LOT of my time. But maybe that's just because of income level and failure to delegate?

Yeah, straight out denial of somebody's experience is another debate killer.

FWIW, no matter the issue, this kind of tone makes me close up for any rational arguments. I wish I could look past it, but I just can't.

> “Under German law scanning books – regardless of the purpose – would not be permissible and constitute a clear violation of copyright law,”

Excuse me, what?

Or maybe it's a case of "I want to own the place where I live and the only way to achieve that while supporting a family is participating in the economy based o hostileOS"?

It's easy to have a principled stance when the only currency it costs you is your own convenience and standard of living. For many people, life is not as simple as that.

There is an odd assumption behind this post that you can put effective guardrails on a model before releasing it. I thought nobody had such technology?

The amount of engagement this gets... guys, I think we are being played here.

On some days, half of my promts to Copilot at work are spoken to a local whisper medium model running on an Intel ARC GPU.