liqilin1567

Seems like there are many useful suggestions for the author. Here is mine: maybe an interactive style would work much better for educational content.

There is a well praised post on HN: https://www.nan.fyi/database, built with the framework: https://github.com/nandanmen/NotANumber

Someone else also shared a similar experience, so it seems true that we should avoid store critical data in mongodb

https://news.ycombinator.com/item?id=45694376

It's kind of nostalgic and make me curious: Were there many magical things like this in the early internet days.

Just curious - how does React.Native implement native UI if it doesn't use chromium

But I feel like some creativity comes from breaking existing patterns

Is it like this?: It's one of those things you can't really describe - you just feel it

HaHa, it's empty too :)

Thanks for sharing, I didn't even know about this useful feature.

Great points! Yes memory can be a force for trust—by enabling users to verify, correct, and audit past interactions.

It really resonates with me, I often run into this situation when I'm trying to fix a bug with llm: if my first prompt is not good enough, then I end up stuck in a loop where I keep asking llm to refine its solution based on the current context.

The result is llm still doesn't output what I want even after 10 rounds of fixing requests.

so I just start a new session and give llm a well-crafted prompt, and suddenly it produce a great result.

Optimizing at the algorithmic and architectural level rather than relying on language speed

This reminds me of stalwart's spam filter feature claim: "LLM-driven spam filtering and message analysis." :D

https://github.com/stalwartlabs/stalwart

I tired Joplin, Logseq and Appflowy but ended up with siyuan

Thanks, there were few errors after testing.

1. stripping fist X bytes: dd bs=1 skip=X

2. stripping last X bytes: truncate -s -X

One of my biggest headaches is stripping specific number of bytes from the head or tail of a binary file. and I couldn't find any built-in tool for that, so I wrote one in C++.

Manga may come to life with this :D

That means the Safe Browsing abuse could be weaponized against self-hosted services, oh my...

This reminds me of another post where a scammer sent a gmail message containing https://site.google.com/xxx link to trick users into click, but gmail didn't detect the risk.

The framework OP uses for his blog: https://github.com/nandanmen/NotANumber

Off the topic: Why does my chrome become unresponsive every time I collapse a sub-thread on this page with a huge number of comments.

But works fine on small threads though.

Out of curiosity, would it be possible to attach pitch, emotion, tone info as text-based metadata to each word during ASR, so that the asr output retains these metadata?

I like the gradient font style in the overview :)

Thanks, this is very detailed! Could you share some real-world use cases for these tools?

Actually I am looking for tools to automate DevOps and security for self-hosting

Thanks for sharing I think these are very useful suggestions.

Mind sharing these tools and what each one does?

This is my self-hosted list--all of which are hosted on a 1-cpu, 1G RAM VPS :)

1. siyuan - a lightweight note-taking website like obsidian

2. readeck - a lightweight bookmark website scrapes original page to the local server while keeping the original formatting.

3. leantime - a lightweight project management website

I self-host a pretty lightweight bookmark website——readeck, it scrape original page to the local server while keeping the original formatting.

I would like to put the request body into a file and use it with `@file.json`.

> Chunking Strategy: this takes a lot of effort, you'll probably be spending most of your time on it

Could you share more about chunking strategies you used?

sentence embedding models like all-MiniLM-L6-v2 [1], bge-m3 [2]

[1] https://huggingface.co/sentence-transformers/all-MiniLM-L6-v...

[2] https://huggingface.co/BAAI/bge-m3

In my recent project I used openai's embedding model for that because of its convenient api and low cost.