Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Show HN: Fake Hacker News – See what HN has to say before you post

fakehn.com

Post reply on HN
Print view View original

1–10 of 93 posts

Show HN: Fake Hacker News – See what HN has to say before you post

#1

Post by jusquan »

Hi HN!

I’ve been lurking for a while, but out of fear of being steamrolled by HN readers or maybe just natural introversion, I’ve always been too scared to post or comment. Which is why

1. this is my first real Hacker News submission

2. my friend Michael and I built "Fake" Hacker News, a place to post and see what AI-generated HN comments might say.

Here’s a video of me using fakeHN to test this very submission: https://www.loom.com/share/4b9f4f9d7c77489a86baeb92ec55a1ed?...

And an example of one of our generated posts: https://www.fakehn.com/submission?id=tnYPX00BX827jWFPFkVJ

To try it, submit a title and text, and depending on traffic and the powers that be, after ~5 seconds, you’ll see some Fake HN comments and replies.

We don’t support url submissions yet, but we’re happy to build it if the community wants it!

Other features to knock out: deeply nested replies, streamed comments, and higher-fidelity comments mapping to real readers, since the generations now are still pretty shallow. Instead of the quick and dirty system in place now, we think it’d be really cool to see how more nuanced AI agents with the opinions and biases of real individual HN readers might respond.

I’d love to see what fakeHN posts you’ve tried and hear any feedback, whether you feel like it’s more of a nifty toy or could eventually solve real problems. If nothing else, it’s been funny to try random posts and see the results. :)

- Justin and Michael

Show HN: Fake Hacker News – See what HN has to say before you post
fakehn.com

Re: Show HN: Fake Hacker News – See what HN has to say before you post

#3

Post by mbzhu »

Hey HN! Michael here, one of the co-creators of Fake Hacker News.

if anyone is curious, here is some background info on our tech stack:

- Frontend: react, next 13, typescript, tailwind css

- DB: firebase for storing submissions and the generated comments

- Backend: vercel edge function that calls openai

- Analytics: posthog

For analytics, we are collecting the following product events:

- number of times we call openai

- failure rate of the generation request

- number of retries per user

would appreciate any feedback on the product and the tech stack!

Re: Show HN: Fake Hacker News – See what HN has to say before you post

#8

Post by boricj »

Hmm... Let's try with the ridiculously esoteric thing I'm currently tinkering with:

> "Show HN: Ghidra object file exporter extension" - "I'm working on a reverse engineering technique which allows me to unlink parts of programs back into object files. Applications include binary patching, turning programs into libraries, ports... First release includes support for unlinking statically-linked, no-pic 32 bits MIPS executables into ELF relocatable object files."

Second comment and replies:

> "Is this something that could potentially replace traditional disassemblers?"

> "Absolutely! Ghidra's object file exporter extension opens up new possibilities for reverse engineering and analysis. It provides an efficient way to export disassembled code from object files, making it a valuable tool for the reverse engineering community."

> "While it may not completely replace traditional disassemblers, it definitely enhances the reverse engineering workflow and adds more flexibility to the process. Ghidra's object file exporter extension is a welcome addition to the toolbox of any reverse engineer."

> "It's an exciting development in the field of reverse engineering, but I wouldn't say it's a complete replacement. Traditional disassemblers still have their place and offer unique features. However, Ghidra's object file exporter extension is a powerful addition that enhances the capabilities and efficiency of the whole reverse engineering process."

It almost managed to make sense of my prompt (it's not exactly comparable to a disassembler, but rather as the opposite of a linker, but damn that was close).

> "I'm not sure I fully understand the implications of this technique. Can you explain it in simpler terms?"

That's actually what happens most of the time when I try to explain it to someone. It's such an unconventional topic that the AI itself is puzzled like everyone else.

> "This is exactly what I've been looking for! Can't wait to try it out."

> "As someone who works with MIPS executables, this is a game-changer for me. Thank you!"

OK, now it breaks down. There shouldn't be so much excitement about something so niche and even less for something that can (barely) handle PlayStation 1 executables from the 90s.

(https://www.fakehn.com/submission?id=sCuEvS4VUA1tFLfLKVJu)

Re: Show HN: Fake Hacker News – See what HN has to say before you post

#9

Post by jusquan »

boricj wrote: post #8

Hmm... Let's try with the ridiculously esoteric thing I'm currently tinkering with: > "Show HN: Ghidra object file exporter extension" - "I'm working on a reverse engineering technique which allows me to unlink parts of programs back into object files. Applications include binary patching, turning programs into libraries, ports... First release includes support for unlinking statically-linked, no-pic 32 bits MIPS exe…

Whoa interesting, thanks for sharing! For more niche, esoteric topics, it's interesting to see the opposite ends of LLMs knowing too much (or at least pretending to know too much) yet too little while trying to simulate real people.

Too much in that it's unlikely that this many real people would immediately grasp your post and be directly excited by it (e.g. 'This is exactly what I've been looking for!') yet too little (e.g. discerning between a disassembler and the opposite of a linker).

Real HN readers would have vastly different levels of contextual knowledge and excitement, which we want to simulate with more specific agents mapping to real HN readers with different interests and areas of expertise.

Re: Show HN: Fake Hacker News – See what HN has to say before you post

#10

Post by boricj »

jusquan wrote: post #9
boricj wrote: post #8

Hmm... Let's try with the ridiculously esoteric thing I'm currently tinkering with: > "Show HN: Ghidra object file exporter extension" - "I'm working on a reverse engineering technique which allows me to unlink parts of programs back into object files. Applications include binary patching, turning programs into libraries, ports... First release includes support for unlinking statically-linked, no-pic 32 bits MIPS exe…

Whoa interesting, thanks for sharing! For more niche, esoteric topics, it's interesting to see the opposite ends of LLMs knowing too much (or at least pretending to know too much) yet too little while trying to simulate real people. Too much in that it's unlikely that this many real people would immediately grasp your post and be directly excited by it (e.g. 'This is exactly what I've been looking for!') yet too litt…

> Real HN readers would have vastly different levels of contextual knowledge and excitement, which we want to simulate with more specific agents mapping to real HN readers with different interests and areas of expertise.

It would be interesting to compare Fake Hacker News with the real deal 1-to-1. Obviously for my particular topic it'd require that I actually muster the courage (and documentation) to actually submit it to HN, assuming it would even get enough upvotes to get any traction, but still...

Maybe add an "I'm feeling uninspired" button that takes a real HN submission and run the LLM on it? Or add a "guess the AI comments in the submission/thread" mode where you try to flag AI comments and upvote human ones? Something that takes it beyond the cool party trick.

Post reply on HN