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

Show HN: ChatGPT UI for rabbit holes

delve.a9.io

Post reply on HN
Print view View original

161–170 of 171 posts

Re: Show HN: ChatGPT UI for rabbit holes

#161

Post by djeastm »

spdustin wrote: post #3

If "delve" was meant to be an in-joke, I just wanted you to know: I got it. I also have a Custom GPT "AutoExpert (Chat)" [0] that several reviewers have called "the perfect Rabbit Hole GPT" due to the way it leads users through learning a topic. You might dig it, especially since free tier users have access to these now. [0]: https://chatgpt.com/g/g-LQHhJCXhW-autoexpert-chat

I like the whole approach this takes, but man I really do wonder about "learning" through something that is so eager to hallucinate. The vast majority is correct, but I know with my luck I'd unknowingly remember something that was hallucinated and some real expert would look at me funny if I mentioned it.

Re: Show HN: ChatGPT UI for rabbit holes

#162

Post by glonq »

ivanhoe wrote: post #120

I really like this. How about enabling user to select any piece of text and use it the way the links work now? For instance, I've noticed it doesn't linkify peoples' names, and one thing I love on Wikipedia is that you can easily lookup people mentioned in the text. So, rather than having thousand links in the page, it would be handy if I could just select the name in the text, click some button (or right-click menu…

Yeah I asked it about 8-bit Atari computers and it mentioned that they had MOS 6502 CPU's but did not linkify that part, which would have been where I wanted to go down the rabbit hole next.

Re: Show HN: ChatGPT UI for rabbit holes

#164

Post by richardgill88 »

wiremine wrote: post #78

Taking a step back: The UX/UI for LLMs in general are very immature. We're in the very early days of to best interact with these tools. We need more experimentation like this to help figure out what works, and doesn't work. Kudos!

Totally agree. It feels like we've barely scratched the surface.

I'm working on a project where I'm experimenting with pretty obvious things like removing the annoying markdown syntax all the LLMs show you for a split second and smoothing out how it renders the characters to match your frame rate.

https://llm-ui.com/

Re: Show HN: ChatGPT UI for rabbit holes

#165

Post by maxutility »

It would be great to implement a browser extension that lets you highlight a term or phrase on any webpage and open a GPT rabbit hole for that term or phrase.

@maxkreiger - if I were to build one as a proof of concept, would you object to me having it hyperlink to your UI?

Re: Show HN: ChatGPT UI for rabbit holes

#167

Post by froggit »

richardgill88 wrote: post #164
wiremine wrote: post #78

Taking a step back: The UX/UI for LLMs in general are very immature. We're in the very early days of to best interact with these tools. We need more experimentation like this to help figure out what works, and doesn't work. Kudos!

Totally agree. It feels like we've barely scratched the surface. I'm working on a project where I'm experimenting with pretty obvious things like removing the annoying markdown syntax all the LLMs show you for a split second and smoothing out how it renders the characters to match your frame rate. https://llm-ui.com/

> I'm working on a project where I'm experimenting with pretty obvious things like removing the annoying markdown syntax all the LLMs show you for a split second and smoothing out how it renders the characters to match your frame rate.

> https://llm-ui.com/

I have a suggestion for the page you linked. The sample on the top page that shows it doing its thing, while it is useful seeing it in action, as a result of looping the example output it was a bit difficult to read. I would suggest extending the amount of time the loop lasts to give slow readers a little more time. An extra 3-5 seconds after it finishes output would be helpful from a mobile UX viewpoint.

Post reply on HN