Recent submissions

Comment · 1 month ago
Indeed the terrorists did win, as we can see from these images.
Comment · 2 months ago
This seems like such an unimaginative and cynical claim. You're saying you can't think of any way of putting an LLM in service of developing your creativity and understanding?!
Comment · 2 months ago
sure the electric motor didn't generate text, but it's ridiculous to think it was anywhere near reliable in its early days
Comment · 2 months ago
I've been working on something to help us pursue our long term interests in a reasonably more structured way: https://pond.li. The basic problem/idea motivating it is that it'd be valuable to reach more often into our archives/collections (or have them reach out to the present) than be pulled into the great big, novel firehose of information we're exposed to everyday.

It's also an attempt to make use of LLMs as not primarily chat partners. Keen for feedback if any of this strikes you as interesting

Comment · 3 months ago
I feel similarly. I have a lot of tendencies characteristic of ADHD, but I'm undiagnosed, and I don't think I'll seek out the diagnosis either. I don't intend to take any stimulants, and the particular identity of having ADHD means little to me, so the diagnosis wouldn't change much. As you say, I'd still need to figure out how to live this life in these conditions anyway.
Comment · 3 months ago
Unfortunately it's nearly here thanks to Starlink
Comment · 4 months ago
I'm making https://pond.li to try and make our personal archives of text and other media we engage with more available to the present. It's an interesting design space at least, partly because it requires taking a stance on what meaningful/cognitive agency preserving human-LLM interaction looks like
Comment · 5 months ago
but they did?
Comment · 1 year ago
my side-project is Possible World Wikis: https://www.possibleworldwikis.com/

LLM/procedurally-generated fictional wikis with worldbuilding history/context so the wiki stays coherent. Fun project to make the most of LLM hallucinations

Comment · 1 year ago
Possible World Wikis! https://www.possibleworldwikis.com/

LLM/procedurally-generated fictional wikis with worldbuilding history/context so the wiki stays coherent. Fun project to make the most of LLM hallucinations

Comment · 1 year ago
Sorry I botched that. Fixing now and will post again, thank you for bringing it to my attention!
Comment · 1 year ago
Hey, HN, this was inspired by Borges’ fictional encyclopaedias and the idea (Brian Eno’s I think) that the flaws of a system/tool (hallucinations and LLMs here) are what we come to value over time. Hope you find it somewhat interesting/fun!

Live from the official Hacker News API

A modern reader for Hacker News · SvelteKit + Skeleton