rriley

The biggest gap in this paper is the condition they didn't test: Skills built through human-AI collaboration. They found fully self-generated Skills are useless (-1.3pp) and human-curated ones help a lot (+16.2pp), but that's a false dichotomy. In practice, especially in tools like OpenClaw, skills will emerge iteratively: the AI drafts procedural knowledge while solving a real problem, the human refines it with domain expertise. Neither produces the same artifact alone. The +16.2pp from curated Skills is likely the floor for this approach, not the ceiling. Would love to see a fourth condition.

I was going through my low-code bookmarks and I found this jewel from November 2021 (1 year BC-GPT = Before ChatGPT):

Low-Code and the Democratization of Programming: Rethinking Where Programming Is Headed

https://www.oreilly.com/radar/low-code-and-the-democratizati...

Making complex simple again.

https://unrav.io

Niiice! I finally understand the origin of this JS syntax used in SQL queries and GraphQL:

sql`SELECT * FROM users WHERE id = ${userId}`

const q = gql` query GetUser { user(id: ${userId}) { name email } } `;

We’ve optimized the internet for producing information, not for humans consuming it. Most of us are overwhelmed not because content is bad, but because it’s all delivered in the same rigid format.

I’m working on unrav.io : a way to reshape any web content (article, video, or PDF) into the form that actually fits how you think (summaries, mindmaps, infographics, podcasts, chat, etc.).

We just launched a Chrome extension, so it’s one click on any page. No login, free to try.

https://unrav.io

Great compilation. The ".cgi" in the URL clearly tells me this is an old collection of links :-)

Another fun esoteric music language missing in the comments is ORCA: https://git.sr.ht/~rabbits/orca

https://unrav.io . Lets users reshape any article, paper, or video into the form that actually helps them think, mind-maps, summaries, podcasts, or interactive Q&A. Launched as part of the Bolt.new hackathon in August and growing steadily. Going from a 100% vibe coded web app to a full production system has been quite a ride!

Just released the browser extension for https://unrav.io . It transforms any article, paper, or YouTube video into your perfect view (infographic, TL;DR, mind map, podcast, etc) with one click.

It’s for people who feel smart but overwhelmed, drowning in tabs, skimming everything, remembering nothing.

You don’t need more information. You need clarity.

Great! Now I desperately need this Stirling engine for my morning coffee: https://a.co/d/6Ja2LeF

Video of how it works: https://www.youtube.com/watch?v=Q5QEBqjkNjo

Everyone’s excited about Nano Banana Pro’s infographic skills, so here's what I've been building: turn any link into a crisp, auto-generated infographic instantly.

Curious what the HN crowd thinks and what you'd want next.

You can also install the extension/bookmarklet and transform any online content instantly to an infographic, a mindmap, a conceptual art piece, or any other format that you prefer. One click away!

https://unrav.io

Nice work. I am bookmarking this. Looks really good!

I’m building https://unrav.io : A tool to fight information overload.

It lets you turn any article, YouTube video, or PDF into summaries, mindmaps, podcasts, chat conversations or infographics that match how you learn with just one click.

We just launched this week the Chrome extension so you can do all this in one click on any page, no login needed (with generous freemium usage).

https://unrav.io https://chromewebstore.google.com/detail/unravio/mbnapibcjcf...

Would love feedback from fellow builders.

Good point :-)

The study introduces the "LLM Brain Rot Hypothesis," asserting that large language models (LLMs) experience cognitive decline when continuously exposed to low-quality, engaging content, such as sensationalized social media posts. This decline, evident in diminished reasoning, long-context understanding, and ethical norms, highlights the critical need for careful data curation and quality control in LLM training. The findings suggest that standard mitigation strategies are insufficient, urging stakeholders to implement routine cognitive health assessments to maintain LLM effectiveness over time.

TL;DR from https://unrav.io/#view/8f20da5f8205c54b5802c2b623702569

I am mostly concerned with the irreversibility part. More developed brains probably would not be affected as much.

This paper makes me wonder the long lasting effects of the current media consumption patterns by the alpha-gen kids.

Everyone’s drowning in long articles, dense PDFs, and hour-long videos. I’m working on https://unrav.io , it lets you flip any article, paper, or YouTube link into the format you actually want (summary, mindmap, podcast, infographic, etc.) in one click.

Right now I’m experimenting with a simple bookmarklet trigger instead of a browser extension. Curious: how do HN folks feel about bookmarklets in 2025, still viable, or do you prefer extensions?

I’ve been exploring the theme of simplification, how people throughout history have managed to strip complexity down to its essence. Along the way, I collected 111 quotes, principles, and patterns on the art of simplification.

As an experiment, I also built interactive "digital twins" of some of history’s greatest simplifiers. You can chat with them to see how their thinking might apply to your own problems.

It’s a free project, just something I’ve been tinkering with, and I thought folks here might appreciate it. Curious what you think, and whether there are other "simplifiers" worth adding.

Thanks! It's a tribute to Rick Rubin's reference to this new era "punk rock development" \m/

Love that you were inspired by "The Elements of Style". I scanned through the pages of the PDF and looks very promising. Concise and full of great advice. Thanks for sharing this with the world!

I tried it with my recent launch of https://unrav.io and the suggestions were fantastic. I tried to do similar research myself but never got the great results your platform generates. I am really impressed!

Google Earth Timelapse is a great tool to visualize our world in different scales of space and time: https://earthengine.google.com/timelapse/

This reminded me of The Overstory by Richard Powers, once you start seeing the world on the time and spatial scales of trees, everything shifts. Human timelines feel like flickers. It’s not just a change in perspective; it’s a change in what even counts as meaningful.

It's a kind of status jiu-jitsu: the less you want, the more others want you.

Good summary of this paper: https://unrav.io/#view/f1604fd7b327f48a9920f4d2561b9626

This is really well done!

I love that I can start using it right away without needing to sign up, that’s becoming a rare (but refreshing) design choice. Also big thumbs up for the PWA support, being able to install it on my desktop instantly is a huge plus. The presets and project tracking are super handy too.

Congrats on launching your first public side project! it’s clear you built it with care and actual daily use in mind.

roleplayr.ai : a browser extension that lets you talk to any image on the web.

An LLM answers in-character; Try it free: https://roleplayr.ai

Would love feedback.

This is incredibly impressive work, especially at 19! The insight about LLMs being better at critiquing than producing translations is brilliant, and using that to build a hybrid system that outperforms individual models is exactly the kind of innovative thinking the field needs. Congrats on the company talk, well deserved recognition for solid research and execution!

This is absolutely fantastic! What a treasure trove you're building. I love that you're using AI to help with the metadata extraction while still requiring the physical magazines, that's such a thoughtful approach to preserving these cultural artifacts. That 1968 computer ad that "answers riddles" is pure gold. Can't wait to see more gems as your collection grows!

Very promising concept! Any link/video/website you can share to learn more about this idea?

Impressive!