←back to thread

Show HN: News Minimalist – News ranked by significance

(www.newsminimalist.com)
84 points yakhinvadim | 2 comments | 15 Jan 25 20:35 UTC | HN request time: 0.001s | source

Hey HN! I'm the author of News Minimalist — a news aggregator where all news is ranked by significance on a scale from 0 to 10.

The project was born out of personal pain — I wanted a way to read only significant news, like major humanity milestones, or historical political events, filtering out all the celebrity gossip and smartphone releases. But I couldn't find a way to do that — everywhere I looked, the news was ranked by popularity, coverage, or relevance, not significance.

I first tried to solve the problem in the beginning of 2023 with GPT-3 (the top model at that time) by asking it to estimate the significance of some news stories. The results were painfully bad — for some reason, the model preferred tragic, personal stories, completely missing the essence of what makes the news significant. No amount of prompt engineering could fix that.

But it all changed in March 2023 when GPT-4 came out. The scores it gave made much more sense. After a month of work, the first version was ready. News Minimalist had its first successful Hacker News post (https://news.ycombinator.com/item?id=35795388), and I realized that a lot of people had the same problem I had.

I've been working on improving the project ever since. As probably most tech founders, I spent too much time on technical improvements, completely ignoring marketing. But I think that work paid off, and I'm finally satisfied with the scores it gives.

The results are posted on the site: https://www.newsminimalist.com/

Let me know what you think!

Vadim

1. yaj54 ◴[17 Jan 25 00:37 UTC] No.42732755[source]▶
Nice work. Subscribed.

I had a very similar idea a while back. I wanted to rank news by "impact" which might be more concrete than "significance."

For an LLM prompt, it would be something like:

"estimate the number of people who's lives that will be materially changed by this news." and "estimate the average degree of change for those impacted."

Then impact is roughly the product of those two.

Additionally, I want a version that is tailored to me specifically "estimate the degree of change this will have on my life." + context of my life.

Tangentially, I've found that getting ratings out LLMs works better when I can give all options and request relative ratings. If I ask for rankings individually I get different and less good results. Not enough context length to rate all news from all time in one go though. Any thoughts on that? Maybe providing some benchmark ratings with each request could help? Something I'm exploring.

replies(1): >>42732856 #
2. yakhinvadim ◴[17 Jan 25 00:54 UTC] No.42732856[source]▶
What you're describing is super close to the first version I had!

In the beginning I had 3 parameters: scale (number of people), magnitude (degree of change for those impacted) and additionally potential (how likely is this event to trigger downstream significant events).

The point behind including potential was to separate these two events:

1) A 80 year old dies from cancer 2) An 80 year old dies from a new virus called COVID

This worked roughly well but I kept adding parameters to improve the system: novelty, credibility, etc... The current system works on 7 parameters.

---

I never attempted to give LLM all options and rank them against each other.

1) as you said, for me 20k articles is just too much to fit into context window. Maybe some modern LLMs can handle it, but it wasn't the case for a long time, and I settled on current approach.

2) I don't want the "neighbors" to affect individual article ratings. With the current system I am able to compare news spread over months, because they were all rated using the same prompt.

3) I intentionally avoided giving AI examples, like "evaluate event X given that event Y is 7/10". I want it to give scores with a "clear mind" and not be "primed" to my arbitrary examples.