hnrss VS hackernews-TUI

That’s interesting.

I have predictive models that can predict if a headline (w/o the rest of the article and not considering the URL) will (a) get more than 10 votes and (b) if it does get more than 10 votes will the votes/comments ratio be more than 2 (which is roughly average)

The first model gets a ROC-AUC (see https://scikit-learn.org/stable/modules/generated/sklearn.me...) in the low 60’s (not good, the second model gets in the low 70’s (actually pretty good though it is a heat seeking missile for clickbait headlines) and my latest content-based recommender for RSS items gets almost 80. (I saw a paper that one system at TikTok gets about 85)

To do all that you need about 10,000 headlines and don’t get a lot of benefit from having more than 100,000. The ceilings on performance have more to do with the nature of the problem rather than my models: the same article can get submitted twice and get 0 votes one time and 200 the other time so it can never be as accurate as “is this an article about galactic astronomy?”

I had it ingest the HN comments firehose and found the amount of articles was overwhelming, my YOShInOn RSS reader now ingests the “best comments” from

https://hnrss.github.io/

together with 110 other feeds and actually I like the comments it picks out a lot. Now that the system is adding about 3000 items per day it might be able to handle a big feed like the comments firehose since now those comments are diluted with so many quality articles. For a problem like that you might want a two-score system with: (i) is it relevant? (something I like) and (ii) is it popular? (like Google’s PageRank)

I think you could make a model that compares comments in the best comments feed with other comments. I have tried formulating the problems above as regression problems where I try to predict the actual score and it does not work well because of the uncertainty problem but formulated as a classification problem for a score over a threshold it is easy to make a well-calibrated model that tells you “this article has a 20% chance of frontpaging” which is about the best anyone can do.

Subscribe via rss, so you can scratch the curiosity itch and each the FOMO, without coming to the site all the time and looking over the same things 20 times?

https://hnrss.github.io/

https://news.ycombinator.com/item?id=9491978

and this https://hnrss.github.io/

ps i’m ok with some % of false positives, but hopefully a sprinkle of OpenAI could keep that magically low?

thanks

It sounds interesting but I use https://hnrss.github.io/

Unless it had most of the features of hnrss.org I would not be able to use it.

Perhaps you could pivot your approach and submit a PR to hnrss for the feature?

Looks like there's an unmerged PR on the third-party hnrss project that would add this: https://github.com/hnrss/hnrss/pull/84

Check out below link to get a more customized, topic wise rss feeds.

https://hnrss.github.io/

Considering that I only access reddit either occasionally from Relay on my phone or primarily modified version of the terminal TUI tuir (They killed the API token for TUIR, but you can still use it if you setup your own API key.) which I'm using now to write this comment from, it would likely reduce my Reddit usage by >90%. I usually have tuir nested in a tmux session, which I would likely change to a hackernews reader like hackernews-TUI.

Hi everyone, this is my second "Show HN" submission posted in Hacker News. The first one was https://github.com/aome510/hackernews-TUI. I received a lot of good feedbacks and suggestions from the community back then. For the second one, I also look forward to hearing the community's opinions.

A bit background on the project: I started `hackernews-tui` and after that `spotify-player` (both are terminal application) because I want to learn Rust and build applications with Rust which I'm able to use daily.

`spotify-player` is a terminal application that can be used as either a remote player to control another Spotify client or a local player with an integrated Spotify client. So if you already know spotify-tui[1] or ncspot[2], `spotify-player` is kinda a simplified combination of both =).

I have two demo videos for the application, one in youtube https://www.youtube.com/shorts/Jbfe9GLNWbA and the another in asciicast https://asciinema.org/a/446913.

Hope you guys give it a try. Any feedbacks are highly appreciated!

[1]: https://github.com/Rigellute/spotify-tui

-hackernews, https://github.com/aome510/hackernews-TUI/

> something that cuts out all layout, formatting and images and shows me the raw article text in a fixed with font.

FYI, I have implemented a reader view for `hackernews_tui v0.6.0` [0] which seems to satisfy most of the conditions above. Judging from my experience, this reader view works quite well and can cover many use cases.

[0]: https://github.com/aome510/hackernews-TUI/releases/tag/v0.6....