News, changelogs and documentation

Version 1.15 released — feed discovery improved!

I'm glad to announce the release of Feeds Fun version 1.15!

What's New

In this update, RSS/ATOM feed discovery functionality has been significantly improved. Now, Feed Fun sees website feeds more accurately, no matter how the site owners mark them in HTML.

Stay Connected

Short summary of the last two months of Feeds Fun

It's been a productive two months for Feeds Fun!

I don't want to bother you with a post for every release (there have been 15 since the last post). Instead, I'll periodically post summaries of the most significant updates. I hope this approach is more convenient for everyone. If you're interested in every release, you can always subscribe to the repository or check the CHANGELOG.

The current version of Feeds Fun is 1.14.2.

Here are the most significant improvements from the last two months:

  1. The time to load the latest news has been reduced by a factor of 4!
  2. LLM prompts have been improved to produce fewer tags of better quality.
  3. Fixed an issue where, after removing a feed on the Feeds page, the feeds list was not updated.
  4. Added tracking of some user activity events.

A Few Words About Tracking Events

Event tracking is essential to better understand how users interact with the reader, how they use it, where they face difficulties, and more. This is particularly important for Feeds Fun because much of its functionality is experimental, and it's easy to head in the wrong direction — such as implementing an interface no one needs or prioritizing the wrong feature for improvement. Tracking these events will help deliver the right features more quickly.

All new events are strictly for internal use:

  • None of the events leave the Feeds Fun server.
  • None of the events are directly tied to email addresses or other personal information.

Version 1.12 released!

I'm excited to announce the release of Feeds Fun version 1.12!

This update brings a lot of cool features and improvements.

Important: Due to significant changes, all user settings, including API keys, have been reset. Please check and update your settings.

What's New

Improved collections interface and logic

  • You can now subscribe to individual feeds from a collection or to the entire collection.
  • Feeds from collections are marked in your feed list for easy recognition.
  • Each feed in a collection comes with a detailed description.
  • Best of all, all feeds in collections are tagged for free for all users!

The old collections (gamedev, artificial intelligence) have been removed because they didn't meet the new quality standards. From now more care will be put into creating collections.

The first new collection contains all scientific papers from arXiv! Now, you can always stay up to date with the latest scientific research!

Gemini support

Support for the Gemini API, Google's alternative to OpenAI's ChatGPT, has been added. You can now use Gemini for tagging in addition to OpenAI's services. This is especially exciting because Gemini offers a free tier with a limited number of requests, so you can test all the features of Feeds Fun without any extra cost.

You can set your OpenAI or Gemini API key in the settings section of the Feeds Fun website.

If you're self-hosting Feeds Fun, please see the additional notes in the CHANGELOG.

Reduced number of duplicated news entries

Sometimes, users can see the same news item multiple times in the list of recent news. That is because the same news item can appear in different feeds from the same website. For example, that can happen for Reddit and arXiv feeds.

Now Feeds Fun will do a better job of detecting and removing such duplicates.

Thank you for your support! I hope you enjoy the new features!

Stay Connected

Hello World, It's Feeds Fun!

Hello Everyone!

My name is Aliaksei, or Tiendil on the web. This is the first post in the blog of my pet project feeds.fun.

Feeds Fun — is an open source web based news reader specialized in organizing extensive streams of news content. Currently, by automatically assigning tags to news items and scoring them by rules you create. In the future, I'll research and implement more approaches.

I created the reader because I subscribe to a lot of news feeds (~500 at the moment) that generate 500-1000 news stories per day. It is impossible to read even all the titles in a reasonable time, and I did not find a satisfactory solution to organize this flow. So, I decided to create my own. The recent progress in neural networks and LLM, especially, was a great help in this.

How Feeds Fun works:

  • You subscribe to news feeds: RSS, Atom, etc.
  • Reader automatically assigns tags to each news item according to its content.
  • You create rules to score news by tags.
  • Filter and sort news how you want ⇒ read only what you want.

Feeds Fun helps me to save 4-8 hours per week, and I hope it will help you, too.

In this blog, I'll post documentation, news about significant project updates, best practices for using it, and other related topics.

So, subscribe to the RSS if you want to follow the news.

Also, feel free to join Discord and Reddit.