Really? Have not noticed anything, but will check metrics at home… And thanks!

FYI I also plan to open source the full source code (frontend, backend, Hetzner build pipeline) to create more trust and grow a community.

Also people could then create new apps for all kind of topics (not necessarily tech related…)

Not yet, but I will create a codeberg repo to share all feeds and also to contribute.

just to not raise wrong expectations: Currently only those two params are available.

Fully customised feeds with your own search queries and more is in the loop. I hope it is ready in the next 2 or 3 weeks. This is still a solo project I work on…

Thanks!

You are welcome! FYI you can also tweak the feed with search params:

• limit: must be between 1 and 20 (default 10)
• threshold: must be between 0 and 1 (default for linux is 0.77) - the larger the more articles you get in the feed (but maybe also unrelated)

I plan to provide fully customizable feeds that every user can configure for their needs.

Who is they?

Why do you think I don’t know how my software works? I exactly explained what is currently possible.

Search of articles is my own implementation with Postgres and pgvector.

I don’t understand why the hostile tone… I did not ask to sign up just mentioned the current possiblities.

Honestly, I did not even know this “-” trick so far. Thanks for mentioning.

And unfortunately this is currently not possible. but a great feature which I will implement soonish.

At the moment you can only block feeds, which you don’t like. But you need to be registered to do so.

Great, feedback is very welcomed.

I plan to create also generic feed generator where you can create your own feed based on a search query and further parameters.

My point is just that by publishing anything online, you feed the AI. Even with our conversation here, we feed the AI. It does not matter if you provide it in a prompt or not. Obviously, I did not mean that it is a good thing.

Regarding not feeding AI, I just want to mention that basically anything that is publicly accessible in the web is used to feed the AI. Crawlers are permanently screening.

No I was not aware of it so far.

Thanks I already added the SwitchedToLinux feeds.

Will let you know once I also added the Brazilian feeds.

I also plan that users can submit feeds on their own in the very near future.

In general yes, but I would need to think how to solve it technically first. I do not want mixed language feeds.

I can imagine that I assign a language to each feed and then the user can simply switch between the languages.

So yes, please send me some sites and I will try to implement a solution in the next couple of days.