@jerrej understandable; i too am paranoid.

but if you feel like looking into it check the #Fedialgo hashtag and/or search posts for fedialgo, you'll see other people vouching for it. someone even made an awesome video about it.

For those curious in the answer - #FediAlgo seems to be the closest thing to what I was looking for, and it's very cool!

https://github.com/michelcrypt4d4mus/fedialgo

Demo app: https://michelcrypt4d4mus.github.io/fedialgo_demo_app_foryoufeed/

3/X “But there are algorithmic feeds for Mastodon?” — Yes, great tools exist, for example:

➡️ #BYOTA by @mala
➡️ #FediView by @adamghill
➡️ #Fedialgo by @pkreissel

.. but most are client-side. They can rerank your existing timeline, but can't make #recommendations from outside your follower bubble.

@bonn.digital Hier hatte ich schon mal ein bisschen ausführlicher zu dem Thema geschrieben, ging da um das coole Projekt #FediAlgo .

https://mainz.social/@fabian/115100920303166577

p.s. @paige the "resource consumption" concern is grade A concern trolling. #FediAlgo is highly optimized. my profile is only 34 MB of storage in my browser for months of algorithm and ~2MB of code for the app.

other than the small wait at the beginning of loading the app (which could indeed be improved if it were running on the server side) there are virtually no resource constraints.

also #FediAlgo is published as a javascript package that serves up augmented data objects that are the same as what is returned by the Mastodon API so any mastodon web app (e.g. #Phanpy etc) should be able to pretty much just install it.

i think theoretically it could even be installed server side as-is because the main mastodon web app is ruby + javascript, though i'm not exactly sure if that could be done seamlessly / without some tweaks.

#Mastodon