krash

Short answer: a lot 😉 its an authentication protocol to have a single identity provider take care of all your users passwords, access rights etc., like those "login with Facebook" buttons.

It's a bit of advanced topic, but a solid way to minimise authentic based alley on username and password.

Just tried it, and liked it. Too bad there isn't support for OICD right now.

I'm really curious about Papra, but don't see the benefits it provides over paperless. Besides, I won't migrate unless there is a tool to brings over the tags, metadata etc...

It's a matter of when, not if, that happens. And in that situation there's headscale but also Netbird, among other services. And of course, there's also just plain wireguard 😏

I wanted to use this on my RPI2 buy I think the CPU is too old 🙃 I to however have a openWRT router and I suppose I can achieve similar functionality with a bit of hacking on the OS.

Welcome to the deep rabbit hole :-) how much do you know about how computers work? In general, you're going to need to understand some basic networking and general Linux administration, but if you already have a grasp on that then I'd say you just need to start small (simple service, aim to have a resilience goal with backups and restoration) and other metrics that motivates you. Perhaps you want to learn something new with every service you host? You decide, this is your hobby :-)

I think for matrix to be usable in a homelab setting, Matrix needs to enable a way to handle these huge data storage with prune or something similar.

I found snikket to be quite decent, give it a whirl.

There are various obstacles to "just forking" a project; it requires times to understand the frameworks / libraries used in the project, understand the code and its different parts and last but not least, have a interest to invest that time and energy (most often, that time could be spent developing your own solution that would fit your usecase better).

As for the stage I was referring to, both the theories of enshittification and rot-economy see software and services going through stages to attract new users, before going in for the profit maximizing.

What's wrong with Ubuntu and RH? Is it because of the snaps / source code debacle? Both of those had solid business cases to them and while I dislike the outcome, I do understand why they made that choice and most importantly - I still appriciate what each company does for FOSS.

My two examples are of OS SaaS that got their plug pulled before they got to that stage. See skiff.com and omnivore.

Awesome <3

If you need feedback, testing etc. on this feature, I'm happy to help. Just pm me and I'll give you my github account.