The same Zuck that is training “AI” off their platform’s user data?

Proxmox sounds like it fits their use case , it’s a useful and tweakable solution, and because it’s based on KVM you can pass through hardware with IOMMU. Personally, I run Proxmox on my (admittedly not very good) home server with like 12 gigs of ram and a processor from the early 2010s, handles a few VMs just fine with hardware passthrough to a TrueNAS VM. I do run a lot of my micro services on some cheap thin clients (DNS mainly) for redundancy as I mentioned, they were cheap. Home Assistant OS is happy on Proxmox as is Jellyfin with hardware acceleration.

but but muh thermodynamics

there are lots of cheap domain registrar options but if you’re looking for a cheap .com I always go for cloudflare, they also offer .org for pretty cheap and many other options as well. The domains other than the common ones are pretty decently priced as well. I migrated all my domains there last year and it’s really simple, integrates with their DNS really well and payment is pretty streamlined.

administrative overhead costs

The fundamental component of the Institute on Race and Political Economy’s plan is expanding the EITC, or Earned Income Tax Credit, an already existent program. Implementing MTBI through the EITC doesn’t increase costs anymore than a UBI would as the internal infrastructure already exists in the IRS. If you were to implement a UBI without the EITC, you’d either have to create an entirely new program through the Treasury Department or otherwise, and be able to find every person in the US to pay them with cash or cheque. That doesn’t sound like more administrative overhead? Maybe I’m biased because I particularly like the idea of an MTBI but just the implementation of a UBI sounds more of a practical nightmare than MT.

any stigma from receiving it

Cremer & Roeder, '15 suggest that a means tested system will have comparable stigma to other existing programs such as SNAP, which is high, but in the US political climate, there will be more support for a means tested system, and “political economy considerations do not appear to justify a universal system.” Although there is still a stigma associated, the net benefit of having political backing that’s miles ahead of a UBI makes it a much more realistic plan to pass in current day.

flawed at the foundation

I have given examples in other comments showing that MT works, mainly the Stockton trial, but I’m more than happy to provide empirical studies from other countries implementing MT on a larger scale.

See my other reply and if you want to argue please reply there, but TL;DR, means testing is basic income and arguing about the qualifications of a basic income doesn’t help when it comes down to whether or not we should implement one.

Literature on the topic suggests otherwise. I said earlier I’ve debated on this topic and so I know what I’m talking about to an extent. According to David A. Green et al. In 2021 from the Vancouver school of Economics, “[…] there are also many alternative designs. The alternatives can be viewed from two perspectives, related to placing conditions on the payments. The first type of conditionality is related to whether the basic income applies to everyone […] or to a specific group of people.” In the end the definition of basic income doesn’t come down to economic theory but what we can agree on, and by saying MT ‘is not basic income’ doesn’t help to implement any kind of BI.

Source here

The broader US doesn’t have a means tested program though, sure you could argue that programs like SNAP etc are MT but they aren’t BI programs. According to the LISC Institute for Community Power in 2022, a lot of guaranteed income pilots in the US are targeted to certain groups, or means tested, and show “extra funds are typically spent on food, health care, paying down debt and household needs. Full-time employment among recipients actually increased[…]” This is data from the Stockton pilot, but you can read more from the full source here

A means tested basic income is a type of BI that, as proposed by the institute on race and political economy in the US, expands the Earned Income Tax Credit program to include those who aren’t earning an income, providing every adult in the country up to $12,500 per year calculated on a sliding scale based on income, as well as up to $4,500 per child. These numbers are as of 2021 so they could’ve changed by now, but basically it gives everyone a certain amount of money if they are below the poverty line (calculated by their current income), to lift them above the poverty line and keep them out of poverty.

It’s more, I guess you could call it a niche, type of basic income so it’s on me for not explaining it, just used to everyone in our debate season already knowing what it is lol, sorry.

I dislike UBI but not because I’m not for a basic income, I just think Means Testing would be better. I’ve said this before but now after being the runner-up in my state for debating on this topic I feel more confident talking about it. Ultimately there are many ways of implementing fiscal redistribution but means testing is substantially cheaper than a full UBI (especially in countries with higher populations, e.g. US), while also providing social utility and enabling recipients of the basic income to have more resources. Not only is MT better from this standpoint but a UBI can also worsen inflation by increasing the dollar’s velocity (1 dollar changes hands more). I won’t deny that most people could use money, especially right now, but a UBI is not the best approach in my mind because of these reasons. Of course I am still in highschool, am not an economics expert, and MT was the plan that we ran in tournament so I’m a bit biased.

ETA: This is all keeping in mind the current political and economic climate of the United States, where realistically neither of these plans will pass but I believe MT has more merit to being passed compared to UBI. If you’d like any sources on what I’ve said I’d be happy to share!

Some people seem to hate on it, but I love Docker, it works well for what it has to do and has relatively low overhead as far as I can tell. I personally virtualize a Debian server on Proxmox for my containers just so as to keep everything even more compartmentalized, but it takes more work than it’s worth to set up.

And if you don’t like Docker for whatever reason, you can also try Podman which is API compatible with Docker for the most part.

That’s a fair point, I self host stuff more out of convenience over privacy (although that’s still a factor) so I guess I just care less about them watching my traffic I suppose. CF is just so easy with their Argo Tunnels and domain registrar service.

I use their tunnels in conjunction with internal split horizon DNS so I don’t have to forward any ports and can access things locally faster so I’m probably breaking this rule but I haven’t gotten any emails or letters about it yet. Crossing my fingers they don’t care lol

It’s been a while since I’ve heard anything about this but didn’t they change their ToS in regards to media on their network? I thought I read something about that clause getting removed at some point a few months back.

ARM software support is just generally rough, yeah it’s good on RPi (and Mac) but on other boards it typically sucks, namely the cheaper boards OP would be buying. Here’s a couple software examples though, I’m a big docker user and just the other day I was trying to run I believe Mastodon and Lemmy on an ARM device but there was just no image for it. I’m sure I could build an image myself but for someone just getting into Homelabbing (like OP), x86 is the platform to use.

I’d recommend an x86 board because as great as the RPI and similar can be, ARM just doesn’t have the same support for a lot of things you might want to self host. I personally like to spring for a used thinclient PC off of eBay, because they have about the same resources as a Raspberry Pi but on an x86 platform. With my thin clients I typically install Alpine but a really light Debian install could work as well, and then from there you can go about installing Docker etc for a little homelab. Even better, if you get lucky and get a couple of them you could mess around with clustering them and some light Kubernetes at home. I’ve got mine running PiHole and Unbound on Alpine to serve my whole house with DNS and it works great. I don’t think I’ve had hardly any downtime issues or anything of that sort.

TL;DR: try a couple cheap thin clients from eBay and you can run some light stuff on them for cheap.

Thanks for the insight, ElPussyKangaroo

As recommended by others, you might want to use a real public domain that you own, and a reverse proxy for split horizon DNS. I personally run Bind9, Unbound, and PiHole as my DNS servers. Bind9 handles split horizon, so if I request my domain internally it gets routed through Bind. Then bind hands it off to PiHole for adblocking, and PiHole makes requests through Unbound set up as a recursive DNS server which doesn’t rely on any external DNS. I also use Traefik as a reverse proxy for all of my services. My set up is more complicated than necessary, and if you want just a few local domains, PiHole + a reverse proxy is plenty good for your needs.

Here is something I wrote previously under a similar post: “Check out the OpenWRT Table of Hardware, it has a list of firmware mod-able off the shelf WiFi routers that work with, you guessed it, OpenWRT. It’s rather versatile as it’s Linux based and can handle VLANs, multiple SSIDs, and of course, you can change the DNS servers.” As I said, OpenWRT is very versatile and runs on many different routers, just find one you like and install it! Many of the supported routers provide Gigabit switching, and some even have multigit for your server connection.