Quite a few model years of Synology were stricken with a network transceiver issue where they would just stop talking randomly.

I have two WD-EX4’s and the six bay Synology, and none of them have the kind of uptime I get out of my unraid.

If you want something rock solid I would probably scratch build a box with truenas

The opposite of self-hosted would be managed service.

You run it yourself at your own location however you want it

Vs

Someone runs it for you at their location. However the want it

VPS is someone loans you a VM at their location that you run yourself however you want to.

It’s still relevant to self-hosted because you still have to do all the work, you were just using their network, power, air conditioning, hardware and fire suppression. You’re still in the hook for installs and patches, configuration, and software issues.

There are about half a dozen US chains that have battered fries.

On average they don’t fare very well here.

We batter potato wedges here too but they don’t sell all that great either.

By it’s not too difficult, are you actually expecting average users to run certbot cli?

We need to get out of the mindset of jellyfin being self hosted and into the same mindset Plex has of you’re just running it.

Hosting is one of my professional duties so I don’t have problems doing all this. But any idiot can install PMS and have secure shared communication with their friends and family. And we need those idiots.

Jellyfin needs the ability to request certificates and install them without any serious user intervention beyond the initial setup, maybe just an email address. And none of this should require users to touch CLI. This probably needs to be dynamic DNS, maybe we also partner with duck DNS. Right in the GUI make an account, store off the URL in the configs.

I’m presuming this means a le API that will not change from the let’s encrypt side, or advanced clear notice when things are going to change, with opportunities to delay if possible and necessary. That’s where your actual partnership comes in.

We need that thing that Plex has that shows you that your server is remotely accessible from inside the admin. This will help the uninitiated set up a port forwarding and test it.

Once the server is set up and working we don’t need centralized login but we need something. Start with the main settings page, where you drop down in your account on the admin We need an invite users option. It just takes you to users add.

Users add needs to have email or slack or something so that when you add the user it can notify them that they’ve been added and send them a link back to your server. It could be a mailto:// or maybe just a page saying here’s the link to share with your family.

That link would contain the dynamic DNS previously set up and whatever port you’re able to use.

It’s just a handful of creature comforts that plex does particularly well that is barely touched on the jellyfin side. But there’s some of the most important comforts.

Blocking tor at the firewall level isn’t difficult. Anything with packet inspection can do it.

Edit: I’ve looked into this some more and you’re kinda right. It’s SSL and with very little to give it away except talking to exit nodes. Most of the systems I’m looking at are just discerning it from IP, the few exceptions require SSL MITM, which I also wouldn’t put by the government.

Now if they could just tidy up remote access so that everyone is comfortable being able to use it.

They really need to partner with let’s encrypt. If they implemented automated SSL generation and regeneration in the app and a dynamic DNS/Port registry, they would get mountains of new users.

Just tidying up remote access would probably be enough to sync Plex.

When they get serious about encryption they will make tor illegal as well.

Tor will not hide you from the feds once they decide they really want to go after encryption. They can either own enough endpoints to find you directly or simply go and shut down all the endpoints. Or, If they have other IP leaks that are unpublished…

On the upside they are firing most of the competent people in government so there’s a chance the CIA can’t do that anymore

You make laws like the Online Safety Act in the UK. You then attach a multi-million dollar fine to anyone who doesn’t adhere to the bonkers unenforceable stipulations in the text.

All of a sudden, no one but a corporation with a legal department can safely run an instance without putting their money and eventually freedom on the line.

They might not be able to just stop it, but you can force us into a pirate scenario where we have to do it in the dark.

We are likely starting to slowly head into 1984 territory. IF Fascim continues to rise, eventually, non-state-run media will be deemed unlawful and they’ll do what they can to make it go away.

There’s more than one style of tar pit. In this case you obviously wouldn’t want to use an endless maze style.

What you want to do in this case is send them through an HA proxy that would redirect them on user agent, whenever they come in as Claude you send them over to a box running on a Wanem process at modem speeds.

They’ll immediately realize they’ve got a hug of death going on and give up.

Yeah, you can get away with really cheap operations up until you start blowing through your cdn and communication budget

The way lemmy (and federation) works, it needs to do a bunch of operations that can’t happen simultaneously, so there’s a job queue. The queue needs to do some database operations and a bunch of communication operations and each of the jobs needs to reach out to distant servers that may or may not be overwhelmed themselves.

You start with one server it costs almost nothing to host. Sooner or later you want to split out the job servers, then you end up needing to split out the database, when you start getting that many people on your server now you want to consider fault tolerance, Even after tuning you can only fit so many simultaneous users on a web server, you end up needing to do some load balancing. The next step would be trying to split it up geography-wise.

That’s scaling up and it’s what big companies do and it’s very expensive but easy for a small team to manage.

Lemmy on the other hand is designed to be scaled out, running smaller individual user bases on lighter hardware with a bunch of individual administrators instead of a organized team.

If people want to be on a large single cluster application Reddit is still there.

I like what we have a lot better.

They are, probably even too quiet. But the level of griefers, bots and trolls is very very low, and selfishly, I REALLY like that.

which are already doing all the work and should definitely be compensated somehow.

That’s why I donate monthly to my instance :)

A pretty decent sized instance managed will uses a few boxes and some CDN, runs a couple to a few hundred a month, it doesn’t take that many people paying to cover it.

It’s not as bad managing the smaller instances. The app works like it says on the tin until you get really big.

IMO lemmy.world let themselves get WAY bigger than they should have. They had to start doing a hell of a lot more work to keep the thing up.

a while ago where yall saying shit like this became the “normies”

Ohh it’s normal for us. But there are still a lot of sports all-consuming, rabble rousing, linux hating, clue collar people still on Reddit, any I don’t mind them staying over there.

I’m really not happy about bluesky their fragmentation of the fediverse protocols

shrug, I wish they were with us, but they are also a big ole corporate entity, so I’m kind ok with us staying our our side of the fence. As they need to implement payment and corporate protections to their network, we’re free to be free over here.

is only going to harm us in the long run.

We don’t have to play ball. not with them anyway,

I think, If we have any credible threat, it’s going to be from the Governmental gross anti-tampering laws, forced moderation, or backup regulations. They could make it legally difficulty for us to exist.

Do you really think Lemmy could handle the amount of people that Reddit has?

yup. no question. Not one instance mind you, but Reddit is also a giant cluster. (and clusterfuck)

As far as I know the existing instances are usually running on capacity and always in need of donations,

We just need the big bois to stop stuffing themselves. There’s 0 reason to have 2/3 of the totally traffic flooding into world because people are scared of Federation that they never even have to deal with.

Maybe Lemmy would benefit of some way to get people to pay, such as purchasing the ability to give people awards etc.

Maybe we make some premium pay servers with baller architecture, killer response time, user capacity limits and high speed storage?

But the point is that without a business model, the Fediverse will only be able to handle a limited number of enthusiasts before it faces scaling problems.

Eventually, it’s going to be ads, donations or payments. It’s all someone else’s computer, someone has to foot the bill. But at great scale, you should be able to have an ad-free experience for something in the range a dollar or two a month.

How can anyone not love the guy?

YES! #winning

It’s fun for now until the state gets a hold of it. They’ll be over here sooner or later, just let them trickle in and get everyone used to it.

It’s good to have some friends, but don’t let too many normies in…