So, if you would, help me out with the ‘why’ part

It eliminates a single point of failure, can be used to bypass censorship, and allow for community support/engagement in a way that is harder to track and suppress (in that there’s no ‘central’ hub and you have to go after nodes individually. From an opsec point of view, you’re still broadcasting a signal that someone in range can pick up). Obviously it requires many devices to make a good mesh work, but short of DOSing every channel or just blowing out the signal space, it’s gonna be hard to take that down.

I see it as something like tor or i2p, not something for general use at the moment, but definitely has good uses.

There’s not really too much of a debate, just a lack of deep understanding of how the infrastructure works under the hood.

The other person (rightly) doesn’t want to share their local network (what’s behind your wifi router) with their neighbors. My only point was that, much like current ISPs, you don’t share any networking with your neighbors. The only thing remotely close to ‘shared’ would be the individual uplinks (your ISP connection) from each residence to the (shared) networking gear of the ISP.

A local ISP and a Telco aren’t (shouldn’t) going to be handling the base networking layer any differently. They’ll all have individual connections between them and subscribers, and the only way that I could get into your network is to setup services and configure either side to talk to the service on the other.

To actually ELI5 (which I am exceptionally bad at with actual 5yos), Alice and Bob both get their toys from Charles (Telco ISP) who charges a lot of money, and doesn’t treat them well when they try to use the toys they got. Dan comes a long and works with Ed and Fred to set up a local toy store and try to treat customers better. Bob (irmadlad) is concerned that the new local toy store means he’ll have to share the toys he bought with Alice, not realizing neither store makes you share your toys.

why would I want to hook my uplink to someone else’s network

Well, the biggest reason I could think of is that you want to access the Internet.

Your local network is only as good as the services you run, and most people don’t self host. If you choose not to hook your uplink to your ISPs network, you’re not gonna be able to do all that much.

Oh I get how it would all work, I’m not into sharing my network.

See, I’m struggling to think that you do. You’re not sharing your network with anyone. You’re just hooking your uplink into someone else’s network, who will take as much (or more, given how fucky current ISPs are) care to keep you and your neighbors from talking to each other without your own config letting it happen.

There is a user here that mentioned he is in funding talks for a local, independent ISP. I’m not really sure I’m ready to be connected to my neighbors intimately. Good fences make good neighbors.

Why do you think an independent ISP would operate any differently at the networking level on a per-customer basis? This is basic network segmentation, and my home gear can do that pretty easily. Throw each customer on their own vlan that’s a /30 and they can’t do anything more than talk from their node to the central router.

Good firewalls make good digital neighbors, and an independent ISP isn’t going to survive long if Alice can access Bob’s home network over the ISP without having something specifically configured in Bob’s network to allow that.

Top 10 reddit moment: worthless agreeing meme response to an unhelpful comment that doesn’t even answer the question asked.

Some hobbies have minimal levels of skill/knowledge/equipment to properly do them, and I’d argue that self hosting is one of them. You can say people are hostile to beginners, but I might say people are trying to save them from themselves by not just telling them how to slap shit together so they can put it on the Internet and get owned by Internet Background Radiation in a short period of time.

My personal opinion is that beginners are too over confident in their skills or expect setting things up is like setting up an online account, and expect everything to be ready for them to install in their preferred method, and get upset when people tell them they need to upskill to be able to accomplish their goal.

An example of this is a conversation I had with someone online about some docker distributed app, and people were trying to get the person to use docker like the install doc says instead of trying to figure out how to just install it directly into the OS, because that’s the way they’re used to doing stuff and they were determined they weren’t going to change now despite the software author’s supported path not including direct install. If the person was willing to learn docker (which is not very difficult if you can follow a tutorial and use compose files), they’d be able to quickly accomplish what they want while also opening more doors for them in the future.

My hero, thank you!

Can I ask how you do that? I have some debian and fedora boxes I should configure for that

had to re-test people and strip their badge/gun if they failed

And I wonder how many of those stripped of their badge and gun went a county over and got hired?

Problematic officers are usually just moved around when they get in trouble, so I have no faith that the officers with mental health issues actually got help or actually removed from the force entirely. And that’s assuming those who didn’t fail the second time legitimately got that score and didn’t have a thumb in the scale…

Good luck arguing that a missed config counts as an ‘unforeseen issue’. If they go that route, people will be all over them for not being SOC compliant wrt change control.

99% uptime in a year gives you 3.65 days of downtime, which I think would still be within SLA (assuming nothing else happened this year). Though, once you get to 1 9 reliability (99.9%), you’ve got a shift and change you can be down before you breach SLA.

If their reliability metrics are monthly, 99% gets you less than a shift of down time, so they’d be out of SLA and could probably yell to get money back.

Interesting, I mainly use audiobookshelf for podcasts along with my audiobooks, I’ll have to look at what the ebook integration is like.

But the ai means it has to call a server with the information, so I can’t get past that.

That just means they need to ship the model and a way to run it locally. I’d love that, and wouldn’t give a shit if it took a long time to run on my hardware for something like that

I usually just do

Docker compose down
Docker compose up -d

As I would with any service restart. The up -d command is supposed to reload it as well, but I prefer knowing for certain that the service restarted.

Out of curiosity, what did you update and what broke? I had that happen a lot when I was first getting started with docker, and is part of how I learned. Once you have a basic template (or have dec supplies example files), it makes spinning up new services less of a hassle.

Though I still get yelled at about the version entry in my fines because I haven’t touched mine in forever

Brb, setting up tons of instances for my area so it looks popular

Docker compose pull; docker compose down;docker compose up -d

Pulls an update for the container, stops the container and then restarts it in the background. I’ve been told that you don’t need to bring it down, but I do it so that even if there isn’t an update, it still restarts the container.

You need to do it in each container’s folder, but it’s pretty easy to set an alias and just walk your running containers, or just script that process for each directory. If you’re smarter than I am, you could get the list from running containers (docker ps), but I didn’t name my service folders the same as the service name.

All your docker data can be saved to a mapped local disk, then backup is the same as it ever is. Throw borg or something on it and you’re gold.

Look into docker compose and volumes to get an idea of where to start.

Really wish we would have gone hyper v instead of Stack HCI

Congrats on doing it the way the website owner wants! You’re now into the content, and you had to waste seconds of processing power to do so (effectively being throttled by the owner), so everyone is happy. You can’t overload the site, but you can still get there after a short wait.