Hasn’t changed much. I have accepted the risk and, for now, the desire to have publicly accessible stuff outweights my lack of desire to deal with the realities of a VPN. I mean, I want to share stuff with others, and I want to be hosting it myself because I have grown to distrust someone else’s service… so what else can I do but risk it?

Besides: I expect tailscale or the underlying VPN services to already be a target. I would be surprised if they don’t find anything. What then?

Sunsetting to me implies that existing users may continue using it for an indeterminate amount of time, with or without ancillary services, but that the team/business has moved on and no one should pick it up. There may, or may not be, accommodations for existing users to either continue using, or migrating to another service.

To continue in analogy land, the sun has set, the store is technically still open because there are people inside, and they are not being kicked out, but you shouldn’t, or won’t be able to, come in as a new customer. It also implies that the people in should know it is getting dark outside and should maybe find their way home (or to the bar).

A shutdown instead implies a hard deadline after which the service will no longer work.

To give examples:

• an email service that is sunset may still allow existing users to receive or send emails, but will not allow new sign up. It may, for example, retain the email services to facilitate migrating to another provider, and otherwise allow setting up address forwarding. If it was a paid email account, then I would expect to continue paying, or at least for the service to continue up until renewal time, with renewal impossible.
• an email service that is shut down will stop allowing users to receive and send emails by the shutdown date. I would not expect email forwarding to another provider. Archive access may or may not be provided after the shutdown date. Service will terminate. Any money exchange for this service would end at the specified dates, and refunds, if necessary, should be handled for anything after the shutdown day.

My current setup is Uptime Kuma sending notifications via Ntfy.

I need to resolve the issue of what to do when the internet is down (and thus Ntfy cannot be reached), but it has so far done everything I want it to do. It also means I get push notifications on all the devices I care to receive them on.

Do they actually solve the issue of trust? I don’t dispute that they make it a simpler problem, but I don’t think you can claim that you don’t have to trust anyone.

My understanding is that if the blockchain decides (i.e.: people in general) that you do not own the NFT, they could create a fork in the database. You may still have your version, but if no one uses it, aren’t you left in the same place?

To actually answer the question: because bad things happened.

First of all, this is a public road used by a lot of people. This is a shared space/resource that we all want to benefit from, and that means that it needs to be safe for everyone to use it (otherwise, people are not going to use it and thus no benefit from it).

By speeding (or letting a Tesla do the self-driving without paying attention), they broke important rules. The other point is that had there been an accident, then we would have punished the person for the accident too.

We can argue about whether a given speed limit is adequate for a given stretch of road, but the core of it is that someone (usually experts) settled on a speed under which it is safe to travel. Going above that speed means that you are now taking on more risks than accounted for when designed.

Why is breaking these rules bad? Because, for example, normalization of deviance. The long-story short is that safety rules are an important part of keeping a system safe, and ignoring them removes the safety margins built into the system and either makes accidents more likely, or makes any accident that does happen more disastrous.

To stay around the speeding example, consider the following: people take a set amount of time to react. The faster you go, the more distance you need to react. If you are traveling down a road, and an animal crosses 100 meters ahead, the faster you go the less time you have to react.

Also consider the following: roads are not perfect. If you go over a patch with sand, or a recently-rained-on surface, you may end up with much less grip than you thought you did, and you end up spinning on the turn, or not breaking enough. And you won’t know you need it until you do, and now you’re too late to not speed.

If you were on a private road? Good luck. Its on you. But you are in a public space, and it’s likely you’re putting others at risk, either directly, or by (as linked above) normalizing deviance and making it more likely for others to not follow the rules when it matters.

Sure, and using the AUR is something to pay attention to when you do. You are right on that¹.

But, here is the thing: I just don’t have to update anything if I don’t want to. There is no Windows Update saying “next reboot we’re going to update and you don’t get a choice”. That is what I meant when I said that I just don’t worry about it.

Though I guess that says more about what one gets used to when dealing with Windows, than necessarily being high praise for Linux. My work machine has been having issues with a bt headset for reasons I cannot phantom despite working for years, so forgive me if the joy of being able to leave well enough alone has got me excited.

¹ I would argue that updating in general has become more fraught with danger, so I don’t really see the AUR as particularly dangerous in comparison, but my tolerance for danger may just be out of whack. See Notepad++ vulnerable versions, or just npm in general, for non-Linux cases.

I got tired of Microsoft. I had seen that Linux was now good enough for what I wanted to so, so I decided to jump ship.

It wasn’t a quick decision. I only transitioned from Windows because I finally got around upgrading the PC.

So far? I mean, there was the kerfuffle with the AUR recently. And I still haven’t figured out a few issues here and there. But damn it just works and there is no worrying about updating or not updating or anything like it.

I’ve been slowly setting up my own services locally, and one of the issues I quickly ran into was simple: what’s the point of running the uptime monitor in the same machine? Granted, it still serves is purposes, but…

So, if you do find a way to get something going, I think it’d be really cool. My biggest concern would be, though, on how to ensure you don’t accidentally build a DDoS network instead. A though issue I have no idea how to solve, but which should be solvable!

I think that with enough community effort, it could.

Some of what I think is missing is just community documentation (manuals, tutorials, troubleshooting pages) that are easy to find and recent. While yes, solutions from 5 to 10 years ago still work, they often don’t reflect the full recent reality, never mind the tendency for CLI solutions (which are great, but plenty of people are intimidated by the CLI).

The other thing that I think is missing is polish around things that are just off the beaten path… the kinds of things that not everyone will do, but that most will do at least one of.

I missed not having to think about Samba to share a drive with a Windows machine.

On the other hand, I was rightly pointed to better alternatives for my use-case, so perhaps the learning will make the pain worth it?

Standardized parental controls would be great, actually. But it should be proper parental controls, not whatever this is. Because at the end of the day, the parent* should be involved in what their child is up to, and allow (or not) based on what the child needs and/or wants, instead of whatever we are doing now.

Or, to put it another way, if your teen has read Games of Thrones (the physical books), I don’t see much of a point in forbidding them from going to the wiki of it, and I’d be hard pressed to justify stopping them from talking about it online with other people who have read the books. The tools should allow for this kind of nuance, because actual people are going to use it and these kind of situations happen all the time.

* some parents are awful and would abuse this, see LGBT+ related things, but that’s a social issue, not a technological issue.