I guess it automatically locallizes the price. Shows me 23.99CAD.

Doesn’t show me a captcha either.

I don’t see why not, and for $24 you may as well give it a go.

Oh, I misunderstood; I thought this code/sign-off was a commit to tesseract (I’ve not actually looked at the repo)

What a child. People don’t want to play in his sandbox, so he goes and takes a shit in theirs.

Good riddance.

‘I’m not the problem, everyone else is’. 🤦

I tried to carve out a space where those people and other rougher edges were not drowning out niche communities and hobbies and where people who choose to use the platform to harass others were also out of sight, out of mind.

All he had to do was make his own instance, where he can defederate from/block whoever he likes. Let users decide if they want to be a part of it.

Good intentions don’t justify bad actions, and abusing the trust instilled in you as a developer of a popular tool is rarely taken lightly.

Imposing your own personal opinionated blacklists on other peoples platforms without their knowledge and consent was never going to go well, no matter what your intentions are. ESPECIALLY in the fediverse; a system built to prevent exactly that behaviour.

The DDOS code however…

Frankly that should be considered criminal. This dev should be facing jail time for that alone.

That’s one of the joys of FOSS, when the dev for a popular project can’t continue; there’s often someone willing to pick up the torch and carry it forward.

I switched to this fork when the one you’re thinking of ended development.

https://github.com/nicholas-fedor/watchtower

Last update 3 days ago.

Homepage is my main front end/landing page for reaching/monitoring all my services; but I’ve got both my dockge instances linked together as well, so all my services show up in both.

I run watchtower on a cron schedule. It had already fired for today, so it wouldn’t have run again until tomorrow. I could have waited, but manual is just as easy.

I do love docker + a front end like dockge.

Saw this post, switched to my browser, clicked three buttons; vaultwarden is now updated. Didn’t even move my lazy ass off the couch. (I’m on my phone)

Thanks for the PSA! :)

Fair points.

I’ve been lucky enough to have never been behind cgnat, so I keep forgetting about it.

My bigger concern with tailscale is being required to install software on the client. Not every device I use, I have permission to install a vpn client, nor would I want to.

For example, I have a fileshare using Filebrowser where I store work related files that I don’t want to loose access to or need access to from multiple machines (non proprietary info, stuff IT/MGT wouldnt get mad at me for ofc. I’ve actually cleared it with my managers, so no worries). That’s also a handy way to (temporarily) share large files with people or provide a way for friends to upload large files to me.

I also like to access my emby server (using sufficiently limited accounts), from things like the TV in the work break room, or a friends PC while I’m visiting.

Tailscale is a hurdle that I just don’t need/want.

Glad I could help. I’m not always immediately available, but I don’t mind answering questions if you run into troubles. Just send me a DM and I’ll do what I can. :)

You don’t need a static IP, you just have to keep track of what your current dynamic IP is.

You can do this with either a free or a paid DNS service.

There are a few different ‘free dns’ services that will delegate a subdomain of theirs to you at no cost. Admittedly, I’ve never actually used one of these so their names escape me. Hopefully someone else can point one of those out if that’s what you really want.

I purchased a domain via google domains, when they existed. It’s now transferred to squarespace, because they bought out google domains a few years ago.

It was around $13/year when I first got it a decade ago. It’s now around $28/year.

This allows me full control over the domain: I can use as many subdomains as I want to give each service I use it’s own unique name. (Instead of using their own separate ports that you’ve gotta remember) My domain will also forward all inbound email to my gmail account; this lets me use email addresses like <servicename>@mydomain.example. This way, I don’t share my real email and can immediately tell who sold my info to the highest bidder when I get spam. (I could also host my own email service if I really wanted, but I haven’t bothered)

Add Cloudflare ontop (for free); and it can filter out known attacks, ddos attempts, geofence your services to regions you’ll actually be in, provide/autorenew ssl certs for https, show you usage analytics, cache static data reducing server/network load, etc.

Ultimately, the paid option is well worth it IMO. $2/month (which I typically pay in 3-10 year blocks) is hardly anything.

/edit; vpns are good and all, but they require you to setup software on the remote device to connect to it, and that typically routes most if not all your traffic back to the vpn server then out to the internet. That can create speed/bandwidth issues.

A domain allows you to access your services from any Internet connection with 0 configuration on the client side. Just accessing it like any other website.

I also host a vpn directly from my network, that I access/find via my domain. This means I’m not dependent on a public service like tailscale, but can still add additional security to access private only services (stuff I don’t expose to the open internet)

for when they do get a license

We asked Valve whether there was any way to keep the project alive: properly licensed, with their blessing, on their terms. They said no.

Valve’s refused to sell them a licence.

There is no ‘when they do get a license’.

Beg forgiveness instead of ask for permission is a gamble. dBrand lost the dice roll.

Ouch… Getting injection moulds made is really REALLY pricey. That’s an expensive lesson.

https://en.wikipedia.org/wiki/Network_address_translation#NAT_hairpinning

TL;DR Your router sees you trying to reach your external address and routes the connection back to your LAN without leaving the network.

This does still depend on a functional internet connection however, as your client gets your public IP from a public DNS server over the Internet.

If you were to run a DNS server locally (I use pihole for this), you could have that DNS respond with your local IP, allowing clients within your LAN to resolve the name without needing to reach out to public DNS. This means your local connections will still work when your internet is down; it also provides more privacy by keeping those requests local and can let you make local-only names that aren’t publicly listed.

Of the ~28 FQDNs in my setup, only 4 are public. The rest is local/vpn only and not publicly listed due the above. The reverse proxy then drops all connections that don’t use one of those recognised names, before even completing the TLS handshake. (So direct connections from someone port scanning my IP or using a domain name someone else has pointed at my IP are completely ignored/dropped without response. The server doesn’t even send the TLS cert so as to not expose the names defined in it.)

Hmm…

I don’t like captcha, (or more specifically; I don’t like being farmed for AI training) however I understand the need to filter out bots vs humans.

I really don’t like the idea of being forced to use a google approved operating system and/or provide documents to prove my legal identity. (With a vague promise that my data will deleted, after it’s distributed to a hundred different affiliates that don’t bother with privacy…)

But this is kind of intriguing, at least on the surface. If you can just point a camera at a bland surface like paper or a blank wall, then perform some hand gestures, to satisfy the ‘I’m human’ test. Without being forced to use anything more. I’d be far happier with that, than I am with the current proposals of ‘upload your govt ID to everyone that asks, to prove you exist. Trust us, we’ll all delete it immediately after, perhaps throw in a facial scan while we’re at it, wouldn’t want to be uncertain…’

So, what’s the catch? I’m sure the footage can and will be used to improve AI video rendering, particularly hands obviously. At least you can choose a background that’s not identifiable, and keep the rest of the scene empty (hopefully?)

Is it going to function on linix PCs with whatever flavor of FOSS or highly customised browser people want to run? Or are people still going to be told, ‘pick a corporate overlord, buy one of their devices and don’t you dare modify it, or good luck accessing modern society.’?

I’m not hopeful of the answers to those questions.

I try to be in the habit of making a full image onto a demonstrably working spare card every couple weeks.

That’s a whole lot of writing to an sd card, wearing it out. It may fail by the time you want to read it. You also destroy each previous backup by creating a new one.

Each of my rpis backup to my main server nightly using dd via ssh. The server then keeps historical backups of those .img files via Borg so I can pull any version from any day of the last year or so.

I use Emby instead of Plex or Jellyfin; mostly because it has an Xbox client, and I’ve already got a lifetime licence. One of my most active users only watches via Xbox.

Really don’t like Plexs centralised user system or the overall direction they’ve been headed for years, so I moved away from that long ago (8+ years ago at least). Jellyfin wasn’t up to par at the time (though they’ve made leaps and bounds of progress in that time), and Emby has always supported more types of devices\clients. Their device limit (the client count limit with premeir) has never come into play for me, but I know there are larger user bases out there where that is a problem.

Embys development is extremely slow though, taking YEARS to implement simple features or even address major concerns. Plus their support sucks without the community stepping in and providing it on behalf of the staff. Luke (the main dev) is better at copy+pasting candid responses than he is at actually interacting with human beings.

For anyone else not sure what they’re looking at…

Tree Spiking

Tree spiking can be an extremely effective method of deterring timber sales, and seems to be growing more and more popular. Mill operators are quite wary of accepting timber that may be contaminated with hidden metal objects, — saws are expensive, and a “spiked” log can literally bring operations to a screeching halt, at least until a new blade can be put into service. The Forest Service and timber industry are very nervous about spiking — when they or the media raise the subject of monkeywrenching, this is the form most commonly discussed. Agency and industry officials are loath, however, to raise the subject. Indeed, the Forest Service (FS) often fails to publicize incidents of spiking, on the theory that the less the practice is publicized, the less likely it is to spread. When the Freddies (FS officials) do publicly acknowledge that a spiking has occurred, they often make a considerable effort to find the perpetrators, even to the point of offering substantial cash rewards. (No modern-day tree spiker has been caught, however.)

To avoid this, you will need an IPv4 address on your client, or an IPv6 address on your server.

This confuses me because I have an IPv4 address on the client, and that IPv4 is what the server is seeing make the connection…

/edit

I think I get it.

The client actually only has IPv6. The IPv4 address I’m seeing in the log and whatismyipaddress.com is the address of my mobile providers NAT.

Thanks. I still haven’t totally wrapped my head around IPv6. Stubbornly happy with IPv4 tbh, but it seems the rest of the world is moving on, understandably.

Or similar.

Sold at Homedepot, tool rental usually has a fogger machine to coat the room, but the spray bottles work fine too.

That insulation needs to go too.