Telstra in Australia were at one stage providing routers to homes with a hidden SSID for their premium ‘public’ wifi.

I think it was solidly isolated from the home wifi, and did not eat into the homes speed quota, but still…

It’s no longer accessible from a desktop, only from the Google Maps app.

The router is set as a subnet router, that is how I am able to access other machines on my lan remotely.

I don’t want to, and sometimes can’t, install tailscale on every device I want remote access to.

So I may have duplicate routes- Does that explain the behaviour in my original post? And how would I go about avoiding that?

I could turn off subnet routing, and only turn it on when needed, but I’ll be putting up a bunch of other services that will want to talk to each other- I’m assuming this will break whenever I turn subnet routing on.

I kind of follow what you’re putting down.

I am not using an exit node. How do I go about splitting my routes?

What I want to achieve is ‘normal’ access for within the lan, as well as remote access over tailscale for things I cannot run tailscale on.

I have a commercial VPN, but I am not connected. What tinkering did you have to do?

I set up subnet advertisements by doing tailscale set --advertise-routes=192.168.1.0/24. I did not touch ACL.

The home PC is Windows, the context menu for the tray app give the option to ‘use tailscale subnets’ which is enabled- I assume this is the equivalent of accepting advertised routes.

From the home PC, tailscale ping 192.168.1.2 returns a pong, from the tailscale IP. tracert fails.

That’s very clear, thanks.

I’m guessing you’d have to search the database to make the index, right? To search for ‘gazter’ you’d have had to go over the whole dataset and assigned each entry with a starting letter value, and so on?

Largely ignorant, but data-curious person here.

…what?

I’ve got the fourth Sharma, I used points to get a large Austrian man to walk over the left side of my body.

I’m interested in why you chose the i5 for the automation, rather than the video server?

I’m no expert, but things like transcoding (or even just re-encoding) take a lot of grunt, which it seems the i5 would be good for.

The i3 would be good for more constant, lower power tasks like automation.

At least, that’s my thoughts, happy to be shown your reasoning…

I use an old Stream Deck- not the Steam deck- from Elgato. It’s essentially a small touchscreen with a transparent button pad laid over the top, making for a fully programmable macropad with fully customisable screen-per-key.

Not only can I have esoteric shortcuts, but I can also dynamically label them, depending on layer. I have a ‘home’ layer with icons representing each other layer. So, for example I can load up a video game, and press the corresponding icon on my macropad. It will then change the icons to match whatever command it does- various whistle commands in Ark, for example. I can then change programs into my CAD, and have the icons now be various shortcuts for modelling tools.

Oh jeez, I hadn’t even thought about capitalisation in the file extension. That would be especially confusing if extensions are hidden- the user would be presented with two files that look exactly the same.

But do I type ‘ImportantFile’, or ‘importantfile’?

As I understand it, if I searched for either of these strings in a case sensitive file system, I would not find a file called ‘IMPORTANTFILE’.

At best, a case sensitive file system makes naming conventions more complex. At worst , it obfuscates files. I just can’t imagine a scenario where it would be helpful. Do you really see a need to have a file called ‘aaaAaa’ and a totally separate one called ‘aaAaaa’?

So if someone tells me to look for a file amongst a long list, I need to look in two different areas- the uppercase and lowercase areas.

I get why it’s more technically correct to differentiate, but from the perspective of a human user, it’s a pain in the ass.

Ascending order implies going from low to high

If I have four files, a.txt, A.txt, b.txt, and B.txt, in what order do they appear when I sort alphabetically?

edit: I don’t understand why this was downvoted?

I have one in theirs, they have one in mine.

Wait, ease of installation? As someone who had to walk away from a semi-homebrew, mildly complicated cloud storage setup recently, that’s not the experience I had. Networks within networks, networks next to networks not talking to each other, mapped volumes, even checking logs is made more complicated by containerising. Sure, I’m a noob, but that only reinforces my point.

Jesus fucking Christ… Is this the reality people in the US are living? I can’t comprehend the environment that would lead to someone giving this advice.