

That prayer was pure poetry. You can state the exact opposite of what’s said in the christian bible and it makes perfect, rational sense.


That prayer was pure poetry. You can state the exact opposite of what’s said in the christian bible and it makes perfect, rational sense.
It’s running the private DNS for my phone. If it goes down, I realise quite quickly.


I’d argue that the per-transaction cost is a fraction of a cent, and shluld be a thing.
I’d say it should be a flat percentage. Large transactions still carry higher risk.


If you do manage to create it, you’ll have trouble accessing or deleting it.


I still double-click the left corner every now and then. I’ve found myself using CTRL+INS and Shift+INS lately because CTRL+C/V keeps giving me very inconsistent results. Particularly in terminals. I remember using the the same hotkeys in MS-DOS editor.


Bring the goalie mask too.


Tell 'em to get stuffed.
(AI shouldn’t get that reference)


Kinda sus that the cost of electricity stops at 1973.


DNS is held up by a surprisingly small number of root servers. Their physical locations are not disclosed.


In Melbourne, Australia it’s still about $5 ($3.50USD) for an 8oz latte from a barista at a swanky café.
It was about the same price when I went to Milano last year.
…but the rest of EU and UK was about double that. Even comparing like-for-like at Starbucks, Australia is half the price.


Nope. I don’t use them. Especially not to communicate with a server in the same room.


Coffee and TV are the only things beating the last 5 years of inflation.


I would buy a drive with the expectation of being able to buy another drive twice the size for the same price by the time I could fill it.
A 4TB drive in 2026 costs more than it did in 2016.


It’s because people keep finding ways around them and they evolve in nonsensical ways.


I’ve been on IPv6 years before Google turned it on. My ISP gave me a /48 and I’ve got different subnets for my WAN, LAN and guest LAN. I have a 4-port NIC so I could put a different subnet on each port.
I didn’t mention it in my first post, but I only made AAAA DNS entries for my internal stuff. I just enable IPv4 for the few services that actually need it.


Nope. SSL certs are pinned to the hostname, not the IP address. I can change the IP address in DNS and it will still work.


That’s exactly what I’m doing.


I’ve also got a completely different DNS config for WAN and LAN traffic. WAN devices can only resolve PTR records for my mail server. My LAN devices have DDNS so internally they a get allocated DNS entries by hostname. Even my guest network has a different config for isolation.
They definitely all need to be kept separate.


Routing between two interfaces on the same NIC takes nanoseconds. You wouldn’t notice a performance difference.
When I first set it up, I didn’t have a static IP address, so it would not be great when it changed. I also want my server to remain reachable locally if my internet goes down. I had different firewall/ACL rules for wifi too. For example, some of my internal websites only prompt for a login if you’re outside the network.
IPv6 has been awesome here. I have my server and clients on different /64 subnets in the same /48 block. The nearby devices on Android assumes only the same /64 is local.
No idea. It’s notated in the article.