ipv6 isn’t without flaws. but have you SEEN the general software landscape? the things that we put up with on a daily basis?
if we waited with software to be “perfect” as much as we seemingly do with networks, computers would still only be strictly experimental mainframes
@emilia i’ve heard this exact story echoed from others a bunch of times, but. it’s even worse when you’re self-taught.
i’m being told that my style of technical writing is really easy to understand, so I feel like I may be a good pick to work on something better…
@domi It's *checks watch* 2026 and my ISP doesn't support IPv6 to residents, like not even a /64 (a reminder that a /48 is appropriate for a residence, I would get a /32 if I had a big network).
They assigned /127s last time I turned off bridge mode.
@lucasmz @domi https://www.ripe.net/publications/docs/ripe-690/
/56 is the minimum, but let's be honest I'm a bit of an outlier with 12 "top level" internal subnets at home (not counting VPNs) and then even more networks inside my lab. It would be possible to fit in 256 subnets with everything but I'd prefer to just assign new /64 per test deployment without reusing them.
ripe-690: Best Current Operational Practice for Operators: IPv6 prefix assignment for end-users - persistent vs non-persistent, and what size to choose
@domi Big education problem. Especially with infrastructure that isn't the "cool" thing anymore. Everyone wants to do shiny software these days.
Sadly, a lot of teachers didn't touch operational stuff in the last decade and won't in the next and those are the ones that are training the techies that will touch the infrastructure for the next 20 years... Had to go through that
dmi đź’˝ 
emilia
Haelwenn /элвэн/ 
rewby
Trix
Jana
Leszek
lucasmz (en)
altf4
moth
jn (big endian energy)
Eloy.
Ozzelot 