wejn
Joined 5 years ago
Karma
29
Submissions
34
Member since
Feb 17
[ my public key: https://keybase.io/wejn; my proof: https://keybase.io/wejn/sigs/Q5ZKPYzRZhM3jYAgyQuflCM46bgQ05ZifJpxHT1bIRk ]
View on Hacker NewsRecent submissions
Comment · 2 years ago
Svalboard is the next generation Datahand.
In other words, a keyboard with zero hand movement, optical switches, and mostly querty layout (or Dvorak, Colemak, if you want it).
Unlike DataHand, it’s available, hackable, and the finger positioning is more fine grained.
Comment · 3 years ago
Why don’t you get a svalboard.com then? (It’s a commercialized next gen of lalboard)
Comment · 3 years ago
Yup. Been running my own for past two decades, still works.
Comment · 3 years ago
If you destroy the CA, how do you issue new certs via ACME?
Comment · 3 years ago
This isn’t strictly true.
If you want to uphold the name constraints in your CA cert, mark the field as critical. At that point clients that don’t understand them should fail validation of the CA cert.
Comment · 3 years ago
Thanks, I've incorporated the name constraints into the article now. (it is indeed supported by Apple and FF just fine)
Comment · 3 years ago
"only" out of scope.
But based on the comments here, I guess you could use the smallstep CA with Nitrokey HSM if that's your jam...
Comment · 3 years ago
Because:
1. ACME is a dumpster fire prone to mitm attacks.
2. without HSM (an additional investment) it's super bad idea to host your root CA signing key somewhere.
Comment · 3 years ago
That's really neat, thanks for the pointer. ;)
Comment · 3 years ago
And yet, my homegrown root CA cert with 3650 days of validity hums along just fine...
[edited: but since I also want to have host certs that are on various internal servers, the short validity applies to them]
Comment · 3 years ago
A friend of mine runs dns01 thusly: https://ipng.ch/s/articles/2023/03/24/lego-dns01.html
Comment · 5 years ago
Hmm, that's an interesting point.
I would encourage them [= someone who only spent a couple of hours on Unix] to dig further... broader. Mostly by demonstrating that there are real world problems for which Unix is better suited than competition.
IMO in Urbit -- there was not much more to explore after the few hours. Short of digging even further into the guts of it... and that was very unappealing.
Other people here in the comments mentioned that there's no killer app for Urbit. So my intuition was probably correct [to stop digging].