Profile: db2f6a66...
The Use of Knowledge in Society [F.A. Hayek]
https://fountain.fm/episode/12643340106
This is an essential piece diving into the complexities and profound depth in the market establishment of a price. And why the very notion of the price being "wrong" or fixing it, is self-evidently impossible, as the only information we have to make a comparison, is given to us by emerging prices in the market.
#[1]
#[1]
Qubes is a cool idea, but it isn't enough for me. If it works for you, it's good. But I never want my bare metal touching anything directly. I would suggested a small CLI Tor gateway (like Whonix), and then a few full-blown VMs connecting through it (or NAT, depending).
#[5]
Data can be saved in isolated, encrypted machines. Qubues is a resource hog, and there's no reason to run 20 VMs, really. Two or three dedicated, fully-encrypted VMs can isolate just fine, with more security.
#[6]
The point of isolating VMs is to isolate compromise. It makes no sense to have VMs that are easily identifiable and unencrypted. If a running machine is compromised FDE doesn't help. But isolated encrypted machines with correctly configured permissions do.
#[5]
Too many people think full disk encryption solves everything. It doesn't.
Even possessing unencrypted VMs is a huge attack vector. Plausible deniability should be the first line of defense.
If a VM can't be run in an isolated, encrypted container, it loses 99% of its security.
Two problems with Qubes:
1. Can't run it in an isolated VM. It has to be bare metal. This is a huge problem for plausible deniability.
2. It's own VMs can't be encrypted. If an adversary gains access to the machine, all VMs are compormised.
It's way too vulnerable.
You can already do that https://brb.io/n/note1as2vhpy0p0z6z60qu6wqnuxuan9al0nxrtnc8zlyhcw4exq7wy6qpzht2u
#[5]

