[Cryptography] Useless side channels

Ryan Carboni ryacko at gmail.com
Mon Feb 12 15:31:30 EST 2018


I'm pretty sure it is illegal for airplane mode not to halt all
transmissions.

Of course this won't matter.

I looked into my BIOS settings and apparently there is spread-spectrum
modulation to reduce EMI? No idea if it reduces or masks side channel
emissions. MAYBE THIS REQUIRES RESEARCH.

There are serious vulnerabilities, and everyone looks the other way. It is
a great mystery.

Somehow there is demand for hardware random number generators when you
could fit an antenna to the microphone jack and pick up low frequency
cosmic background radiation. Or even higher frequencies, wifi requires
error correction because of imperfections within the circuitry and echoes
from walls requiring full knowledge of the device's placement and design.

In the end, the most amazing thing is that every purported benefit is
really a hindrance, and real beneficial things without secret problems are
ignored.

No really, look for every touted benefit that seems to provide marginal
benefit over a simpler solution, and it is really an elaborate backdoor.
x86 is designed for the unlikely scenario that a computer only needs a few
dozen threads.

P.S. Buy iPhones, not Android, because the Apple tax is really the price
for cybersecurity. In some mystery, carrier-specific forks of Android
should really use a subset of apps, unique sprites, and unique bootloader
screens. If Google cared about cybersecurity, they would bar outdated
phones from being sold that can access Google Play. Android 4.4 devices are
STILL BEING SOLD.
Ideally when people buy products, money goes towards a trust for long term
maintenance of them. Like cemeteries. A properly designed product would be
supported for a long time.
Purism's products seem to be from an irrational design basis that the GPL
will cure everything.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.metzdowd.com/pipermail/cryptography/attachments/20180212/7051d576/attachment.html>


More information about the cryptography mailing list