News for hackers. Stuff that matters.
Re:Academics Make Theoretical Breakthrough in Random Number ... by jaunkst on Wednesday May 18, 2016 @01:03AM (#11719877)
I have always wondered why not introduce physical randomness into cryptography. Let's take scalability out of the question and look at the problem at the fundamental level. If we used a box of sand that shifted each time a random number was requested and a camera to scan and produce a number from this source would it not more random than any other method? I'm not a professional in this field I am just truly asking why not..
Re:Academics Make Theoretical Breakthrough in Random Number ... by sobellian on Wednesday May 18, 2016 @01:51AM (#11720036)
New Intel chips use a special bistable circuit to generate random bits in hardware, but many people don't seem to trust it (talk of back doors and such).

Frankly, it seems to me that if you're worrying about the intentions of your CPU manufacturer, you're screwed anyway. I guess that I also haven't put that much thought into it though.

Re:Academics Make Theoretical Breakthrough in Random Number ... by im3w1l on Wednesday May 18, 2016 @02:33AM (#11720173)
Most kinds of CPU misbehavior are detectable: "-Did it add the two numbers correctly? -No it did not" Randomness is different: "-Did it generate a random number? -Well it generated some kind of number...."
Re:Academics Make Theoretical Breakthrough in Random Number ... by topkekz on Wednesday May 18, 2016 @04:58AM (#11720622)
> but many people don't seem to trust it

This paper explains how to weaken the intel RNG by modifying only one logic gate and still make it passes the integrity tests.

http://sharps.org/wp-content/uploads/BECKER-CHES.pdf

Re:Academics Make Theoretical Breakthrough in Random Number ... by Cyph0n on Wednesday May 18, 2016 @02:51AM (#11720234)
Yeah, Intel's new design blows everything else out of the water when it comes to throughput, but it consumes more power.

I think not trusting a hardware RNG is a valid concern. As a consumer, it might not be a big deal, but what if you're buying CPUs for use in top-secret government machines? Given Intel's interests, it may likely comply with a backdoor request depending on the target.

One of the areas I'd like to look into is hardware integrity checking. It's most likely impossible, but it's worth a shot.

Re:Academics Make Theoretical Breakthrough in Random Number ... by flamedoge on Wednesday May 18, 2016 @02:13AM (#11720117)
Only truly truly dependable with 99.9^ % certainty is to build your own CPU and RNG.
Re:Academics Make Theoretical Breakthrough in Random Number ... by stephengillie on Wednesday May 18, 2016 @01:33AM (#11719979)
I've often visualized a Trouble Pop-o-matic[0] hooked to a webcam and a solenoid. The solenoid would pop the die, and simple computer vision could count the pips or number on the die. I'm unsure how many would be needed to make SSL work.

[0] http://i.imgur.com/fL9Bg0M.jpg

Re:Academics Make Theoretical Breakthrough in Random Number ... by VMG on Wednesday May 18, 2016 @02:17AM (#11720129)
Why not just put your cat in front of it and use sha3 on the image?
Re:Academics Make Theoretical Breakthrough in Random Number ... by pbhjpbhj on Wednesday May 18, 2016 @04:09AM (#11720462)
Just capture an image from any webcam. I seem to remember something like SSL cert creation using mouse movement for entropy seeding.
Re:Academics Make Theoretical Breakthrough in Random Number ... by crawlerie on Wednesday May 18, 2016 @05:48AM (#11720770)
That would be PuTTYgen[1] actually.

[1]: http://www.chiark.greenend.org.uk/~sgtatham/putty/

Re:Academics Make Theoretical Breakthrough in Random Number ... by VMG on Wednesday May 18, 2016 @10:40AM (#11722287)
Well if it's pure black or white or sth like that, it's within the reach of brute-forcing.

Ideally it should be private and contain some amount of entropy.

Re:Academics Make Theoretical Breakthrough in Random Number ... by gcp on Wednesday May 18, 2016 @05:51AM (#11720778)
PGP/GPG traditionally did this too.
Re:Academics Make Theoretical Breakthrough in Random Number ... by pbhjpbhj on Wednesday May 18, 2016 @05:18PM (#11726057)
Ah, I think that's where I used it, when trying PGP for mail signing.
Re:Academics Make Theoretical Breakthrough in Random Number ... by bootload on Wednesday May 18, 2016 @02:51AM (#11720241)
"I've often visualized a Trouble Pop-o-matic[0] hooked to a webcam and a solenoid."

I've had the exact same idea, derived from a machine to roll dice.

I was going to use a simple motor rotating with a cam, beneath a die in a small enclosed box. Stops after N cam rolls, snaps a picture of the die face, bit of OCR and continues. You could generate quite a few random numbers using say two 32 side dies. A simpler method with less moving parts would be what you've described. I remember playing this game with a dice enclosed with a spring. Press it down and the spring bent, popped back and the dice rolled.

Re:Academics Make Theoretical Breakthrough in Random Number ... by ricardobeat on Wednesday May 18, 2016 @04:09AM (#11720459)
Check random.org - they go the easy route and use antennas to capture atmospheric noise as input.
Re:Academics Make Theoretical Breakthrough in Random Number ... by eru on Wednesday May 18, 2016 @02:02AM (#11720072)
Cynically: you can't write papers about this.

Practically: lots of cryptographic protocols just ask for a source of randomness, but don't care where it's from.

Re:Academics Make Theoretical Breakthrough in Random Number ... by DigitalJack on Wednesday May 18, 2016 @01:05AM (#11719883)
There was some service (maybe still around) that did this with cameras and lava lamps.

edit: https://en.wikipedia.org/wiki/Lavarand

Re:Academics Make Theoretical Breakthrough in Random Number ... by grahamburger on Wednesday May 18, 2016 @01:12AM (#11719903)
Also random.org. I thought I remembered seeing one that used radioactive decay to generate random files for download but I can't find it now.
Re:Academics Make Theoretical Breakthrough in Random Number ... by Jach on Wednesday May 18, 2016 @01:28AM (#11719969)
Re:Academics Make Theoretical Breakthrough in Random Number ... by pbhjpbhj on Wednesday May 18, 2016 @04:12AM (#11720475)
Well CERN produce terabytes of data from experiments (when running!); particle collision images must be random with much more disorder than mere decay timings.
Re:Academics Make Theoretical Breakthrough in Random Number ... by jaunkst on Wednesday May 18, 2016 @01:18AM (#11719931)
Awesome, at least I am not crazy. A lava lamp was not my first choice but somehow fitting :)
Re:Academics Make Theoretical Breakthrough in Random Number ... by jlgaddis on Wednesday May 18, 2016 @02:21AM (#11720137)
I can't remember their names off the top of my head (not near a PC) but there are packages available for some Linux distributions that use your audio (i.e. microphone) and video (i.e. webcam) input devices for generating entropy.
Re:Academics Make Theoretical Breakthrough in Random Number ... by keypusher on Wednesday May 18, 2016 @02:59AM (#11720265)
This is already the most common source, no? /dev/random uses environmental noise collected from device drivers.