On Wed, Mar 16, 2005 at 09:07:03AM -0800, Gus Wirth wrote:
>At 11:01 03/16/2005 -0500, George Georgalis wrote:
>>On Tue, Mar 15, 2005 at 08:14:25PM -0800, Stewart Stremler wrote:
>>>begin  quoting George Georgalis as of Tue, Mar 15, 2005 at 10:33:36PM -0500:
>>>[snip]
>>>> the subject "hardware fingerprinting" is no less contiguous then the
>>>> thread. ...if you don't reset your ttl then they will really know it was
>>>> you, the men in the black helicopters don't even have to unerase your
>>>> disks to tell what you've been up to.
>>>
>>>Ah, I see. Since all data comes from the network and none of it of
>>>any worth is created locally, they have an archive of all the bytes
>>>you've downloaded anyway, and so they know what your system may look
>>>like.
>>
>>Yes, now the fire is burning...
>>
>>>With regards to the fingerprinting... it seems like a trivial thing
>>>to fuzz the clock skew randomly, and so render the whole scheme useless.
>>
>>I don't see people skewing their clocks for anonymity any time soon.
>>Besides isn't the test pretty short lived, your clock would have to
>>be changed between test packets, it's not simply a matter of adding
>>a sine wave slew method and changing the clock every hour. Though, I
>>haven't taken the time to really read the article.
>
>You may already have some degree of clock randomness and not even know it.
>One of the methods used to reduce EMI (ElectroMagnetic Interference) is to
>use a variable clock rate to achieve spread-spectrum operation. This
>reduces the single frequency harmonic strength and allows the equipment to
>pass the FCC test for emissions. It doesn't carry over to the peripheral
>clocks yet that I know of but it is on newer motherboards already for RAM
>and CPU clocks.
>
>>>-Stewart "Need a cron syntax for 'random', I see." Stremler
>>
>>probably should do it with mon, so there is no concurrent time reset
>>overlap.
>>
>>(Was it Barry G that came up with the aquarium based random number
>>generator a few years back?)
>
>Sounds interesting. What was the source of the random numbers? An optical
>sensor on the bubbles from the aerator or a microphone? Sonic emmisions
>aren't as random as you might think. Just ask carl.

I think it was temperature based, but we all know true randomness
requires human intervention. ;-)

Hasn't spread-spectrum been a bios option for a while? Not having
applied my flimsy electrical knowledge on its operation and not having any
interference problems, I always accepted the assumption it slows things
down and never enabled it.

// George


-- 
George Georgalis, systems architect, administrator Linux BSD IXOYE
http://galis.org/george/ cell:646-331-2027 mailto:[EMAIL PROTECTED]
-- 
[email protected]
http://www.kernel-panic.org/cgi-bin/mailman/listinfo/kplug-list

Reply via email to