Other Directory Sites: SeekWonder | Directory Owners Forum
The following archive was created by hippie-mail 7.98617-22 on Sat Apr 10 1999 - 01:11:01
David Honig (honig@sprynet.com)
Wed, 26 Aug 1998 09:18:56 -0700
At 05:34 PM 8/25/98 -0500, steven.soroka@mts.mb.ca wrote:
>
>> Certainly, although the original poster did not say how he generated the
>> OTP. If he had some external random source, then he could use that as a
>> basis for OTP use. Certainly he couldn't if he had a deterministic
>> program generating them. Sorry for not clarifying that in my response...
>
>Are there any non-deterministic factors that can be gathered from within
>the
>computer? (internal random source)..? And can one be built and installed
>as
>an extention to the computers abilities?
FM Radio tuned to hiss, high volume ---> Soundcard, 44Khz, 16 bits --> 8:1
parity 'distillation' passes Diehard's statistical battery and is not
deterministic.
Since local RF propogation and the internal instantaneous state of the
amplifiers are not externally known, they are unpredictable.
Running pretty much anything through a good block cipher will whiten
it, but not increase its entropy/symbol ratio. Distilling with
parity increases the entropy of the output and reduces bias.
There are plenty of hazards associated with taking measurements of any
computer activity. RFC 1750 is required reading.
Both the RFC and all the dongle-based <10kbit/sec True RNGs were written
before soundcards became ubiquitous.
There is a secret message embedded in the phosphor of this period.
The following archive was created by hippie-mail 7.98617-22 on Sat Apr 10 1999 - 01:11:01