Re: Electronic envelopes

M****** R****** (m***swirl.com)
Mon, 27 Sep 1999 21:16:21 -0700

• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
• Next message: mgraffamidsi.net: "Re: Electronic envelopes"
• Previous message: Criptyk Hayz: "Re: DH:secure Prime & RndNum Sizes for key generation?"
• In reply to: Anonymous: "Re: DH:secure Prime & RndNum Sizes for key generation?"
• Next in thread: jeromepsti.com: "Re: Electronic envelopes"
• Next in thread: jeromepsti.com: "Re: Electronic envelopes"

>Does anyone have any *practical* ideas?

ntp - http://www.eecis.udel.edu/~ntp/ - uses a convergence method to
synchronize clocks between multiple machines fairly well. My understanding
is that you are computing the time at one node based on propagation times
and times from multiple other nodes (which are all time servers). It is
not secure.

Might it be possible to work up something along the order of secure time,
and encrypt messages flowing in the graph between the shared time entities
in order to give it some properties of resiliance ?? Obviously there will
be limits to what kinds of accuracy can be achieved dependent on the
topology of the graph and how many "real" time clocks are tied into it (ntp
uses many gps receivers in addition to atomic clocks and other things).

One secure time is achieved, many things become possible ..

(Caveat: I have not followed this thread well, neither do consider myself a
cryptographer, which is why I am generally in watch and learn mode.)

--m

• Next message: mgraffamidsi.net: "Re: Electronic envelopes"
• Previous message: Criptyk Hayz: "Re: DH:secure Prime & RndNum Sizes for key generation?"
• In reply to: Anonymous: "Re: DH:secure Prime & RndNum Sizes for key generation?"
• Next in thread: jeromepsti.com: "Re: Electronic envelopes"
• Next in thread: jeromepsti.com: "Re: Electronic envelopes"

This archive was generated by hypermail 2.0b3 on Tue Sep 28 1999 - 01:29:35 CDT