[642] in WWW Security List Archive
RSA vs PGP trust models
daemon@ATHENA.MIT.EDU (Jesse Mundis)
Wed Apr 26 21:05:38 1995
From: jesse@oes.amdahl.com (Jesse Mundis)
To: hfinney@shell.portal.com (Hal)
Date: Wed, 26 Apr 1995 14:34:16 -0700 (PDT)
Cc: www-security@ns2.rutgers.edu
In-Reply-To: <199504251706.KAA05842@jobe.shell.portal.com> from "Hal" at Apr 25, 95 10:06:38 am
Errors-To: owner-www-security@ns2.rutgers.edu
Hal recently spake thusly:
>
>Re web of trust versus hierarchy models:
>
>The big difference I see is that the web of trust at least as implemented
>in PGP is "non transitive". That means that there is no mechanism to
>follow a chain of trust from one signer to another. If A signs B's key,
>and B signs C's, then just because I trust A as a signer that gives me no
>basis to conclude that C's key is valid, and in fact PGP has no support
>for this kind of reasoning.
>
>With the hierarchical approach, OTOH, there is the assumption that trust
>is transitive in this sense.
[...snip...]
>
>The problem with the web of trust used by PGP is you need to know and
>trust one of the signers of a key you want to use (unless you are going
>to try to validate the key yourself independent of any signatures). This
>works OK within localized groups where in fact most discussion occurs, but
>will not work so well when you are talking to strangers.
>
>Hal Finney
>hfinney@shell.portal.com
Someone else added the statement that with PGP either everyone is a CA, or
nobody is. It has been a while since I've played with PGP, but I seem to
recall that it let you specify two attributes for each key. The first was
how much you trusted a given public key to really belong to whomever it
claimed to represent, and the second was how much you trusted that entity
to act as a CA. With this model, the web of trust allows each individual
to decide who they want to trust as a CA. This buys you the misssing
"transitivity." If I go to my bank, and get their public key, in person,
from their branch office, I can give that key a maximum value for the
"Is it really their key?" attribute. Now, depending on how responsible I think
they are, I decide the "CA-ness" of that key with respect to its certification
of other keys. Likewise, I may trust my friend Bob more or less than my
bank in that role.
Jesse Mundis
jesse@amdahl.com
--
Any opinions expressed above are mine and do not necessarily represent the
opinions policies of Amdahl Corporation.
Jesse Mundis | Amdahl Corporation | Remember:
jesse@oes.amdahl.com | 1250 East Arques Ave M/S 338 | Quality is job 1.1
(408) 746-4796 | Sunnyvale, CA 94088-3470 | -Heard from Maintenance
