[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: SDSI and key distribution



At 07:37 PM 5/7/96 -0700, Wei Dai <weidai@eskimo.com> wrote:
>I think you're missing my point somewhat.  The reason to explicitly state
>that the index strings have no meaning other than as indices to Principal: 
>objects is to avoid the need for the administrators of the database to
>verify that meaning.  Because the index allocation is first come first
>serve and there is no meaning attached to indices that need to be
>verified, the administration can be almost completely automated after the
>system is set up. 

It's a nice approach; as long as the servers can safely sign things
without having their keys stolen, and can prevent index reuse,
and as long as you can reliably locate the keys for the servers,
you can verify that keys://keymaster.fnord.com/alice really has your key
and is signed by keymaster.fnord.com; the difficulty is verifying
keys for the keyservers (but you've reduced the scale of the problem by 
a couple orders of magnitude, so it should scale well) and making sure the 
keyserver is secure.  By contrast, the PGP keyservers don't provide
security; you still have to depend on the web of trust, though you could
integrate the two if you wanted to; their benefit is providing convenience.
#					Thanks;  Bill
# Bill Stewart, stewarts@ix.netcom.com, +1-415-442-2215
# goodtimes signature virus innoculation