Re: [Trans] Goals and generic mis-issuance fgramework

[Stephen Kent <[e-mail]>]

Russ,

Sorry I didn't get to this sooner, vacation, etc. I see that Rick 
already replied, but for
completeness ...


> ...
>
> Can't anyone build the certificate chain?  Since the logs are posting 
> the supported trust anchors, anyone can build a chain for the end 
> entity certificate.
Anyone can build a cert chain that should be accepted by a log, if the 
chain is "valid" relative to
one of the CAs for which the log says it will accept certs. I think this 
is why Rick suggested that I
make the fist sentence of the second paragraph more general, which I have.
> ...
>
> Alternatively, a log could check to see if the reported certificate is 
> already present, and if so, return the older entry to the party that 
> reports the certificate.  I seem to recall reading this idea at some 
> point, buy I admit I did not look into the current I-D to see if that 
> was where i read it.
I believe you're correct, i.e., a log may, at its discretion, return the 
SCT it generated previouslythen
if a duplicate cert is submitted.
>
>> ...
>
> If the log performs some validation checks, are you suggesting that a 
> relying party can leverage the work already done by the log?  If so, 
> it puts the log checking at a different place in the certificate 
> validation than I was imagining.
That's not what I meant to imply. I was saying that if an entity 
submitting a cert to a log asserted
the type of cert, then  a Monitor and/or a client knows what additional 
checks it ought to perform,
based on the assertion. So, for example, if a client gets a cert that 
the log entry and SCT say is
an S/MIME cert, but the cert has been submitted in a Web PKI context, 
this might raise an alarm. Also if
a log states that it performed syntax checks for the indicated type of 
cert, and a Monitor or client
performs the same checks and gets a different outcome, this should raise 
doubts about the log.

Nonetheless, I guess a client might choose to rely on cert validation by 
a log, although
one should do so only if one has confidence that the log (or a set of 
logs that all agree the
cert was valid) can do a better job than the client. Given experience 
with some browsers, that
might not be a bad idea ;-).

Steve