3. Either has "unique" advantages
see above. If they had, why not listing them and join all advantages
into one standard?
Perhaps I should have said "unique and mutually exclusive" to drive the
home, though I thought I had done that with my examples. For instance it is
mutually exclusive to allow anyone to be a signer and to allow only those
meeting certain standards to be signers.
With respect David this is simply not true. An S/MIME client can be
used in precisely the PGP web of trust mode. I will take Outlook
Express as an example but the same holds for other S/MIME
clients such as Navigator only the process is slightly different.
If someone sends me a certificate that is either self signed
or signed by a root not trusted by the browser I simply click
a few buttons to add the sender to my address book then go
to 'properties->digital ids' I select the untrusted cert and
select 'explicitly trust this certificate'. If I want to trust the
key as a signing key I click 'inherit trust from this cert.'.
The only sense in which the built in client certs are in any
way special is that the browser company (Microsoft/Netscape)
has reviewed their Certification Practices Statement and
their operations and determined that their certificates
may generally be considered trustworthy. If you disagree
with this assement then uncheck the box next to VeriSign,
MCI, Thawte etc and you are entirely OK.
This is precisely what many people using VeriSign
private label (aka OnSite) certificates do. There is no
reason to force them to recognize the public certs
on an Intranet.
There is a rather nice poster put out by PGP which shows
all the various options for arranging certificate issue. As
the poster shows PGP can be used in hierarchical fashion
if desired. Unfortunately it overlooks the fact that X.509v3
certificates are equally flexible.
Please don't start the tiresome flames about personal interest.
Its not as if PGP Inc.had no vested interest. I happen to believe
that the more ways certificates can be used the larger the
market for them will become. It is certainly not in our interest
to be dogmatic.
Let us all stop being dogmatic about all this then. If there was
indeed a problem with the certificate structure of S/MIME that
would not make its transport formats invalid. There would be
no need to obsolete 25 million clients simply to introduce
a different trust model. Evolve the certificate spec to a new
level, either in ISO or if need be in the IETF.
This is starting to look like one of those programming language/
editor/ Operating System wars*. Ho hum...
* BTW the answers are occam, LSE and VMS or Java, does not