A basic problem in using certificates (and attribute certificates) from
multiple issuers (CAs) is that each may issuer has a different set of
extensions (where an extension may be a composite set of attributes). With
X.509, there may be an ASN specification of the extension, but I'm not
aware of a standard way of obtaining it or interpreting it - is there?

I think an important use of XML would be for defining such extensions and
attributes in a well-defined way which will allow interoperability amoung
multiple certificate issuers and applications using the certificates. For
example, our Trust Establishment system provides a pretty easy system for
allowing applications to use (x.509 or other) certificates from diverse
issuers, some of which may not be known in advance, but presently we assume
each certificate has an XML certificate profile associated with it (using a
simple schema/DTD we defined). Clearly, to really allow such
interoperability in practice, it is desirable that such a
certificate/extension/attributes definition would be standardized.

BTW, I'm not too happy with our current profiles and therefore, while I'll
be happy to post them if people are interested (you can also get them as
part of the package if you download), I actually think we need something
different. In particular I believe we could have a spec which can reuse the
ASN definitions, as well as much of the ASN logic. In particular I've been
recently looking into ways to define profile which are compatible with ASN,
a particular one seems to be XER, or XML Encoding Rules (for ASN). I wonder
if others have been looking into XER or have other ideas on what would be
the right way or what are the requirements, to describe such
certificate/extension/attributes format.

Another BTW, I think this discussion should belong on the new XMLCERT list
(archive  at http://jcewww.iaik.at/mailarchive/xmlcert/xmlthreads.html).
I'm copying this initial note to other relevant lists as xmlcert is very
new but I suggest people really interested would follow up there.

Best Regards,
Amir Herzberg

IBM Research Lab in Haifa (Tel Aviv Office)
http://www.hrl.il.ibm.com



Reply via email to