- From: Brian LaMacchia <bal@exchange.microsoft.com>
- Date: Mon, 6 Jul 2009 17:18:27 +0000
- To: "public-xmlsec@w3.org" <public-xmlsec@w3.org>
- Message-ID: <03703CBF081F9E4CA49CF5FF131362E60D539DF4@DF-M14-05.exchange.corp.microsoft.com>
Folks, I've committed revision 1.73 of xmldsig-core-11\Overview.htm, which includes text to resolve ACTION-142. Specifically, now that FIPS 186-3 is out defining DSAwithSHA224 and DSAwithSHA256, I've made the following updates: In Section 6, define the identifier DSAwithSHA256 (http://www.w3.org/2009/xmldsig11#dsa-sha256) as an OPTIONAL signature algorithm. In Section 6.4.1, added DSAwithSHA256, updated the language in paragraph 1 to describe the four variants of DSA, and updated the Security Considerations section (there was a duplicate paragraph there, among other problems). (Cynthia, I made these changes before seeing your comments - see if you're OK with the new version or if I need to update/revise.) In keeping with the way we did RSA, where we didn't put the key size in the algorithm URI, I chose to do the same thing with DSA. So the intent is that the DSAwithSHA256 AlgID should be used for both 2048-bit DSA and 3072-bit DSA with SHA-256. Similarly, since we don't use SHA-224 anywhere else in the XMLDSIG spec, I did not define a corresponding DSAwithSHA224 (which would be 2048-bit keys & SHA-224). We can add that if people think it's necessary, but I didn't see a compelling reason. --bal
Received on Monday, 6 July 2009 17:19:15 UTC