- From: Martin Franklin <MFranklin@STELLCOM.com>
- Date: Mon, 29 Jan 2001 13:16:21 -0800
- To: "'Joseph Ashwood'" <jashwood@arcot.com>, meadowsj <meadowsj@nobs.ca.boeing.com>, xml-encryption@w3.org
- Message-ID: <6BCDCDFAB19FD4119CE300508B9B06EA188593@izar2.stellcom.com>
Joe Here's ome example I found Asokan, Shoup, and Waidner: Fair Exchange of Digital Signatures At EUROCRYPT '98, Asokan, Shoup, and Waidner extended optimistic protocols to fair exchange [1]. In their protocol, Alice and Bob send encrypted signatures to each other. Then they prove to each other that the encryptions really do contain digital signatures, and that these are signatures on the correct documents. Finally, they send each other decryption keys. If something goes wrong, a third party can step in and use the "proofs of correct encryption" to verify that all has gone well up to a point. Then the third party can take some course of action, such as terminating the protocol or decrypting the signatures for both parties. This is from http://www.acm.org/crossroads/xrds7-1/contract.html > Martin Franklin > Principal Engineer > s t e l l c o m > San Diego, CA > mfranklin@stellcom.com > (858) 947-1572 > www.stellcom.com > -----Original Message----- From: Joseph Ashwood [mailto:jashwood@arcot.com] Sent: Monday, January 29, 2001 12:55 PM To: meadowsj; xml-encryption@w3.org Subject: Re: Signing and Encryption ----- Original Message ----- From: "meadowsj" <meadowsj@nobs.ca.boeing.com> To: <xml-encryption@w3.org>; <jashwood@arcot.com> > If signing a document is akin to making an assertion about a document, > I could perceive some value in keeping certain assertions made about a > document private from third parties. I'm hard pressed to think of an > example where storing those assertions with the document would be an > absolute necessity however, so perhaps it's a non-issue. > > Cheers, > Joe Meadows Actually that's a very good point that hadn't occured to me. I'm still at a loss of any examples where knowledge of the document must be public knowledge but the signer cannot be, except possibly to give anonymous testimony to something, where the testimony could be verified later. Maybe that's enough. I'm not sure it would seem to be a business end of things, and I'm an engineering end. Does anyone have any examples? Or is this a non-issue from the business end also. Or would simply forcing detached signatures with out of band ordering information be enough? Joe
Attachments
- application/octet-stream attachment: Martin_Franklin__E-mail_.vcf
Received on Monday, 29 January 2001 16:19:11 UTC