ietf-smime
[Top] [All Lists]

Re: ESS-05 comments - MLA withdrawal of a signed receipt request.

1998-05-08 08:14:53
All,

a signed receipt from the MLA seemed to be the simpler way to inform the
originator from the withdrawal of his/her request for receipts without
creating a new content type.  I agree that the originator is not expecting
a receipt from the MLA, but it is expecting receipts from other recipients.
 These receipts will never be submitted.  An unauthenticated message from
the MLA without any reference to the message that has been denied receipts
does not help the originator/originator user agent.

That being said.  There appear to be no other expressed requirements for
that feature.  Consequently, it is probably best to stick with Mr.
Pawling's proposal.  

Yves Thériault.  



At 09:12 AM 5/8/98 -0400, John Pawling wrote:
All,

I respectfully disagree with Capt Theriault's proposed additions.  It will
not always be appropriate for the MLA to send a signed receipt to the
originator.  For example, if the originator requested that only certain
recipients return a signed receipt (via the receiptRequest attribute
receiptsFrom receiptList field) and the ML is not listed as one of those
recipients, then the MLA should not send a signed receipt to the originator
because the originator is not expecting a signed receipt from the ML.

If the original signer of the signedData requested a signed receipt from the
ML itself (ex: receiptRequest ReceiptsFrom allOrFirstTier set to
allReceipts), then it would be appropriate for the MLA to send a signed
receipt to the originator.

Please note that the ML's mlReceiptPolicy only applies to the ML members,
not to the ML itself.  For example, if the originator requested a signed
receipt from all recipients and the ML's mlReceiptPolicy is none, then the
MLA would send a signed receipt to the originator, but the ML members would
not send a signed receipt to the originator.

I strongly disagree with the Captain's other proposal:
"It is up to the originator to interpret this receipt as an indication that
no other receipts will be received."

There is nothing special about the signed receipt received from the MLA, so
there will not be any special action taken by the originator's software when
it receives the MLA's signed receipt.

In summary, I believe that my proposal should be accepted without the
proposed modifications.

- John Pawling


At 08:20 AM 5/8/98 -0400, Capt Y. Theriault wrote:
All,  

I would add at the end of John's proposal: "... the MLA MAY inform the
originator of that fact by sending a signed receipt."

I would also add the following:

"It is up to the originator to interpret this receipt as an indication that
no other receipts will be received."

Yves Theriault.

At 11:54 AM 5/6/98 -0400, John Pawling wrote:
All,

I don't object to adding the following text to ESS:  "In the event that a
ML's signed receipt policy supersedes the originator's request for signed
receipts such that the originator will not receive any signed receipts,
then
the MLA MAY inform the originator of that fact."

I would object to this being a SHOULD or MUST requirement.

- John Pawling





Capt Y. Theriault
National Defence Headquarters
PMO Defence Message Handling System Hardware Engineer
MGen R. Pearkes Building, Ottawa K1A 0K2
office:  (613) 995-6476
email:   aa862(_at_)issc(_dot_)debbs(_dot_)ndhq(_dot_)dnd(_dot_)ca

<Prev in Thread] Current Thread [Next in Thread>