RE: proposed addition to application/pkcs7-mime smime parameter

["Jim Schaad" <[e-mail]>]

Blake,

I have both a very basic and a very complicated answer to this
questions.

Is the message document correctly titled "How to do secure MIME with
CMS" or "How to do secure messaging with MIME and CMS"?

If the answer is the first, then this should be done.  If the answer is
the latter (and this is the position that most people think from) then
this should not be done and a separate draft should be written on how to
do the additional CMS security types.

I don't really want to bifercate the current Message and Certificate
drafts to have different documents for both the first and the second
(although the latter documents would be a "simple" profile of the former
documents). But I think we need as a group to make a decision on what
document we are writing.

jim

> -----Original Message-----
> From: [e-mail] 
> [mailto:[e-mail]] On Behalf Of Blake Ramsdell
> Sent: Thursday, June 19, 2003 3:51 PM
> To: 'Rohan Mahy'
> Cc: [e-mail]
> Subject: RE: proposed addition to application/pkcs7-mime 
> smime parameter
> 
> 
> 
> > -----Original Message-----
> > From: [e-mail]
> > [mailto:[e-mail]] On Behalf Of Rohan Mahy
> > Sent: Friday, June 06, 2003 7:59 PM
> > To: Blake Ramsdell
> > Cc: [e-mail]; [e-mail]
> > Subject: proposed addition to application/pkcs7-mime smime parameter
> > 
> > I have included some proposed text to add the other CMS types to the
> > smime-type mime parameter.  Alternatively a new cms-type mime 
> > parameter 
> > could be defined, but this seems a but pedantic to me.
> 
> We are in a strange situation here, and I'd like to get 
> feedback on this.  One side of me says that the 
> "application/pkcs7-mime" means "MIME packaged in PKCS #7 
> (which then became CMS) for the purpose of moving around 
> secured MIME entities".  I don't know if it's a better idea 
> to a) overload the application/pkcs7-mime type to mean "CMS, 
> possibly not wrapped in MIME", or b) introduce 
> application/cms in a separate draft, along with a cms-type 
> parameter that explains the inner type.
> 
> I know that there was much discussion about application/xml 
> in a similar context, and I don't know if there's anything we 
> can learn from that in order to resolve this.  It seems that 
> the application/xml semantic would be very similar to the 
> application/cms semantic, but I may not understand it correctly.
> 
> I'm going to release 2633bis-05 shortly, and if there's no 
> discussion on this topic I'm not going to include anything in 
> that draft.  If it's important, we should work through it.
> 
> Blake
>