Re: LAST CALL:draft-ietf-pkix-time-stamp-05.txt

[Magnus Nystrom <magnus@xxxxxxxxxxxxxxx>]

I could not agree more. IMHO, these context tags adds confusion,
bits-on-the-wire, and makes type-checking harder. Please remove them.

-- Magnus

On Thu, 27 Jan 2000, Peter Gutmann wrote:

> 
> >This document is hereby issued for 14-day WG Last Call.  Please submit any
> >comments to this list.
> 
> I have the usual ASN.1 style complaint, the whole thing contains large amounts
> of unnecessary tagging which obscures the underlying data types (as well as 
> making the encoding more complex than necessary).  For example in TimeStampReq:
> 
> TimeStampReq ::= SEQUENCE  {
>      version                      Integer  { v1(1) },
>      messageImprint               MessageImprint,
>        --a hash algorithm OID and the hash value of the data to be
>        --time stamped
>      reqPolicy                [0] PolicyInformation      OPTIONAL,
>      nonce                    [1] Integer                OPTIONAL,
>      certReq                  [2] BOOLEAN           DEFAULT FALSE,
>      extensions               [3] EXPLICIT Extensions    OPTIONAL
> }
> 
> only the extensions actually need a tag, and that doesn't need to be explicit
> (all the elements except the last are distinct, why are they given context-
> specific tags?).  In addition the primitive types should really be 
> capitalised, ie use INTEGER instead of Integer.  For the rest of the PDU's,
> you can remove almost all the tags without causing any problems, which both
> simplifies the encoding and makes the encoded data easier to work with (for
> example you can actually see an INTEGER rather than just an opaque [0] blob).
> 
> Peter.
> 
> 
> 

-- Magnus
Magnus Nystrom		Email: magnus@rsasecurity.com
RSA Laboratories