I did receive some comments on this draft at the last second. Two were
on the key size text:
- SHOULD- on 1024-bit DSA key ought to be SHOULD
- In 4.3: 1023 <= key size <= 2048 : MUST needs to be
1024 <= key size <= 2048 : MUST
(this is the same rationale as for 3850bis) The 1st one I think is a
good suggestion because SHOULD- on a key size doesn't seem to make much
sense to me especially in the light of DSA with SHA-256 being a SHOULD+.
The 2nd one is editorial because we can't have two different
requirements for the same key size.
I also received some other non-key size related comments:
- RSA-PSS should be replaced with RSASSA-PSS
- RSA-OAEP should be replaced with RSAES-OESP
- RSASA-PSS in references should be RSASSA-PSS
- Change noted to 3.2.2 not implemented and should also
be applied to 3.6 (replace encrypted with enveloped)
- Rephrase the last sentence in the 2nd para of 2.3:
OLD:
As AES 128 CBC is the mandatory to implement content encryption
algorithm thus, when DH ephemeral-static is supported, AES-128 key wrap
algorithm MUST also be supported.
NEW:
As AES-128 CBC is the mandatory to implement content encryption
algorithm, the AES-128 key wrap algorithm MUST also be supported when DH
ephemeral-static is used
- Rephrase 1st sentence in 4th of Security Considerations
OLD:
The choice of 2048 bits as the RSA asymmetric key size in this
specification is based on the desire to provide 100 bits of security.
NEW:
The choice of 2048 bits as the RSA asymmetric key size in this
specification is based on the desire to provide 112 bits of security.
All but the last I consider editorial and will incorporate. For the
last one, instead of replacing 100 with 112 I'd like to make it say "at
least 100 bits of security". I think this is what we intended though
technically a 2048-bit key size does offer 112 bits of security.
Unless there's strong objections to these resolutions I'll post a new
version of this ID Friday afternoon for Tim to continue processing.
spt
Sean Turner wrote:
To date, I have received no comments on this version of the ID.
spt
Sean Turner wrote:
This ID incorporates comments necessary to resolve IESG DISCUSSES, one
IETF LC comment I missed, and incorporates key size text from our Area
Directory. The changes were in some cases significant and as a result
I'm asking for a one week comment period that will end 22 April 2009 @
8am EST.
spt