Re: Approving next version of the HTTPS Specification

[Dave Lemire <dave.lemire@g2-inc.com>]

Small group at today's IC-SC meeting were supportive of the changes in PRs 62 and 63. I'm going to wait one more day, then apply those to the working branch. Final step will be a sanity check review of the working version before submitting it to the TC as the proposed next CSD / PRD. I would appreciate any extra eyes assisting with the final quality check. I'll post another message when the PRs are applied and the working branch is ready for review.

Dave

--

David P. Lemire, CISSP

G2, Inc. (AÂHuntington Ingalls company)
 OpenC2 Technical Committee Secretary
 OpenC2 Implementation Considerations SC Co-chair
 Contractor support to NSA
Email: dave.lemire@g2-inc.com
WorkÂ(301) 575-5190 |ÂMobile (240) 938-9350

On Mon, Feb 18, 2019 at 4:03 PM Dave Lemire <dave.lemire@g2-inc.com> wrote:

I've created pull request #63 (https://github.com/oasis-tcs/openc2-impl-https/pull/63) to implement the adjustments mentioned above, with the exception of softening the MUST / SHALL language (suggestions on that are welcome).

Dave

--

David P. Lemire, CISSP

G2, Inc. (AÂHuntington Ingalls company)
 OpenC2 Technical Committee Secretary
 OpenC2 Implementation Considerations SC Co-chair
 Contractor support to NSA
Email: dave.lemire@g2-inc.com
WorkÂ(301) 575-5190 |ÂMobile (240) 938-9350

On Mon, Feb 18, 2019 at 3:03 PM Dave Lemire <dave.lemire@g2-inc.com> wrote:

IC-SC Members:

I have one remaining PR I need to create to update the alignment of the HTTPS Specification with the Language Specification with regard to the request_id message element. My intent for the PR is:

• Update the description of the request_id to be a string supplied by the Producer, and adjust the examples accordingly (using either UUIDs or simpler strings like id1234, id2345, id3456, ...; I'm leaning toward the simpler strings for readability)
  
  
• Change the header to be used from x-correlation-idÂto x-request-idÂto better align the terminology with the OpenC2 request_id name for the message element
  
  
• Somehow soften or otherwise adjust the MUST / SHALL language about populating x-request-idÂto acknowledge the inability of an OpenC2 Producer or Consumer to guarantee what happens to that field in the underlying transfer implementation.
  
  
• Remove any mention of command_id from the transfer specification because it's optional message content that should be opaque to the transfer spec

I will create that PR before COB Tuesday, hopefully before COB today, so that there's time to examine the details before Wednesday's TC meeting..

Realizing that the devil is in the details, I'm seeking conditional approval from the IC-SC membership to report to the TC at Wednesday's meeting that all PR01 issues have been resolved and the SC recommends approval of the updated document as a CSD04 / PRD02 and that it be released for public review in conjunction with the other two OpenC2 specifications when all three are ready.

If you have objections, now would be the time to raise them.

CRM:Â https://docs.google.com/spreadsheets/d/15RUBgNRoSZzuxhCt5TPX46qU5m9uPYbWkOHUbjNwH54/edit#gid=0

-- cross-posted to Slack #implementation channel and the IC-SC mail list.

Dave

--

David P. Lemire, CISSP

G2, Inc. (AÂHuntington Ingalls company)
 OpenC2 Technical Committee Secretary
 OpenC2 Implementation Considerations SC Co-chair
 Contractor support to NSA
Email: dave.lemire@g2-inc.com
WorkÂ(301) 575-5190 |ÂMobile (240) 938-9350