Uploaded image for project: 'Technical Advisory Board'
  1. Technical Advisory Board
  2. TAB-1300

YAML is problematic normative reference

    XMLWordPrintable

    Details

    • Type: Bug
    • Status: New
    • Priority: Critical
    • Resolution: Unresolved
    • Affects Version/s: TOSCA Simple Profile in YAML Version 1.0
    • Fix Version/s: None
    • Component/s: None
    • Labels:
      None
    • Environment:

      Normative

    • Proposal:
      Hide

      Perhaps a better suggestion than I had for the CAMP TC, why not copy both YAML documents into the TC document repository and point to those as the canonical references? That insures they will never change, which addresses my concern about drafts under uncertain management being used for normative references and involves the least amount of effort for the TC.

      Show
      Perhaps a better suggestion than I had for the CAMP TC, why not copy both YAML documents into the TC document repository and point to those as the canonical references? That insures they will never change, which addresses my concern about drafts under uncertain management being used for normative references and involves the least amount of effort for the TC.

      Description

      1.6 Normative References reads in part:

      [YAML-1.2]
      YAML, Version 1.2, 3rd Edition, Patched at 2009-10-01, Oren Ben-Kiki, Clark Evans, Ingy döt Net http://www.yaml.org/spec/1.2/spec.html

      [YAML-TS-1.1]
      Timestamp Language-Independent Type for YAML Version 1.1, Working Draft 2005-01-18, http://yaml.org/type/timestamp.html

      As I remarked for the CAMP TC on YAML-1.1:

      ************

      • From "Status of this Document" I read:

      *****
      This specification is a draft reflecting consensus reached by members
      of the yaml-core mailing list. Any questions regarding this draft
      should be raised on this list. We expect all further changes to be
      strictly limited to wording corrections and fixing production bugs.
      *****

      So the YAML normative reference is to a draft dated 2005-01-18, that
      reflects a "consensus" of members of a mailing list.

      Is that a fair characterization?

      There is no formal organization charged with its maintenance and no
      known process other than email "consensus" as far as any changes?

      **********
      First, draft work should never be used in normative references at all.
      Under any circumstances. As the TC Process notes (Section 1, w:

      *****
      "Normative Reference" means a reference in a Standards Track Work
      Product to an external document or resource with which the implementer
      must comply, in order to comply with a Normative Portion of the Work
      Product.
      *****

      Note the "must comply" language.

      Drafts by their very nature change and reliance on a draft invites a
      lack of interoperability.
      **********

      • From examining the draft it is clear that conformance to YAML, a
        mailing list consensus draft, is critical for use of this standard.

      The only solution that comes to mind, given that YAML states it can be
      copied if not modified, would be to include a copy of YAML 1.1 as an
      appendix and cite that appendix as your normative reference.

      That fixes the text of YAML 1.1 to be what is included in the appendix
      and provides implementers with a stable target for their implementations.
      **********

      The CAMP TC choose to maintain their original reference and supplemented that by pointing to an archive copy of YAML-1.1 at http://xml.coverpages.org/yaml-spec-v1.1-archive-copy.html

      I searched the coverpages but did not find archive copies of either YAML document cited by the TC.

        Attachments

          Activity

            People

            • Assignee:
              Unassigned
              Reporter:
              patrick Patrick Durusau
            • Watchers:
              1 Start watching this issue

              Dates

              • Created:
                Updated: