Dear all,

I'd like to get some progress on the node requirements draft. I have some outstanding 
issues to address, which I will in the next few days.  However, the high-order bit to 
settle is about the requirements language.


The current draft can be found here:

        http://www.ietf.org/internet-drafts/draft-ietf-ipv6-node-requirements-01.txt


The design team decided to use the following terms, in concert with the keywords from
2026.  

   Unconditionally Mandatory

      If a node claims compliance to this document, then it must support
      the behavior specified within each conformance group listed of
      type unconditionally mandatory.

   Conditionally Mandatory

      Conditionally mandatory groups include those which are mandatory
      only if a particular condition is true, such as whether a specific
      type of hardware is present, or whether another given group is
      implemented.  When a conditionally mandatory specification or
      group is described, the condition will also be described.  A given
      RFC or portion thereof can sometimes appear in multiple
      conformance groups, with different conditions.

   Unconditionally Optional

      Behavior that is neither unconditionally mandatory nor
      conditionally mandatory is unconditionally optional for compliance
      to this document.

The reason we felt this type of language was used becase:

        1) It was felt that IPv6 nodes may be have very different
        applicability, therefore some functionality are mandatory in 
        some circumstances, but not in all.

        2) 2026 is not a mandatory reference, and in many cases, some
        of the normative references in this draft do not use the 2026
        language.

        3) MUST, SHOULD, MAY, etc. do not cover cases where there is 
        dependency on other functions.  For example, support for IPv6
        over Ethernet is not a MAY, SHOULD or MUST.  One can imagine
        that in the case of a wired server running IPv6, IPv6 over
        Ethernet would be a should, but in the case of a purely 
        wireless device IPv6 over Ethernet is useless.

If you have an opinion on this, please speak up.

thanks,
John

--------------------------------------------------------------------
IETF IPng Working Group Mailing List
IPng Home Page:                      http://playground.sun.com/ipng
FTP archive:                      ftp://playground.sun.com/pub/ipng
Direct all administrative requests to [EMAIL PROTECTED]
--------------------------------------------------------------------

Reply via email to