Uniform Resource Name (URN) Syntax
draft-ietf-urnbis-rfc2141bis-urn-03

Internet-Draft                 URN Syntax                   October 2012

   Filled in Appendix B contents.

   Numerous editorial and grammar improvements.

D.6.  Changes from WG Draft -01 to WG Draft -02

   Added note at the beginning of Section 1.3 highlighting the purpose
   of this section.  The URNbis charter excludes a revision of RFC 1738,
   and hence the changes suggested on the list to alter and update this
   section have been dismissed.

   Added hint to URN Namespace designers in Section 2 that ":" is
   customarily used in URN Namespaces to provide further level(s) of
   hierarchical subdivision of NSSs.

   Reworked text on fragment identification issues and resulting
   specification, mostly based on Juha Hakala's evaluation of the
   consensus evolving from the list discussion.

   Modified ABNF rule for NIDs to better align it with rules for similar
   identifiers used in IETF protocols.  The new rule now prohibits a
   trailing hyphen, but defers further restricting rules on NID syntax
   (based on the kind of NID) to RFC 3406bis.

   More clearly documented and marked (still open / already closed)
   ISSUES.  The related text will be removed in the next draft version,
   whence it should have been transferred into the IETF issue tracking
   system.

   Text of Section 3 revised, based on Juha's suggestion.

   In Section 5, added removal of <query> part (but not <fragment> part)
   to canonicalization steps for the purpose of determining lexical
   equivalence of URNs (Juha's comment).  Also added examples showing
   this.

   Elaborated a bit more on Encoding Consideration in the URI Scheme
   registration template (Juha's comments).

   Numerous editorial corrections and improvements.

D.7.  Changes from WG Draft -02 to WG Draft -03

   Added text in s1.1 to reflect a comment from SM on other, legacy
   interpretations of "URN".

   Added note in old s1.2 to reflect importance of the name binding
   established by a URN (derived from list discussion on other topic,

Hoenes                   Expires April 19, 2013                [Page 34]
Internet-Draft                 URN Syntax                   October 2012

   Keith Moore et al.).
   However, (despite comments from SM and PSA) preserved excerpts there
   to keep document self-contained and avoid normative down-references
   (as discussed during WG chartering process and pointed out in the
   third para of old s1.3).  Doing so should also help to avoid another
   future recurrence of the discussion on these topics that has consumed
   a lot of resources unnecessarily during the WG formation process.

   Swapped s1.2 and s1.3 (note from SM); however, for logical reasons,
   motivation (part of s1.1) needs to stay in the text before the
   objectives derived thereof (now s1.2).

   Material on query part enhanced (new subsection 2.3); structure of
   query part formally specified with a rather liberal syntax (could be
   more restrictive, if WG prefers); IANA registry of URN query keywords
   established, with two initial entries for the global scope "s" and
   "c" keywords now specified in s2.3.1 and s2.3.2.

   To avoid further confusion (as seen on the list discussion), this I-D
   uses the term "fragment" only for the trailing component in the
   Generic URI Syntax and the semantics associated with it in RFC 3986;
   otherwise this I-D talks about "components" of structured resources.

   Material on fragment part heavily revised and stripped down, put in
   new subsection 2.4.  New text is intended to reflect least common
   denominator of list discussion; i.e., mostly just enable usage by
   specific URN Namespace and otherwise point to RFC 3986 and RFC
   3406bis.
   Namespace designers now have three options to design-in component
   resource designation (if warranted for the namespace), whichever is
   the best fit for their underlying identifier system: (1) media-
   specific designation using fragment part, (2) media-independent,
   abstract designation using query part (to be dealt with by resolution
   system, not resolution client), and (3) media-independent designation
   via assignment of distinct NSSs to component resources.
   (That is being elaborated upon to a greater extent in the -03 version
   of the rfc3406bis I-D.)

   Added text to percent-encoding considerations (Bengt Neiss'
   concerns).

   Amended text on support of existing identifier systems (s3), based on
   various comments received.

   Revised part of text in s5 and s6 on lexical/functional equivalence
   to reflect the new specification for query and fragment (new s2.3,
   s2.4) and to address several comments received; changed s5.1
   accordingly.

Hoenes                   Expires April 19, 2013                [Page 35]
Internet-Draft                 URN Syntax                   October 2012

   In spite of the challenges raised by serious evidence of improper
   management practices for the ISBN system and hence the URN:ISBN
   Namespace (Lars Svensson), the I-D still contains one (hypothetical)
   example based on URN:ISBN; this is being thought acceptable because
   it is in the tradition of earlier documents and we can expect that
   every potential reader of the memo will have an understanding what
   ISBNs are for (or should be).

   Modified title of s7.1 to avoid clash with new s9.1.  Added IANA
   Considerations for "URN Query Parameters" registries (s9.2).

   Acknowledgements expanded.

   Amended Appendix A with text regarding <fragment> usage.

   Filled in details in Appendix D.1; added this Appendix D.7.

   Former Appendix E (guide to IETF document repositories) and pointer
   to it removed (comment from SM).

   Multiple editorial enhancements and fixes.

Author's Address

   Alfred Hoenes (editor)
   TR-Sys
   Gerlinger Str. 12
   Ditzingen  D-71254
   Germany

   EMail: ah@TR-Sys.de

Hoenes                   Expires April 19, 2013                [Page 36]