Re: [urn] [apps-discuss] URNs are not URIs (another look at RFC 3986)

[[e-mail] (Dale R. Worley)]

> From: John C Klensin <[e-mail]>

> Because the query model of 3986 is, at least IMO, pretty closely
> tied to the interpretation of queries in http-style URLs, it is hard
> to make those distinctions except, perhaps, by kludge.

This is one a I don't understand.  As far as I can tell, the
constraint on the semantics of the query-part in RFC 3986 is:

   3.4.  Query

   The query component contains non-hierarchical data that, along with
   data in the path component (Section 3.3), serves to identify a
   resource within the scope of the URI's scheme and naming authority
   (if any).

In regard to the urn-scheme, in which URNs do not have naming
authorities, this seems to establish essentially no limitations.  It
says "identify a resource", but a "resource" could be *anything*.
That is, the meaning of the query-part is entirely up to the NID
registration.  In particular, a particular query-part value can be
defined to mean "fetch the metadata", because the metadata of a
resource can itself be considered a "resource", which is designated by
a URN containing a specific query-part value.  Similar definitions can
be established regarding other "forks" of a resource, and modifiers of
how the query/resolution process is to be executed.

Given that no NID defined the use of query-part, we could write a
single standard for query-parts in the urn-scheme that would apply to
all NIDs.

> It can be accommodated using use "?" syntax of 3986, but only by the
> use of some reserved keywords or other instances of horrible kludges.
> One or two such kludges were proposed to the WG.  I think it is
> accurate to report that they got no traction.

I'm not able to visualize how one could denote the "it" in this
sentence (whose meaning I do not clearly understand) by character
strings in a way in which the syntax of 3986 proves to be excessively
restrictive.  The 3986 syntax is a string from this character set:

   ALPHA / DIGIT
   / "-" / "." / "_" / "~" / "/" / "?" / ":" / "@"
   "!" / "$" / "&" / "'" / "(" / ")" / "*" / "+" / "," / ";" / "="
   and %-escapes

I can't visualize a syntax for identifiers that is significantly more
accommodating -- with the large exception of changing to Unicode as
the underlying character set.  Although Unicode can be represented by
%-escapes, so we can cover even Unicode by having the user interface
translate to/from %-escapes.

Perhaps you know of a proposed modifier syntax (e.g., a way of
specifying a "fork" or specifying modification of the query/resolution
dynamics) that illustrates features that are difficult to implement
within the current syntax of query-part.

Of course, none of this deals with the fact that 2141 does not admit
query-part on urn-scheme URNs.  But solution of any of these problems
will almost certainly require some extension of 2141.

Dale