Hello Ray,

URN in its basic form is not HTTP URI, but it can easily be expressed as HTTP URI as long as that is the only way of making the URN actionable. All persistent identifiers are used in this manner for the time being.

For instance, the URN of the National Library of Finland is


As HTTP URI it becomes actionable

This URL is protocol dependent and therefore not nearly as persistent as the URN. If and when URN Global resolver discovery service is in place, URNs may become actionable without the protocol dependent extension, provided that the URN namespace specific string contains a “hint” which makes it possible to find the right URN resolver. All URNs the national library of Finland has assigned are like that.

Anyone planning to use IRIs should be cautious because the standardization status of IRIs is confused. The IETF concluded that RFC 3987 is inadequate and established a working group the revise the RFC, but the WG could not reach consensus on how to revise the specification and gave up.

The problems with IRIs are related to, among other things, security, internationalization and weaknesses in the URI specification (RFC 3986).


From: Bibliographic Framework Transition Initiative Forum [mailto:[log in to unmask]] On Behalf Of Ray Denenberg
Sent: 10. kesäkuuta 2016 2:25
To: [log in to unmask]
Subject: Re: [BIBFRAME] bf:Urn

I meant  “ BIBFRAME only http URIs ….”

On Jun 9, 2016, at 7:16 PM, Simon Spero <[log in to unmask]<mailto:[log in to unmask]>> wrote:

On Jun 9, 2016 6:43 PM, "Ray Denenberg" <[log in to unmask]<mailto:[log in to unmask]>> wrote:
>   A Urn is in fact a URI, but that’s irrelevant.  It is not an HTTP URI.   HTTP URIs, and only HTTP URIs, are used to identify resources that are subjects or objects of RDF statements.
That's not strictly correct.
Resource names in subjects or objects can be any absolute IRI. See §3.2 of the RDF concepts and Abstract Syntax recommendation for details.
The IRI <🐤> is perfectly valid in  subject, predicate, or object positions.