Discussion:
[urn] URN namespace for NBNs
John C Klensin
2018-05-03 19:59:40 UTC
Permalink
Hi.

While I suggested that the URN mailing list have a look at
draft-hakala-urn-nbn-rfc3188bis-00 in my note of 23 April
forwarding the Last Call announcement to the list, Peter has
reminded me that Expert Review was never formally requested.
So, to be sure that ducks are lined up and we don't have any
last-minute surprises, I'm formally asking the Expert Review
panel to do their thing and inviting mailing list recipients to
comment as appropriate.

Noting that, from IANA's standpoint, this is an update to an
existing reservation and that the basic transition issues are
discussed in Section 3 of RFC 8254, I believe that this request
for review is strictly pro forma. But, if anyone has comments,
I hope the members of the expert review team, especially those
who were not involved in the construction of this document, will
pay careful attention.

URN namespace template review closes concurrent with the Last
Call on the document, i.e., on 2018-05-21.

thanks,
john
Peter Saint-Andre
2018-05-04 02:02:31 UTC
Permalink
On 5/3/18 1:59 PM, John C Klensin wrote:
> Hi.
>
> While I suggested that the URN mailing list have a look at
> draft-hakala-urn-nbn-rfc3188bis-00 in my note of 23 April
> forwarding the Last Call announcement to the list, Peter has
> reminded me that Expert Review was never formally requested.
> So, to be sure that ducks are lined up and we don't have any
> last-minute surprises, I'm formally asking the Expert Review
> panel to do their thing and inviting mailing list recipients to
> comment as appropriate.
>
> Noting that, from IANA's standpoint, this is an update to an
> existing reservation and that the basic transition issues are
> discussed in Section 3 of RFC 8254, I believe that this request
> for review is strictly pro forma. But, if anyone has comments,
> I hope the members of the expert review team, especially those
> who were not involved in the construction of this document, will
> pay careful attention.
>
> URN namespace template review closes concurrent with the Last
> Call on the document, i.e., on 2018-05-21.

Robert Sparks pointed out to me offlist that the ABNF might not be quite
right, in particular nbn_string. I suggest:

OLD

nbn_string = <specific per prefix>
; MUST adhere to RFC 3986 <path-rootless> syntax;
; parsers must regard nbn_strings as case-sensitive

[...]

Whereas the prefix is regarded as case-insensitive, NBN-strings
MAY be case-sensitive at the preference of the assigning
authority; parsers therefore MUST treat these as case-sensitive;
any case mapping needed to introduce case-insensitivity MUST be
implemented in the responsible resolution system.

NEW

nbn_string = path-rootless
; the "path-rootless" rule is defined in RFC 3986

[...]

The structure (if any) of the nbn_string is determined by the
authority for the prefix. Whereas the prefix is regarded as
case-insensitive, the nbn_string MAY be case-sensitive at the
preference of the assigning authority; parsers therefore MUST
treat these as case-sensitive; any case mapping needed to
introduce case-insensitivity is the responsibility of the relevant
resolution system.

Peter
Dale R. Worley
2018-05-04 02:36:54 UTC
Permalink
The draft looks fine to me. There are various nits, which I've listed
below.

Dale

----------------------------------------------------------------------

Various terminology like "hand-held", "immaterial", and "digital" is
used. Do you have a clear terminology in mind?

2. Conventions used in this document

"NBN" refers to any National Bibliography Number identifier system
used by the national libraries and other institutions, which use
these identifiers with the national library's support and permission.

s/library's/libraries'/ since it is "support of libraries". (Check
with the Editor!)

3.1. The URN:NBN Namespace

For instance, even if a digitized book has an ISBN, JPEG
image files of its pages get NBNs.

You probably want to say "may get".

3.2. Community Considerations for NBNs

Note that the f-component is not a part of the NSS and therefore the ...

Resources identified by NBNs are not always available in the ...

If an NBN identifies an immaterial work, descriptive metadata about ...

These three paragraphs seem to be more about the resolution process
than community considerations. Perhaps they can be separated in some
way?

4.1. Overview

uniqueness of the URN:NBNs at the global scale [Iso3166MA"/>.

This reference is corrupted.

4.2. Encoding Considerations and Lexical Equivalence

When an NBN is used as a URN, the namespace-specific string (NSS)
MUST consist of three parts:

o a prefix, structured as a primary prefix, which is a two-letter
ISO 3166-1 country code, ...

This doesn't specify *what* the country code must be. Instead,

When an NBN within a national library's identifier system is used
as a URN, the namespace-specific string (NSS) MUST consist of three
parts:

o a prefix, structured as a primary prefix, which is the two-letter
ISO 3166-1 country code of the library's country, ...

5. URN Namespace ID (NID) Registration for the National Bibliography
Number (NBN)

Declaration of syntactic structure of NSS part:

nbn_string = <specific per prefix>
; MUST adhere to RFC 3986 <path-rootless> syntax;
; parsers must regard nbn_strings as case-sensitive

There is trouble with <...> here.

Colon MAY be used as a delimiting character only within the
prefix, between ISO 3166-1 country code and sub-namespace code(s),
which split the national namespace into smaller parts.

Comparing with the text in 4.2, I don't think this is phrased
correctly, as it may imply that colon may not appear in the
nbn_string.

Reading it as: Colon MAY be used (as a delimiting character
only within the prefix) ...

Instead:

Colon MAY be used within the prefix only as a delimiting
character between the ISO 3166-1 country code and the
sub-namespace code(s), which split the national namespace into
smaller parts.

--

See Section 4.2 of RFC XXXX for examples.

There should be a note telling the RFC Editor to insert the RFC number here.

Process for identifier resolution:
See Section 4.3 of RFC XXXX.

Ditto.

Rules for lexical equivalence of NSS part:
...
Formally, two URN:NBNs are lexically equivalent if they are octet-
by-octet equal after the following (conceptional) preprocessing:

1. normalize the case of the leading "urn:" token;

This should be 'the leading "urn:nbn:" token', as otherwise the "nbn"
is not mentioned!

8. Acknowledgements

Revision of RFC 3188 started during the project PersID [PERSID] Later

Insert a full-stop after the reference.

9. Contributors

Alfred Hoenes was the editor and co-author of two of the documents
from which this one is, in part, derived.

Is it possible to provide a description or reference to these
documents? (One appears to be in the references already.)

Appendix A. Significant Changes from RFC 3188

Updated URN:NBN Namespace Registration template for IANA; whole
document adapted to new URN Syntax document, RFC 2141bis, and new URN
Namespace Registration document, RFC 3406bis (now retired and merged
into 2141bis.

This paragraph is hard to read, and it has unbalanced parentheses.

Use of query directives and fragment parts with this Namespace is now
specified, in accordance with the aforementioned RFCs.

The terminology in RFC 8141 is "query components" and "fragment components".

[END]
Hakala, Juha E
2018-05-04 08:30:15 UTC
Permalink
Hello Dale; all,

Thank you very much for reviewing the draft thoroughly! I have made almost all changes requested by you, Alexey and Peter. But I may have misunderstood something so it is necessary to check the attached result. And instead of fixing broken links I may have created a few more ;-).

There were two competing suggestions on what to do with ABNF; I have used Peter's version.

I do not have better terms with which to replace hand-held, immaterial or digital. They are commonly used and well understood within the library community. Work such as Hamlet is immaterial, but it becomes hand-held when it is published as a manifestation such as a printed book. PDF version of Hamlet is another manifestation of the work, not hand-held but digital / electronic.

Under community concerns there were paragraphs which from your point of view dealt with the resolution process. I have modified the text in order to make it more clear why to me these things are (also and perhaps primarily) community concerns.

The draft specified three different ways in which NBNs can be used to identify component parts of resources (assign separate NBN to each component part, use local NSS syntax to indicate component parts within a resource, all sharing the same base NBN belonging to the resource itself, and using URN f-component). I deleted the middle option because I am not aware of any national library using it and without giving a real example it is hard to understand how these URN:NBNs would actually work.

I have added references to the new, RFC 8141 -compliant ISBN and ISSN namespace registrations and deleted the reference to I-D which was an early version of an updated ISBN namespace registration.

Alfred Hoenes is still mentioned as a contributor but the text is shorter. There was no point to refer to any documents especially after I dropped the reference to the outdated URN namespace revision request I-D he worked on..

All the best,

Juha


-----Original Message-----
From: Dale R. Worley <***@ariadne.com>
Sent: perjantai 4. toukokuuta 2018 5.37
To: Hakala, Juha E <***@helsinki.fi>
Cc: John C Klensin <***@jck.com>; ***@dnb.de; ***@stpeter.im; ***@isode.com; ***@ietf.org
Subject: Re: [urn] URN namespace for NBNs

The draft looks fine to me. There are various nits, which I've listed below.

Dale

----------------------------------------------------------------------

Various terminology like "hand-held", "immaterial", and "digital" is used. Do you have a clear terminology in mind?

2. Conventions used in this document

"NBN" refers to any National Bibliography Number identifier system
used by the national libraries and other institutions, which use
these identifiers with the national library's support and permission.

s/library's/libraries'/ since it is "support of libraries". (Check with the Editor!)

3.1. The URN:NBN Namespace

For instance, even if a digitized book has an ISBN, JPEG
image files of its pages get NBNs.

You probably want to say "may get".

3.2. Community Considerations for NBNs

Note that the f-component is not a part of the NSS and therefore the ...

Resources identified by NBNs are not always available in the ...

If an NBN identifies an immaterial work, descriptive metadata about ...

These three paragraphs seem to be more about the resolution process than community considerations. Perhaps they can be separated in some way?

4.1. Overview

uniqueness of the URN:NBNs at the global scale [Iso3166MA"/>.

This reference is corrupted.

4.2. Encoding Considerations and Lexical Equivalence

When an NBN is used as a URN, the namespace-specific string (NSS)
MUST consist of three parts:

o a prefix, structured as a primary prefix, which is a two-letter
ISO 3166-1 country code, ...

This doesn't specify *what* the country code must be. Instead,

When an NBN within a national library's identifier system is used
as a URN, the namespace-specific string (NSS) MUST consist of three
parts:

o a prefix, structured as a primary prefix, which is the two-letter
ISO 3166-1 country code of the library's country, ...

5. URN Namespace ID (NID) Registration for the National Bibliography
Number (NBN)

Declaration of syntactic structure of NSS part:

nbn_string = &lt;specific per prefix&gt;
; MUST adhere to RFC 3986 &lt;path-rootless&gt; syntax;
; parsers must regard nbn_strings as case-sensitive

There is trouble with <...> here.

Colon MAY be used as a delimiting character only within the
prefix, between ISO 3166-1 country code and sub-namespace code(s),
which split the national namespace into smaller parts.

Comparing with the text in 4.2, I don't think this is phrased correctly, as it may imply that colon may not appear in the nbn_string.

Reading it as: Colon MAY be used (as a delimiting character
only within the prefix) ...

Instead:

Colon MAY be used within the prefix only as a delimiting
character between the ISO 3166-1 country code and the
sub-namespace code(s), which split the national namespace into
smaller parts.

--

See Section 4.2 of RFC XXXX for examples.

There should be a note telling the RFC Editor to insert the RFC number here..

Process for identifier resolution:
See Section 4.3 of RFC XXXX.

Ditto.

Rules for lexical equivalence of NSS part:
...
Formally, two URN:NBNs are lexically equivalent if they are octet-
by-octet equal after the following (conceptional) preprocessing:

1. normalize the case of the leading "urn:" token;

This should be 'the leading "urn:nbn:" token', as otherwise the "nbn"
is not mentioned!

8. Acknowledgements

Revision of RFC 3188 started during the project PersID [PERSID] Later

Insert a full-stop after the reference.

9. Contributors

Alfred Hoenes was the editor and co-author of two of the documents
from which this one is, in part, derived.

Is it possible to provide a description or reference to these documents? (One appears to be in the references already.)

Appendix A. Significant Changes from RFC 3188

Updated URN:NBN Namespace Registration template for IANA; whole
document adapted to new URN Syntax document, RFC 2141bis, and new URN
Namespace Registration document, RFC 3406bis (now retired and merged
into 2141bis.

This paragraph is hard to read, and it has unbalanced parentheses.

Use of query directives and fragment parts with this Namespace is now
specified, in accordance with the aforementioned RFCs.

The terminology in RFC 8141 is "query components" and "fragment components"..

[END]
Peter Saint-Andre
2018-05-04 18:10:32 UTC
Permalink
On 5/4/18 2:30 AM, Hakala, Juha E wrote:
> Hello Dale; all,
>
> Thank you very much for reviewing the draft thoroughly! I have made almost all changes requested by you, Alexey and Peter. But I may have misunderstood something so it is necessary to check the attached result.

Thanks as always, Juha.

Unfortunately, the registration template is not quite consistent with
the new template in RFC 8141 (compare the section headings in Appendix A
https://tools.ietf.org/html/rfc8141#appendix-A against the headings in
Section 5 of draft-hakala-rfc3188-nbn-00). Bringing your template into
harmony with RFC 8141 is mostly editorial; the only substantive task is
defining the security and privacy considerations. Let me know if (as the
document shepherd) I can help with this.

Peter
Dale R. Worley
2018-05-05 13:54:05 UTC
Permalink
"Hakala, Juha E" <***@helsinki.fi> writes:
> I do not have better terms with which to replace hand-held, immaterial
> or digital. They are commonly used and well understood within the
> library community. [...]

Ah! I was unaware that those are "terms of art" within the
library/bibliographic community. By all means, continue to use them.

I was only concerned with a problem that shows up in a lot of drafts,
where the authors haven't settled on terminology and use multiple terms
for the same thing throughout the draft. But your community has a set
terminology...

Dale
Loading...