Section D.3.2 Review

[bkasdorf]

Name: authority
Description:
The authority property identifies the system or scheme the referenced element's value is drawn from.

Allowed value(s):
one of the following case-insensitive reserved authority values:

AAT
The Getty Art and Architecture Taxonomy.

BIC
The main UK subject scheme for the book supply chain.

BISAC
The main US subject scheme for the book supply chain.

CLC [Link no longer works.]
The main subject classification scheme used in China.

DDC
The Dewey Decimal Classification system.

CLIL
The main French subject scheme for the book supply chain.

EuroVoc
The European multilingual thesaurus.

MEDTOP
IPTC Media Topics classification scheme for the news industry.

LCSH
Library of Congress Subject Headings.

NDC [This might be a better link: https://www.jla.or.jp/committees/bunrui/tabid/187/Default.aspx]
The main Japanese subject scheme.

Thema
The international subject scheme for the book supply chain, providing codes that work across many languages.

UDC
The Universal Decimal Classification system.

WGS [Returns a 404; this might be a better link: https://vlb.de/. Or somebody may be able to dig more deeply to the vocabulary itself. There appear to be various versions of the vocabulary, and the high-level link I suggested provides access to all of them.]
The main German subject scheme for the book supply chain.

an absolute-URL string [URL] that identifies the scheme. The URL SHOULD refer to a resource that provides more information about the scheme.

Cardinality: zero or one
Extends: subject
EXAMPLE 84: Expressing A BISAC subject heading
<metadata …>
…
<dc:subject
id="subject01">
FICTION / Occult & Supernatural
</dc:subject>

BISAC

…

[OriIdan]

In the IDPF I think we decided to not act as naming authority.
Please add HB - Helicon Books Israel
We act as de-facto naming authority in Israel.

[iherman]

CLC [Link no longer works.]
The main subject classification scheme used in China.

I found this: https://www.isko.org/cyclo/clc, which also refers to http://clc.nlc.cn/ztfls.jsp (I cannot check the validity of that link, because it is in Chinese. @Jeffxz could you help us out here? Is it o.k. to use this as a reference for CLC?)

NDC [This might be a better link: https://www.jla.or.jp/committees/bunrui/tabid/187/Default.aspx]
The main Japanese subject scheme.

@shiestyle or @murata2makoto, can you confirm whether that link is the right one? The page has an English version which seems to be o.k., but there may be a another link as an official reference...

WGS [Returns a 404; this might be a better link: https://vlb.de/. Or somebody may be able to dig more deeply to the vocabulary itself. There appear to be various versions of the vocabulary, and the high-level link I suggested provides access to all of them.]
The main German subject scheme for the book supply chain.

Indeed, this seems to be a fairly high level reference... @WSchindler can you help us out to see if this is the right reference?

[iherman]

Please add HB - Helicon Books Israel
We act as de-facto naming authority in Israel.

I do not know how IDPF decided to add entries to this list (or not). If there is a decision to add this entry (I am personally fine with it) then we would need a link to a page, @OriIdan

[iherman]

This list very much looks like a registry to me. I do not think we should go down the road of making an official registry, but I am a bit worried about this list being part of a recommendation, whose change policies may be complicated on long term in maintaining this list... This may be a more general issue, @dauwhe @shiestyle @wareid

[OriIdan]

I also think the list should not be part of the recommendation. If we do go and define a registry we have to define update policies, update frequency etc.

…

On Sun, Apr 3, 2022 at 1:17 PM Ivan Herman ***@***.***> wrote: This list very much looks like a registry to me. I do not think we should go down the road of making an official registry, but I am a bit worried about this list being part of a recommendation, whose change policies may be complicated on long term in maintaining this list... This may be a more general issue, @dauwhe <https://github.com/dauwhe> @shiestyle <https://github.com/shiestyle> @wareid <https://github.com/wareid> — Reply to this email directly, view it on GitHub <#2200 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAB43QGT32BNKX7DO6ILJRLVDFV4PANCNFSM5SLJ2LBQ> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[mattgarrish]

I believe this list was compiled by one of the metadata task forces, but I can't remember which revision it was in. Probably 3.1. I don't know if @laudrain is still paying attention to these discussions, but my memory is he led the group.

As far as I'm aware, this is not meant to be a general registry. It's probably not clear in this issue, but there are two possibilities: one is to use the predefined values for what the group established as major regional taxonomies and the other is to use a URL. The latter option is open to anyone, so Helicon can always specify to use their URL when using their classification system.

At this stage, I personally would not expand the token list any further, and we may want to make clear that the first option is no longer available. We don't have a metadata task force in place to judge new classification systems, and I'm not sure it's a task we want hanging over the WG forever.

[iherman]

At this stage, I personally would not expand the token list any further,

I would agree with this. The list is historical (maybe emphasize the historical nature of it in the text?) and we do not have the structure for evolving it any further.

and we may want to make clear that the first option is no longer available. We don't have a metadata task force in place to judge new classification systems, and I'm not sure it's a task we want hanging over the WG forever.

I think this issue should be discussed on the WG call @wareid @dauwhe @shiestyle

[toshiakikoike]

NDC [This might be a better link: https://www.jla.or.jp/committees/bunrui/tabid/187/Default.aspx]
The main Japanese subject scheme.

@shiestyle or @murata2makoto, can you confirm whether that link is the right one? The page has an English version which seems to be o.k., but there may be a another link as an official reference...

If this means "the management organization of NDC",
=> https://www.jla.or.jp/
If this means "the description of NDC",
=> https://www.jla.or.jp/committees/bunrui/tabid/187/Default.aspx
If this means name space,
=> http://jla.or.jp/vocab/ndcvocab#

[iherman]

NDC [This might be a better link: https://www.jla.or.jp/committees/bunrui/tabid/187/Default.aspx]
The main Japanese subject scheme.

@shiestyle or @murata2makoto, can you confirm whether that link is the right one? The page has an English version which seems to be o.k., but there may be a another link as an official reference...

If this means "the management organization of NDC", => jla.or.jp If this means "the description of NDC", => jla.or.jp/committees/bunrui/tabid/187/Default.aspx If this means name space, => http://jla.or.jp/vocab/ndcvocab#

Thanks a lot, @toshiakikoike! I think what we would need, in line with the others, is the second link. We will make it sure to update this whenever we make a decision on the general fate of this part of the document.

[Jeffxz]

@iherman the link http://clc.nlc.cn/ztfls.jsp is available and legit.
I am trying to find out the context of linking to http://clc.nlc.cn/ztfls.jsp. It is website of introducing history of the Chinese Library Classification so I am thinking you might just want to use http://clc.nlc.cn (which is entry page for the Chinese Library Classification) instead of http://clc.nlc.cn/ztfls.jsp

[iherman]

The issue was discussed in a meeting on 2022-04-08

List of resolutions:

• Resolution No. 3: Do a final update of the IDPF registry, add note that it will no longer be updated, accept PR 2233, close issue 2200.

View the transcript

3. Section D.3.2 Review (issue epub-specs#2200)

See github issue epub-specs#2200.

See github pull request epub-specs#2233.

Matt Garrish: this was basically a mini-registry that IDPF was making, we said you can use certain limited token values unless you provide a URL.
… there was a request to add to this mini-registry, but my preference is rather to drop restrictions on registry values.
… add note that we are no longer maintaining this.

Ivan Herman: in addition, mgarrish and I went through a number of older IDPF documents.
… Bill_Kasdorf noted that that there are a number of IDPF documents linked to older versions of spec.
… so I updated these links.
… in doing this we found the mini-registry, which is what this issue is about.
… so the proposal is that we use the same IDPF registries that were there before.
… in the meantime, we've gone through the registry values and updated links.
… we've reached out to various wg members, from JP, DE, etc. to ask for clarification on which links would be proper to use.
… zheng_xu_ if you can please help interpret the link to the chinese website, to see if it is the correct one.
… I wonder if there are other IDPF documents that need updating in the same way, but I'm not sure what those would be.

Matt Garrish: main ones are the vocabs, the CMT, so I think we got most of them.

Murata Makoto: do we need to finalize those changes to the IDPF registry today?

Ivan Herman: not right now at this meeting, but I'd like to have all of these changes behind us by next week.
… if you could look at the pointer that Koike-san provided by then, that would help.

Murata Makoto: are there EN versions of the data linked via these links?

Ivan Herman: I'm not sure, but these documents only need to be in their native languages.

Shinya Takami (高見真也): the first link from Koike-san's comment is good for these needs.
… the second one is more an introduction from the committee in charge, so it may not be correct for this use.

Ivan Herman: okay, I will wait until next week before finalizing any changes to the IDPF registry.

Murata Makoto: I will make my comments in the issue.

Ivan Herman: okay, thank you.

Matt Garrish: noting that there is a PR to resolve this.

Proposed resolution: Do a final update of the IDPF registry, add note that it will no longer be updated, accept PR 2233, close issue 2200. (Wendy Reid)

Wendy Reid: +1.

Matt Garrish: +1.

Matthew Chan: +1.

Ivan Herman: +1.

Charles LaPierre: +1.

GeorgeK: +1.

Murata Makoto: +1.

Dan Lazin: +1.

Dave Cramer: +hbg:1.

Bill Kasdorf: +1.

Ben Schroeter: +1.

Masakazu Kitahara: +1.

Resolution #3: Do a final update of the IDPF registry, add note that it will no longer be updated, accept PR 2233, close issue 2200.

[iherman]

Reopening temporarily until the registry document is updated with the various links.

[murata2makoto]

I asked a member of the Japanese Diet Library in the latest meeting of the Japan DAISY Consortium technical committee. He is now studying this issue.

[murata2makoto]

The latest edition of NDC is available as a printed book, as shown at:
http://www.jla.or.jp/Default.aspx?TabId=87&pdid=p11-0000000413

I got this information from my colleagues in the JDC technical committee, who work for the Japanese Diet Library. Thanks, Uemura-san and Ando-san.

[shiestyle]

The latest edition of NDC is available as a printed book, as shown at: http://www.jla.or.jp/Default.aspx?TabId=87&pdid=p11-0000000413

I think this is not proper URL because this will be old when the new one will be published.

It’s better to use organization’s URL, isn’t it?

[iherman]

@shiestyle the fact that it is an ugly URL is not really an issue, it is not a namespace URL. Looking at the current registry items, some URL-s list to organizations indeed, while others are similar; e.g., the French reference is https://clil.centprod.com/information/detailDoc.html?docId=34

But, indeed, the fact that it may become obsolete is a problem; a stable URL is preferred.

[iherman]

@shiestyle @murata2makoto the current registry uses https://www.jla.or.jp/ which indeed looks like the organization's URL.

[murata2makoto]

@iherman Unfortunately, https://www.jla.or.jp/ does not say anything about NDC. But both https://www.jla.or.jp/committees/bunrui/tabid/187/Default.aspx and https://clil.centprod.com/information/detailDoc.html?docId=34 may become obsolete at some point.

[shiestyle]

@murata2makoto We need a solution now. Do you think the URL to the current version’s printed book is proper for this case?

@iherman I think this is the better one for now.
https://www.jla.or.jp/committees/bunrui/tabid/187/Default.aspx

Although this one is more stable indeed.
https://www.jla.or.jp/

[murata2makoto]

I can live with https://www.jla.or.jp/committees/bunrui/tabid/187/Default.aspx

[iherman]

Thank you @murata2makoto and @shiestyle. I will take jla.or.jp/committees/bunrui/tabid/187/Default.aspx

[iherman]

The IDPF registry documents have been finalized, marked for archival, and the repository archived. Closing.