Colleagues
A word of encouragement to jump in with more of the relationship designator discussion. I am looking into the RDF representation of the RDA relationship designators as a follow-up to recommendations to JSC [1]. Outcomes of JSC’s discussion of these recommendations are given informally on the ALA representative’s blog [2].
Although extension to the lists of designators is not in direct scope, I think it will inform the work, so I’d like to hear of any issues arising. The potential need for extensions was discussed during a recent e-Forum organized by the UK’s Cataloguing and Indexing Group [3], so this appears to be a widespread requirement.
We can move discussion to the DCMI Bibliographic Metadata Task Group if it gets too specific.
Cheers
Gordon
[1] http://www.rda-jsc.org/docs/6JSC-CILIP-rep-2.pdf
[2] http://www.personal.psu.edu/jxa16/blogs/resource_description_and_access_ala_rep_notes/2012/11/report-of-the-meeting-of-the-joint-steering-committee-6-november-2012.html
[3] https://www.jiscmail.ac.uk/cgi-bin/webadmin?A0=CIG-E-FORUM [then search archives for October 2012]
From: DCMI Vocabulary Management Community [mailto:[log in to unmask]] On Behalf Of Chew Chiat Naun
Sent: 20 November 2012 15:40
To: [log in to unmask]
Subject: Re: Vocabulary extension
Thanks once again for inviting us to post here and for taking the discussion further. I'm hoping that others involved in the relationship designator discussion who I believe are lurking here will be encouraged to jump in as well.
1. They (an individual, an institution, a specialist community, or a maintenance agency) wish to add a term for a concept that is not already represented in the base vocabulary. There are at least two ways they could do this:
a. By maintaining a supplementary list of terms
b. By proposing the term for inclusion in the base vocabulary. The status of such terms may depend on the submission process used by the relevant agency. For example, the Program for Cooperative Cataloging Subject Authority Cooperative Program (SACO) considers proposed terms “pre-approved” pending evaluation.
It seems to me that there are several factors here that would determine what options are available to a community or institution. First, does the vocabulary have a process for proposing new terms? Does that process include all communities or institutions with needs for new terms, or just a subset? These questions are relevant in your 1b, above, but I think the important factor is whether there IS a process, and whether that process is open or limited in some way. I would call these social, rather than technical concerns.
Your 1a seems to have a number of facets. Certainly anyone can maintain a list of terms, either free-standing or extending another list. But the usefulness of this strategy is based on a number of important factors. Is that list available only within a particular institution or is it publicly available? If publicly available is it in a form that is usable within a variety of applications? Do the terms have URIs and are they freely usable, maintained and documented by the organization that developed the supplementary list? Are they structured in a way that makes the relationships between the two vocabularies available to applications? Many of these questions relate more to technical decisions made by the developers of the supplementary vocabulary.
We spent a lot of time in our PCC group discussing the processes that we now have in place for developing the list of relationship terms, what alternative models are available in the current environment (this is where the SACO process entered the discussion), and what the needs are for different stakeholder communities - for example, whether a group like RBMS would want to maintain their own vocabulary - as well as what this might imply for their relationships with JSC on the one hand and PCC on the other.
Our current practices are constrained by our existing environment, which includes the fact that MARC does not have the ability to flag different source vocabularies for relationship designators. I think it's fair to say we saw potential for what you call our social practices to evolve quite significantly if our technical options widened, and this is what makes this discussion so interesting to some of us.
A couple of follow-up observations on the use cases we've been discussing:
3. They introduce domain-specific application guidelines for certain existing terms in the base vocabulary. The guidelines may be consistent with the original definitions, but may give more specific instructions. Example (proposed extension of RDA relationship designators from RBMS Controlled Vocabularies subcommittee):
• Former owner: A person, family, or corporate body formerly having legal possession of an item (i.e., a specific copy or instance of a resource).
• Donor: A former owner of an item who donated that item to another owner.
• Inscriber: A person who has written a statement of dedication or gift on an item.
Domain-specific application guidelines:
• Donor: Use for the entity who donates a book, manuscript, etc., to the present owner; donors to previous owners are designated as former owner or inscriber
It seems to me that this is a use case where an Application Profile is a more appropriate option than a vocabulary extension. The usage guidelines don't change the existing definition, but instead define the usage for a particular community.
That's a helpful distinction. In practice, I could imagine the two kinds of activity co-existing as part of the same effort -- e.g. a community defining usage for some existing terms while introducing others that are new.
4. They introduce supplementary terms that may not in fact be appropriate for inclusion in the base vocabulary itself, e.g. for cultural or language reasons. For example, the Council on East Asian Libraries’ Committee on Technical Processing gives this example of a relationship designator not provided for in RDA:
Qianlong, Emperor of China, 1711-1799, chi zuan.
乾隆, Emperor of China, 1711-1799, 勅纂. [compiled under his order—
only used after names of emperors]
This is a really interesting use case, because it includes the problem of a term lacking in a standard, but one that seems to apply in a particular language context. It strikes me, though, that the question of how to designate writings or proclamations of heads of state generally, as distinguished from the private writings of that person, is a fairly common issue in traditional cataloging. As I recall from my previous life as a cataloger, the distinction is made at the level of heading, so that there are generally two headings for heads of state and rulers of all kinds, one to be used for official things, one for personal. I'm not really familiar about how this particular problem is handled in RDA. If the guidelines are ported directly from AACR2 instead of being handled in the context of the relationship terms, then an addition to the relationship vocabularies won't solve the problem. I really don't think the issue you've described is specific to the context of Chinese emperors (though it may be that the East Asian catalogers are the first to articulate it).
RDA makes a similar distinction between the same person acting in an official and non-official capacity, and that distinction certainly isn't confined to Chinese emperors. I'm not sure this was the main point my CEAL colleagues intended to raise, though. I think they were raising a more basic question about how to allow for relationship designators that the originating agency may not regard as being in scope for the base vocabulary. The CEAL group provided a second example which I didn't include in my first post but may illustrate the point more clearly. I'm told that the heading below is for a publisher, and specifically one who produced the original (i.e. woodblock) impression of the publication:
710 2 善成堂, ǂe 藏版.
710 2 Shan cheng tang, ǂe cang ban. [woodblock owner]
Putting aside the question of how to maintain equivalent vocabularies in different languages or scripts, there's no reason in principle why the makers of RDA should not regard "woodblock owner" as a legitimate addition to the RDA relationship vocabulary, to be submitted and approved through the standard channels. But it's just as possible that they or CJK cataloguers would regard the maintenance of such terms as an activity best devolved to the CJK community. So maybe (4) is really a special case of (1), where as you say, "the choices are best differentiated by what the relationship is between the vocabulary and the organization with needs for extension".
