RSC Paper on RDA Conformance

Gordon Dunsire,
RSC Technical Team Liaison Officer
May 12, 2020

The RSC is soliciting feedback on the following paper that will be discussed at the July asynchronous meeting: RDA conformance.

This informative paper explains what is RDA-conformant metadata, “presents a draft expansion of RDA Toolkit guidance to accommodate the topic,” and asks three questions:

“Question 1: Should the RSC offer conformance certification in addition to the passive guidance included in the Toolkit?

Question 2: Should the Toolkit include content for assessing conformance for specific application communities? Content might include examples, alignments and mappings for parsing metadata statements, and decision trees.

Question 3: Should the RSC offer additional technical guidance and support for conformance, interoperability, and mappings between RDA metadata and non-RDA applications?”

Print Friendly, PDF & Email

Tagged , , . Bookmark the permalink.

3 Responses to RSC Paper on RDA Conformance

  1. Thomas Dousa says:

    A small typographical comment: On page 8, in the first sentence of the first paragraph of the section on RDA Relationship Data, the punctuation of the clause “In the IFLA library reference model, (LRM) relationships are an essential part …” should be changed to “In the IFLA library reference model (LRM), relationships are an essential part …”

  2. Keith Knop says:

    MLA responses:

    Q1: It is not clear us from the paper what the purpose or desired outcome of offering certification would be. A respondent notes:
    The only other mention of certification in this paper is the sentence “Some communities may wish for some form of certification of conformance.” What would that look like? What can be done in an automated way, especially since, as the paper points out, RDA instructions have changed over time? Or is this all manual work? How would the certification be used? Where would it be reflected? Is certification granted at the community level? At the metadata description set level? At the metadata statement level?

    Q2: Is the intention that RSC would develop this content, or only that it would be hosted in the Toolkit? If so, how would that differ from existing or planned community documentation and policy statements? Would this be similar, for example, to the MLA or OLAC best practices documents?

    Q3: No strong opinions, but in what form is the guidance and support being proposed? Documentation? An automated tool? Personal advising?

    It feels odd to rename the entire chapter from “Data elements” to “Data conformance” when only one segment of the chapter explicitly addresses conformance while the rest focuses on how entities, elements, and relationships function and only addresses conformance implicitly. That seems to obfuscate a good portion of the content.

    An additional question and proofing remarks from a respondent:
    On p. 7, under Legacy elements and deprecation, is there an implied AND or OR between the two bulleted conditions? Could it specify? I note that in the conformance section on p. 5, it uses the very specific language “if all of the following requirements are met.”

    p. 6, under RDA entities, in the bulleted list of two exceptions, the second exception needs a period at the end.

    p. 7, under Choice of elements, “Guidance: Application profiles” needs a period at the end.

    p. 8, under RDA relationship data, in the first sentence, the comma should be after “(LRM).”

    p. 9, top of the page, the bullet point that starts with “The broadest level or granularity” is probably supposed to be “…of granularity.”

  3. Kathryn Lybarger says:

    I think a method of validating RDA metadata as conforming in some ways to the standard (e.g. including minimal elements) is needed for interoperability, and know that some of this can be automated, but this part gives me pause: “The statement records a value that is compatible with the RDA guidance and instructions” — just in data type, or in content as well? If content, would a person need to check the records?

    I think we DO need a way to validate structure in RDA records if we are to import or share them, but I don’t think this notion should be so prominently featured on the Toolkit, or we should at least be very clear about what it means. If we’re only validating structure, I worry that people will mistake “certified RDA conformant” with “having the right content”. (I’m currently working with a big batch of MARC records that are marked as being AACR2, but which have 100 fields like “Anon.”)

Leave a Reply