RDA Content Elements

4 October 2019

RDA Content Elements

Gordon Dunsire, RSC Technical Team Liaison Officer


This paper makes recommendations for developing expression “content” elements to conform to the aggregates model. The paper does not address the need for additional elements to improve the recording of aggregating expressions and aggregating works.

Print Friendly, PDF & Email

Tagged , , , . Bookmark the permalink.

6 Responses to RDA Content Elements

  1. Dominique Bourassa says:


    The propositions, recommendations, and options listed in this paper are:

    Proposition 1: An aggregating expression does not accumulate or inherit the characteristics of the expressions that are aggregated.

    Proposition 2: No descriptive elements for an expression are applicable to an aggregating expression.

    Proposition 3: Only the relationship elements for a minimum description are applicable to an aggregating expression.

    Recommendation 1: Redefine or deprecate Expression: supplementary content because it is not conformant with the model of aggregates.

    Recommendation 2: Redefine or deprecate Expression: accessibility content because it is not conformant with the model of aggregates.

    Recommendation 3: Redefine or deprecate Expression: illustrative content and Expression: details of illustrative content because they are not conformant with the model of aggregates.

    Recommendation 4: Replace Expression: details of colour content with a new element for the specific colours found in the content of an expression.

    Recommendation 5: Redefine or deprecate Expression: sound content because it is not conformant with the model of aggregates.

    Option 1: Redefine Expression: supplementary content as a shortcut.

    Option 2: Redefine Expression: accessibility content as a shortcut

    Option 3: Add a shortcut element for Manifestation: accessibility content

    Option 4: Redefine Expression: illustrative content as a shortcut.

    Option 5: Add a scope note to Expression: colour content to restrict it to specific content types.

    Option 6: Redefine Expression: sound content as a shortcut.

    Option 7: Add a scope note to Expression: sound content to restrict it to specific content types.

    Do you agree with the propositions outlined in this paper? For instance, according to the propositions outlined in this paper, a French edition of the complete works of Shakespeare does not have French as the language of expression. Agree? Disagree?

    Do you agree with the recommendations outlined in this paper?

    Which option(s) do you prefer? Do you have other suggestions or options to recommend?

  2. Tim Kiser says:

    The options would be much easier to interpret and assess if the current Beta RDA language were included. For instance, as I understand option 4, “Illustrative content” would be changed from the current entity defined as:

    “A presence of image content that accompanies the primary content of an expression.”

    To a shortcut defined as:

    “An indication of the kinds of expression of image content that supplement the main expressions that are embodied by an augmentation aggregate.”

    I take it that the new shortcut definition also replaces the current definition of Details of illustrative content (“Details of content intended to illustrate the primary content of an expression”)? Option 4 doesn’t say that explicitly; I’m inferring it from Recommendation 3.

    I might be on board with the idea of redefining this entity as a shortcut, but I think the proposed definition would be widely disliked by catalogers.

  3. Mary Huismann says:

    Summarized comments from the music community:

    RDA Content Elements

    For illustrative content, the scope note says illustrative content applies to an aggregate that embodies an expression with certain content types; “notated music” is not included in this list. Illustrative content can apply to the content type of notated music – I’ve cataloged plenty of old sheet music with striking illustrations, that in many cases are of more interest than the music itself.

    Note: For colour content, there’s a recommendation to replace “Expression: Details of colour content” with a new element for specific colours found in the content of an expression. An option (option 5) would add a scope note to “Expression: Colour content” to restrict it to specific content types (including “notated music.” (p. 6) – which makes the exclusion of notated music from illustrative content more puzzling.

    For sound content, the recommendation is to redefine or deprecate Expression: Sound content with an option to redefine “Expression: Sound content” to a shortcut (option 6) and to add a scope note to restrict it to specific content types including “recorded music” (I believe “performed music” is intended?), “sounds” and “spoken word.”

    This is all very difficult to talk about with drawing entity-relationship diagrams, but I’ll do my best. I’m not sure how I feel about the discussion paper as a whole – it seems like aggregating expressions are the outlier, and it would be easier to slightly change the definition of an aggregating expression rather than redefining multiple expression elements.

    I’m also not sure that I agree with Proposition 2. I can see a scenario where there is supplementary content present in an aggregating expression that is not present in any of the aggregated expressions. For example, what if there was an aggregating expression of the complete works of Beethoven that also included an introductory essay that is not available elsewhere? This introductory essay is not a supplement to any of the expressions of Beethoven works included in the aggregating expression, so the aggregating expressions wouldn’t “inherit” the element of supplementary material through this shortcut. The only way to tackle it (unless I’m mistaken) would be to treat the supplementary essay as another expression in the aggregating expression – in effect, treating it as “primary” rather than “supplementary” (i.e., in accordance with the definitions at the bottom of page 2), thus treating the aggregate as a collection rather than an augmentation.

    Basically, can an aggregating expression have supplementary material that the aggregated expressions do not have? I feel like they can, or at the very least I need to see more to be convinced otherwise. I can easily see a situation where a something is a “supplement” to the aggregating expression without being a “supplement” to any of the expressions that are aggregated within it. Although it could be just that I’m misunderstanding how this is all supposed to work.

  4. Kathryn Lybarger says:

    I have a few thoughts/questions on this discussion paper:

    How are things like “Expression: illustrative content” not conformant with the model of aggregates? Is this paper suggesting replacing the shorter clearer language currently in the beta toolkit (“a presence of image…”) with that longer definition (“…augmentation aggregate”)? If so, would this new definition mean that we’re modeling ALL materials with illustrative content as “augmentation aggregates”? (So the text is one work that is a purely textual work (with no manifestations of just that version) and we augmented it with visual content from another work (that maybe doesn’t exist outside this work))? Is that the model?

    Even if it is modeled that way underneath, is it useful to describe it so precisely in the instructions for catalogers using this site as a cataloging reference? This seems like implementation detail, unless we’d be doing that kind of work in a cataloging interface.

    I’ve been reviewing the IFLA-LRM document (2017), and specifically the section on aggregates. It describes three distinct types of aggregates, which are roughly: a) creations published together (like anthologies, serials, or collected works), b) augmented works (like a book with a new foreword maybe?), and c) parallel expressions (like a book published with its translation).

    For most of these, it doesn’t seem like the typical situation to have data in place for all of the expressions contained, and just want a method to refer to them as a grouped aggregate expression. I’m also not sure why you would want to avoid describing the manifestation, or to avoid having that direct relationship between it and the aggregating expression.

    The IFLA-LRM document also says:

    “Modelling an aggregate simply as an embodiment of discrete expressions may fail to recognize the creative effort of the aggregator or editor. The process of aggregating the expressions is itself an intellectual or artistic effort and therefore meets the criteria for a work. In this sense the aggregation happens on the expression level, because only expressions can be combined (or aggregated).”

    I agree with this; I think an aggregated expression (especially that first type) is often more than a sum of its parts, with fields worth recording. For example: would an aggregated expression not have a title? (because the resulting work and manifestation surely would?)

  5. Karen Stafford says:

    It appears that recommendation to replace “Expression: description of colour content” with “Expression: colour” might no longer allow for an unstructured description of colour content. If this is indeed the implication, this would be troublesome to art catalogers. We need to be able to give nuanced descriptions of the chromatic content of the resource, which includes being able to write notes associating colors with specific parts of the resource (ex. black ink on blue paper). This information has to be understandable for a human, and therefore, it needs to be structured as an unstructured description. Would this still be available in RDA in an unstructured notes field?

    We are also troubled by the term “specific colours” in recommendation 4 and would like clarification. In a color illustration, are we to note all of the colors used? Is this only for paint or ink colors, or does it include things like paper color?

    We would also like clarification as to why “tactile” images and forms can have illustrative content but not color content.

  6. Ryan Tamares says:

    Comment from the law community:

    [The document] is so abstract and so rooted in concepts of RDF and the like, that most catalogers cannot make sense of them or understand how they would be applied. RDA is getting further and further removed from the experience of actual catalogers and will probably not be applied directly. Instead, catalogers will use secondary sources such as workflows in their cataloging.

Leave a Reply