Proposal: Clarifying instructions for Recording Duration (RDA 7.22) and Note on Carrier (RDA 3.21)

Tagged , , . Bookmark the permalink.

18 Responses to Proposal: Clarifying instructions for Recording Duration (RDA 7.22) and Note on Carrier (RDA 3.21)

  1. Robert Rendall says:

    Comments from Francis Lapka, Yale University (and co-chair of CC:DA’s Task Force on Machine-Actionable Data):

    Some additional thoughts, not necessarily endorsed by the MAD TF.

    EURIG says:

    […] We consider that Chapter 7 should describe the Expression of a Work and not the “content of the resource”, i.e. the content of a specific Manifestation.

    In consequence, the reference at the end of “See also for instructions on recording the total playing time, running time, etc. of the content of the resource” and the instructions under should be applicable only in the case of a resource containing only one Expression of one Work, without any bonus or other additional content. Otherwise, the total running time of the content of the resource should be recorded under 3.21 (i.e. at the Manifestation level), if it is available from the resource instead of or in addition to the playing time of each unit. […]

    FL: The EURIG argument is consistent with the final report of the FRBR Working Group on Aggregates (, which suggests that disparate content is aggregated at the Manifestation level – so that the aggregate running time (etc.) would only be known as an attribute of Manifestation. I think the WG/EURIG approach is not in the best interest of FRBR/RDA. We should, instead, acknowledge the centrality of aggregate expressions – where the content of a resource is aggregated in a Publication Expression, as is well articulated in the FRBRoo model.

    Fidelity to the FRBR WG/EURIG mindset would invalidate 7.15 Illustrations and 7.16 Supplementary Content, among others.

    The decision that JSC makes on this question will have considerable impact on the development of Extent of Expression. I believe that allowing for aggregate Expressions is the way forward that will most benefit user needs (while maintaining a logical/sustainable bibliographic model).

    Several constituencies object to the proposal’s suggestion to move certain instructions to chapter 3, on the grounds that anything to do with content should be kept in chapter 7. As I understand it, the primary (or a major) point of the proposal’s is to give us a way to say how the content of a resource has been distributed over multiple carrier units. Such information is known only when it materializes in Manifestations, and won’t be the same for all Manifestations of the single Expression – so I think these objections are wrong. There is a limited but necessary use case for

    DNB asks:

    The proposed changes under instruct the cataloguer to record the duration in the form preferred by the agency. Wouldn’t it be more useful to record the information in a standardized form?

    FL: The MAD TF is keen to hear JSC thoughts on this question. The closest comparable RDA guidelines are probably those for dimensions (3.5), where there is prescribed unit type (cm, usually), specificity (nearest whole cm, usually) and sometimes form (e.g. [height] x [width]). For Duration the implied prescribed unit types are hrs/mins/sec. There is no guideline for the degree of specificity, but perhaps there should be (e.g. nearest second?)? I don’t think the MAD TF would be interested in the idea of a “standardized form” if it means nothing more than display preference.

    DNB also says:

    We propose to add an example illustrating the duration of the content regardless of whether it is issued on one or more carriers. Moreover, we would appreciate an example showing how several parts of content (e.g. a movie and bonus material) are treated with respect to duration.

    FL: The MAD TF will be happy to oblige as we develop Extent of Expression. The latter request demonstrates the potential utility of a “part measured” attribute in the Extent model. The example resource could have multiple duration extents – one for the feature, and one (or more) for the bonus material.

  2. Tracey L. Snyder says:

    Here is what I emailed to Kathy about the CCC response:

    If we go with the overall approach in the proposal, I am fine with their two suggestions for I know they object to the Ch. 3 stuff, but since they do concede that the Ch. 3 stuff may go forward, I considered their suggestion for those proposed examples. It makes sense, since the Toolkit maps 3.21.2 to 500 only, not 300. But we actually intentionally varied those examples specifically to allow for recording this information in either a 300 or 500. I’m realizing that we may not have said anything about adjusting the mapping in the proposal. If the Ch. 3 stuff goes forward (which it may not, given the objections), I would be fine with either approach: either keeping the examples as we propose them and adding the 300 to the mapping, or adjusting the examples so that they all conform to the 500 style. I have a slight preference for the latter.

  3. Tracey L. Snyder says:

    My very long comment on the LC response is stuck in “awaiting moderation” mode, so I am going to try breaking it up into smaller pieces.

    • Tracey L. Snyder says:

      I have mixed feelings about standardizing the style of representing duration. I’m all for simplicity and standardization, but on a practical level, I would not be happy about having to use “min.” and “sec.” for every single track on every single audio recording I catalog, and I think there would be a huge backlash against that. I think the duration of component parts has to be recorded (or at least be allowed to be recorded) using the colon-separated style. I can see the “hr., min., sec.” style being nice for total duration, but if the JSC wants to go with just one style or the other, I would definitely go in the other direction—using colon-separated style instead of the abbreviations.

      • Tracey L. Snyder says:

        FWIW, AllMusic uses colon-separated style for both the total duration and the durations of the individual songs. So does Wikipedia. I guess LC’s proposed Alternative to use a form preferred by the agency technically would allow the colon-separated style, but I still don’t think that is a great solution, because then agencies would have to make policies on whether or not to apply the Alternative, and, more importantly, if we are not allowed to include examples that illustrate the Alternative, catalogers are going to miss it and think that they must use the abbreviations.

        • Kathy Glennan says:

          In April 2008, the JSC decided not to refer to any standard or specify any format for recording duration. (See the JSC minutes of that meeting, line 239.36). The ALA wording & examples in the proposal make it clear that RDA does not mandate a particular standard.

          ALA’s preference has always been for the colon delimited recording of time (at least back to 2007), with the idea that this is more internationally understood. However, we previously have been unable to identify a standard specific to duration (the ones identified actually specify a particular time/date).

      • Kathy Glennan says:

        What’s the rationale for a cataloging agency settling on a single format? This isn’t machine actionable (and I don’t think that anything readily eye-readable would ever be).

  4. Tracey L. Snyder says:

    In discussing this proposal, CC:DA felt strongly about separating Expression and Manifestation, which is how we wound up with the proposed Ch. 3 stuff, but LC is saying that it’s OK to rely on Manifestation somewhat in Ch. 7. I’m fine with that. It’s very practical. I think it’s just about impossible and perhaps even unwise to eradicate Manifestation stuff from Ch. 7, and LC is right that there are other places in Ch. 7 that rely on Manifestation information. So, I guess I ultimately agree with LC that we should leave Ch. 3 alone and just focus on streamlining and clarifying Ch. 7. However, I have some specific concerns about LC’s proposed

  5. Tracey L. Snyder says:

    LC’s b) and c):
    I suggest wording that would make it clearer that catalogers are not expected to routinely estimate an approximate time, because, as we discussed in Las Vegas, we don’t want catalogers thinking they have to supply an estimated performance time for scores. That’s the whole reason we were sticking with RDA’s existing structure that has different guidelines for performance time (time must be stated on the resource) than it has for playing time. I do like the simpler structure of LC’s approach, but I still think we need to account for this factor. Here is my marked-up version of LC’s wording.

    b) If the exact time is not readily ascertainable *but the time can be readily estimated*, record an estimated time preceded by approximately.

    c) if the time cannot be *readily* ascertained or *readily* estimated, omit it.

  6. Tracey L. Snyder says:

    LC’s handling of resources containing sound/moving images plus text/still images in
    I don’t think this covers it. It just refers you to the instruction that tells you how to record the text or still images, but there’s nothing to tell you how to record the sound or moving images. I’m not sure what to do about this.

  7. Tracey L. Snyder says:

    The example in the first example block would benefit from an explanatory note that would make it really clear that in this instruction, “component parts” means tracks, movements, etc., and NOT physical units. Maybe “Playing time of the individual tracks on an audio disc” or “Performance time of the individual movements of a musical work in a score.” Or one example of each.

  8. Tracey L. Snyder says:

    I don’t understand the third example.
    Also, if we are scrapping the Ch. 3 stuff, can we include the “Playing time of the separate discs in a set of 4 audio discs” example here? (The one that starts with 73 min., 33 sec. in our proposed

    • Kathy Glennan says:

      For that 3rd example, would “total duration” make more sense?

      I agree on the idea of moving that example from, if the JSC doesn’t approve our Chapter 3 proposal.

  9. Tracey L. Snyder says:

    As for (Sources of Info.), I still rather like our proposed revision, despite LC’s objection. I do think Duration is comparable to Colour, Sound, and Aspect Ratio, and I thought it was very clever of us to align the wording of with the Sources of Info. instructions for those elements. I do think there is a subtle but important distinction between our proposed wording and “Take information from any source.” However, I will defer to the JSC’s decision on this, since I understand LC’s point.

    • Kathy Glennan says:

      If our proposed language for isn’t accepted by the JSC, then I think the other occurrences in Chapter 7 should also be changed to just any source.

  10. Tracey L. Snyder says:

    Overall, I like LC’s approach, but I don’t think it will work without some adjustments along the lines of what I suggested above.

    • Kathy Glennan says:

      I’m inclined to prefer the ALA approach (no great surprise) in separating instructions for playing time (etc.) and performance time. In the latter case (in our proposal), you only ever take duration information from what’s stated on the resource.

Leave a Reply