5. Attributes

Here is a brief outline of the entities and attributes, first the entity, with the attributes below it:

  • Res
    • Category
    • Note
  • Work
    • Category
  • Expression
    • Category
    • Representivity
    • Extent
    • Intended audience
    • Rights
    • Language
    • Key (music)
    • Medium of performance
    • Scale (cartographic)
  • Manifestation
    • Category of carrier
    • Extent
    • Intended audience
    • Manifestation statement
    • Access conditions
    • Rights
  • Item
    • Location
    • Rights
  • Agent
    • Contact information
    • Field of activity
    • Language
  • Person
    • Profession/occupation
  • Collective agent (no attributes)
  • Nomen
    • Category
    • Scheme
    • Intended audience
    • Context of use
    • Reference source
    • Language
    • Script
    • Script conversion
    • Status
  • Place
    • Category
    • Location
  • Time-span
    • Beginning
    • Ending
Advertisements

12 thoughts on “5. Attributes

  1. Pingback: World-wide review of the FRBR-Library Reference Model, a consolidation of the FRBR, FRAD and FRSAD conceptual models | FRBR Open Comments

  2. Here’s a quote of a quote from my book “FRBR: Before and After”:

    As Robert Maxwell says in FRBR, A Guide for the Perplexed: “Given that FRBR emphasizes the fluidity of the concept behind the entity work, it is somewhat surprising that the document immediately gets down to the business of defining exactly where that line or boundary is (FRBR 3.2.1, 16–17)” (Maxwell 2008).” (p. 130)

    This same could be said for the LRM – it declares that the model is flexible, but it gets right down to defining specific data elements for each some of the entities, thus locking in the meaning of the entities as the data elements define them. Users of the model can extend it by adding detail, but they cannot change what is there. This is especially rigid because the entities are declared to be “disjoint”, which, at least in an RDF-aware world, means that no two entities can be described with the same attributes. (Note, that depends on some design considerations, but it’s not at all a stretch to reach this conclusion.)

    The inclusion of “attributes” (which are expressed as data elements) makes this less of a conceptual model and more of a data model. At the same time, the actual definition of the entities is quite vague, as it is in FRBR itself. It would have made sense to put more energy into creating good, strong definitions of the entities before jumping immediately to pinning them down with data elements. Much of the discussion around FRBR is about confusion around the differences between work and expression, with some people feeling that expression isn’t needed at all, and others that expression is exactly what users are seeking (cf. Taniguchi). An exploration of the meaning of the entities and their CONCEPTUAL attributes would probably be more informative than a list of data elements.

    Like

    • I just want to second the comment about the downsides of lack of flexibility for associating attributes with different WEMI entities. Although the extensibility of RDA is a wonderful thing, many of my frustrations with RDA are rooted in differences in world view and therefore not irresolvable in the current paradigm. For example, RDA has associated a number of roles (costume designer, production designer, etc.) with the FRBR expression that seem to me better associated with the work. FRBR-LRM isn’t that specific, but the disjoint entities appear to mean that the general problem will remain. It isn’t completely clear to me that there wouldn’t still be problems with making data interoperable where one implementation has associated X with the work and another has associated X with the expression. But I would much rather live in a world where we don’t all have to slice and dice the data the same way (unless, of course, everyone agrees to do it my way)

      Like

      • One more thing on this. Stephen Hearn submitted a comment to the CC:DA task force responding to FRBR-LRM, which I don’t think was included in the final report, but which I think is very relevant here and which I will try to summarize (I hope he doesn’t mind). He pointed out that it might be better to consider both adaptations of a work into a different literary form as well as translations of a work as creating a new expression. FRBR-LRM currently considers a change in form to always create a new work. Should a faithful prose translation of the Iliad necessarily be considered a new work because of the change in form? Treating a change in form as an expression-level change would provide more flexibility by allowing a cataloger to treat such an expression as either a translated expression or as an adapted work with a new creator. If the model were less rigid, there would be more room for experimentation and different interpretations for different needs.

        Like

  3. I would love to hear an explanation of the difference between the attribute designation of work (just category) and the attribute designation of expression which has a few particular details for music and maps. Also, there is an emphasis here on audience which hasn’t been emphasized in previous cataloging models. (I don’t know where to ask questions of the LRM group, so I’m posting this here as a place holder.)

    Like

  4. Another question about this area. Again, work has only the attribute category but expression has specific attributes relating to music and maps. Yet, section 2.1 Scope and Objectives (p.5) says “… data elements that are viewed as specialized or specific to certain types of resources, are deliberately not represented in the model.” This seems to be a contradiction, and an explanation about this would be welcome.

    Like

    • Page 60 has the Fig. 5.6 Overview of relationships. Here you see the only relationship of Timespan (except the recursive one) is “is associated with” RES. Same applies to Place by the way. Of course this will not do. These association relationships have to be classified: for instance Timespan – [date of publication] – RES. I am not sure if this is taken care of elsewhere in the proposed model.

      Like

      • As I recall, either from this reading or from my correspondence with Maya Zumer, there is an assumption that relationships on the class Res are inherited by the subclasses of Res. This is something I’m not entirely sure about since I believe it depends on implementation technologies.

        Like

  5. The way time-span is defined, it sounds like it is always to be expressed as a span, with a start and an end date. The only attributes are “Beginning” and “Ending.” It is intended to be used for single dates as well as ranges: “Even a very precise time-span has a measurable duration, however brief it may be.” So a year date would be “beginning=2016, ending=2016”. (?)

    Like

  6. Language: an attribute of a number of entities: Nomen, Agent, Expression. That is a design choice. Personally I would make Language an entity. The language of a Nomen for instance can then be defined by a relationship. Also other things have a language. Not on the conceptual level maybe, but definitely on the physical/implementation level. This refers to textual display labels of attributes (which is more or less equivalent to the Nomen entity, but not only for other entities).

    Like

  7. This is a comment about 5.6. It is part 6 of my personal response to the FRBR review group.

    6. LRM on Serials

    I am troubled by the discussion of serials (page 67). I wasn’t aware that the FRBR treatment of serials (specifically the Wall Street Journal) is now “recognized to have been an error”. If this is so the authors of the statement need to demonstrate it. But even if so it does not follow that any serial work can be said to have only one expression and only one manifestation. This seems absurd. A microfilm copy of a serial is a different manifestation from the original print copy, but it’s the same expression and it’s certainly not another work. The same holds for a digital copy of an original print copy. But even in the print world, 18th century serials were often reprinted later on by another publisher. This isn’t a whole new work, it’s a manifestation of the original serial work, and might or might not be another expression–it probably is another expression since resetting the type inevitably brings in at least minor changes to the text. A translation of a complete serial from one language to another–which does happen–is not two works, it’s two expressions of one work. So I don’t understand how the document can say that any serial work can only have one expression and only one manifestation.

    Like

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s