BIBFRAME Life after MARC?

Posting to Bibframe

On 8/13/2016 01:01, Karen Coyle wrote:

One of the key questions, for which we do not currently have an answer, is: What are the goals of the catalog? Cutter had his, in 1876

The latest version of the International Cataloging Principles has these goals

I think these ICP principles are pretty bizarre

I completely agree, and I think it is important for librarians and catalogers to accept that these are pretty bizarre goals. At least with Charles Cutter, he made it very clear that the goals of the catalog come from the questions asked by the users and he delineated those questions. See at http://tinyurl.com/3x4xpom where he provides the questions (asked in 1876):

1st. Has the library such a book by a certain author? Have you Bell on the Brain ? Have you John Brent, by Theodore Winthrop ?
2d. What books by a certain author has it ?’ What other books by Winthrop have you?
3d. Has it a book with a given title? Have you John Brent ?
4th. Has it a certain book on a given subject ? Have you a pamphlet on the bull-frog, by Professor -I’ve forgotten his name?
5th. What books has it on a given subject? Have you anything on glaciers? What have you on philosophy? I wish to see all the books.
6th. What books has it in a certain class of literature? What plays have you ? What poems ?
7th. What books have you in certain languages? What French books have you? How well provided are you with German literature?

Cutter’s catalog was created to answer most of these types of questions asked by people in 1876. While such questions are still asked today, the range of questions has become much broader–and I suspect the questions asked by the public were already much broader than those provided by Cutter in his day as well. But Cutter had to start somewhere.

An additional purpose of the catalog, and one that I think everyone can agree on, is that it serves as an inventory of (most of) the materials in the library. Beyond an inventory tool, in the early 21st century “search” is changing rapidly (Google changes its algorithm about twice a day) and people’s expectations are changing as well.

The ICP rules are just modern reinterpretations of Cutter’s purposes. It seems to me that if we want to build something for the public, then we should find out what kinds of questions they are asking and analyze them, just as Cutter did. The people who have the best knowledge of the public’s questions are reference librarians–reference librarians of all kinds: academic, public, children’s and so on.

People who are building the tools normally want to find out what the users of their tools want to do and what they expect, especially when there are tons of alternatives available that people find more attractive. (An interesting article in this regard is “The Strange Affliction of ‘Library Anxiety’ and What Librarians Do to Help” http://www.atlasobscura.com/articles/the-strange-affliction-of-library-anxiety-and-what-librarians-do-to-help, which I have certainly seen myself)

If we insist on building tools without knowing what modern society needs or wants (or we insist that we already know better) then it shouldn’t be surprising that anything we come up with will be bizarre.

-115

Share

2 Comments

  1. Ronald Murray said:

    Actually the ICP document initiates what professional data modelers want librarians to do: describe the things of interest to them in their own language. The renowned data modeler David Hay references this as the model of the business (Owner’s View – http://tdan.com/the-zachman-framework-an-introduction/4140).

    The ICP document is about two documents away from the “programmer’s to-do list” hinted at in the e-mail thread.

    Now work your way down the Zachman Framework numbered list in Hay’s article. Recognize yourself?

    What a professional data modeler* would do given the ICP document would be to elicit a much larger document that still represented the Owner’s View (which presently expresses many important-to-the-enterprise linguistic, philosophical, and sociological concepts). This version would be more articulate with respect to the things of interest and the relationships between them. This is simply an improved version of step two of the Zachman model.

    Only in step 3 then does IT-centric thinking come into play in identifying which things of interest can be represented as UML, E-R, Object Oriented, etc. data models. Even at this level we are not nailed to implementation-specific technologies. Much discussion on the thread (maybe) starts at this level rather than the levels above it. It’s no surprise that previous commenters are aghast at the non-executability of the ICP document.

    It is by combining levels three and four that 1960’s state-of-the art IT thinking yielded that amalgam of theory and implementation called the MARC record. (https://archive.org/stream/lrtsv12no3/lrtsv12no3_djvu.txt – p. 246). The same strategy deployed today is yielding BIBFRAME. Who’s going to be mad at BIBFRAME folks in the future for baking in XML/RDF thinking (I’m looking at you, triples) into resource descriptions that need to be incorporated into neural network systems?
    Systematically fleshing out the ICP into the full set of things of interest and their relationships (yes, doing the linguistics, philosophy, etc.) that it implies is the smart next step. We’ve not done this before – perhaps it will make a difference in the fitness for purpose of the computational deliverable.

    If nothing else, library coding thread commenters and the like would benefit from an intermediary who could translate between the two cultures.
    ——
    * See:
    https://www.amazon.com/David-C.-Hay/e/B000APF9Y2/
    https://www.amazon.com/Steve-Hoberman/e/B001ITPLVU/

    August 16, 2016
  2. Rachel said:

    I agree completely with the last paragraph, thogh I wold also add “unusable” to “biarre”. As a former Computer Science major, I know that a lot of programmers have the attitude that they know better than anyone what technology to introduce and how to make it and anybody who can’t use or doesn’t want what they make are idiots. It’s one thing I really wasn’t comfortable with as a Computer Science major, and one reason why I don’t think BIBFRAME has much chance to succeed. If BIBFRAME or some other Linked Data based catalog is going to be successful, it has to be programmed by programmers who are willing to find out from their users (and I include both patrons and catalogers in that) what works for them. Programmers who know about User Interface design know to do this. If BIBFRAME has any hope at all, it needs to have programmers who know about User Interface design.

    August 16, 2016

Comments are closed.