On 17/11/2012 03:07, Karen Coyle wrote:
Rather than beginning with our bibliographic metadata (or any other metadata, really) I think we should be looking at services first, and then deciding what metadata fulfills those services. What do people need? I don’t have data that would allow me to answer that question (and I would love to have such data!) but my own thinking is tending toward:
– they probably need navigation that is based on concepts rather than words — that allows them to go narrower and broader, or sideways. I recall studies when I was in library school that showed that people begin a search with a broader topic term than what actually fulfills their information need. In other words, the questions are less specific than the answer that turns out to be helpful. This makes a lot of sense, because the person asking the question probably knows less that what will be revealed as the answer. Keyword searching (aka “googling”) isn’t conceptual, and Google et al function more as yellow pages than as knowledge organization systems. Libraries do classify their works, but unfortunately we classify only to determine shelf order. I’d like to see experiments with classification as navigation.
– they need to get to stuff that isn’t open access on the web, but is available through a library. Thus: library holdings and licensed content.
– they need recommendations and guidance when exploring new topics. Wikipedia sometimes helps people get started, but I’m finding the bibliographies in WP pages to be somewhat random. I’d like to see the art of bibliography revived.
– they need all kinds of help digging through the mass of stuff that IS available, and I’d like to see a way to do “shared screen” reference, and for libraries to somehow find the means to give personalized help or even group help. The idea of “maker spaces” needn’t be limited to 3-D printing
– etc etc
I agree with all of this completely. But this does show that you believe there is a need for library bibliographic metadata to be on the web–at least the access points, not the descriptive portions. The description is almost always (I won’t say always) aimed at the librarians and collection managers, to ensure maximum efficiency in managing the collection. While xii, 356 p. is mostly unnecessary to a patron, it is vital information for the librarian whose job is to manage the collection. I wrote about precisely this in “Realities of Standards in the Twenty-First Century” http://eprints.rclis.org/handle/10760/15838#.UKod9oZnZMo (Part of “Conversations with Catalogers in the 21st Century”).
Catalogs do not have to be declared dead, but the dictionary part needs to be considered obsolete. Many parts of Cutter’s “Rules for a Dictionary Catalog” just don’t make sense anymore, i.e. the dictionary part. There are probably people alive today who have never looked up anything in alphabetical order. In fact, I wonder how long it will be before people cease to understand the “dictionary” part of what Cutter wrote about–that in his day, everybody looked up everything in alphabetical order and now, it is very rare. Online dictionaries are not searched alphabetically; Wikipedia isn’t. Of course, to experience it, all anybody will have to do is get a printed dictionary (or download a free one) and look something up but perhaps even all of that will become lost.
The “conceptual groupings” that catalogers make, e.g. the set of all resources with the subject “Poe, Edgar Allan, 1809-1849–Knowledge–Book industries and trade” are very useful, but the current methods to access it are obsolete. Teasing something similar out of full-text searching is difficult, and making that same access reliable is almost impossible to imagine using full text or relying on “crowdsourcing”.