Thomas Brenndorfer wrote:
I'm not sure about this fixation on the "unit record." In Scenario 2, a multitude of bibliographic and authority records work together and are linked together. Likewise, in the card catalog paradigm (Scenario 3), separate records are used to create a linked infrastructure using controlled headings. There is a single MARC record now that concatenates information about different entities, but it's not a self-contained unit record that stores all important information about the entities involved. That being said, the idea of broad container records is still a priority, as seen in the work being done on the Linked Data version of the cataloging universe (and the game in town for Linked Data is RDA, not AACR2). This Q&A page (particularly question 12) for Linked Data applications of bibliographic information is quite useful in understanding how bibliographic entities are handled in this new context. http://www.niso.org/news/events/2010/dublincore/questions/
I don't think I have a fixation on the unit record. What I have been trying to point out is that FRBR does not propose anything that catalogs do not provide now: the user tasks are *precisely* what anybody can do in a catalog right now, and what people have been able to do since at least Panizzi's time in the 1840s. The only thing that is "new" is breaking our current records up into works/expressions/manifestations/items. And I tried to show that this is also not new at all, that it harkens back to the days of the printed catalog. Then I asked what seemed to be a highly obvious question: is trying to recreate 19th century methods what people want today? Is this moving forward or backward? The only other kind of display involving FRBR-type structures I have found was in Fiction Finder, which was very interactive. In my opinion, such a tool would be essentially useless for patrons.
Concerning the point about linked data, I agree that it is important, but there are lots of ways to do it. Linked data is nothing new on the web--webmasters have always had to deal with "linked data" every day when they make a page. Any webpage you look at today probably contains lots of different parts of linked data: separate files for headers and footers, for navigation, styles, and so on. There are all kinds of different ways of implementing it: through server-side includes, linking cascading style sheets, images, various scripts, web services, and so on.
For example, what appears to be a single "webpage" in this record from my catalog: http://www.galileo.aur.it/cgi-bin/koha/opac-detail.pl?bib=26319 is actually made up of--I don't know how many different files, dozens at least--linked together using all of the methods delineated above. The Google Translate, Google Books, Sharing sections, are all called "widgets" that links this record to all kinds of other resources elsewhere on the web. If you click on "Get a citation" you are working with the WorldCat API, which they created, but I had to format what I received a little bit, to give my users a citation automatically. My patrons like that one a lot.
To institute linked data *right now* in our records, we could do it now and don't need FRBR or RDA. The problem is: we would have to link into something now that people would find really useful, and we don't really have anything. There is the id.loc.gov service, which could be used to search, e.g. in this case "Television series" (since "Television series Great Britain" retrieves nothing) and the patron could find the additional heading: "Television mini-series", and then--you're at a dead end. It's a good beginning but currently, there is nothing for users to do once they get there.
This is why I have linked into Bernhard Eversberg's system "LCSH Browser" using the blue "S" next to each subject. When you click on that, you search his system where, if you clicked on "Television viewers" you will find the BT, NT, UF etc. for that topic, plus links into different things, including even my own catalog (AUR Library). If Bernhard would turn this into a web service, you wouldn't have to click and it would come up automatically, and I could format it however I would want (using an onmouseover event or something).
We could try searching subj3ct.com, the search for "Television series Great Britain" https://subj3ct.com/search?query=Television+series+great+britain retrieves something that could prove useful, e.g. we see the children's series "Freewheelers" (http://dbpedia.org/page/Freewheelers) which provides lots of additional information and links to all kinds of other resources, such as its Wikipedia page, photos, related subjects, and so on. But at basis, I personally find the current structure of subj3ct.com confusing and unpredictable. Still--it's something!
Unit record or not, I honestly couldn't care less. We need to build a tool that *our patrons will use* and especially in this economic climate, without risking both the family farm and the baby's shoes. So much could be done right now and I think there are lots of people just aching for some ways forward.
I'll say it one more time: FRBR looks backward, while nothing I have seen in RDA changes anything of substance from what we do now. What we need is real change: something that will make people sit up and take notice. Real change can happen, if we decide to do it.