Posting to RDA-L
On 11/12/2013 15.44, Brenndorfer, Thomas wrote:
Why are you repeating your egregious misinterpretation of FRBR? FRBR is not Find-Identify-Select-Obtain by author, title, subject headings.
If I “select” a book because it won the Booker Prize then I am engaging in one of the FRBR user tasks. The user tasks apply to the entire spectrum of attributes and relationships– not just the ones found and implemented in traditional catalogs through the heading structure. In addition, the entity-relationship model can be applied to many other data systems used in libraries. For example, the circulation module in my system is a feature-rich implementation of entity-relationship principles that if anything showcases how much the potential of the data locked in AACR2-MARC records lies untapped.
It is so remarkable that these basic facts are glaringly absent in your posts, and yet not surprising since many of your cited sources are to your own blog posts and podcasts.
For example, in emphasizing the power of free text searching in your blog post, why would not that same power be brought to bear on the topic at hand– retrospective conversion?
If the technology is so “incredible” then how is it that some problems are then so insurmountable that we should give up?
From you blog post — “it is the reliance on alphabetical order that has become obsolete in our new environment” indicates that your interpretation of FRBR is incorrect as FRBR and RDA are not just repeating the alphabetical structure of traditional catalogs. Notably in RDA, the instructions for authorized access points (the equivalent to headings) are relegated to second-class citizen status by being put at the back of chapters, after discrete data elements are covered (with the expectation that these discrete data elements, along with new forms of controlled identifiers which are always emphasized first in RDA, will become the primary operative pieces in catalogs in the future).
I have been on ruder lists, but this one of the rudest lists I have ever been on. Once again, there is a correlation with “understanding” and “agreement”: if someone disagrees, there must be something wrong with that person’s understanding. Because, if the person understands, he or she must see the light and be in agreement. Of course, that is a very modern interpretation. Or it could be argued that it is actually a very old attitude, harking back to the medieval church.
Still, when questioning an unproven system among a group of true believers, I realize it can become quite difficult.
Yes, I quote myself, but otherwise I would keep writing many of the same things. I link to what already exists. After all, that is the essence of what linked data is all about and that is supposed to be the salvation of us all. Still, it is true that I also quote lots of others. For instance, I quoted Amanda Cossham’s excellent paper where she in turn quoted many others who all have major problems with FRBR and have offered dozens of alternative models. Therefore, it seems that I am far from alone in questioning the utility of FRBR.
So now, it seems that FRBR has nothing to do with the user tasks? That’s news. Makes it hard to justify those entities and attributes and relationships since that is the very first step in building an entity-relationship model. But people have been backing away from those user tasks for awhile now as it becomes more and more obvious that people prefer other methods such as relevance ranking and algorithmic connections that work in completely different ways. The public is moving on at a terrifying rate.
And as for the relationships, there remains that uncomfortable fact that if they are to be implemented, then quite literally millions of records will have to be updated by cataloging staffs that are decreasing in numbers; staffs who are already overworked and in many cases with morale not doing all that well. Plus, money does not seem to be pouring in for any of this. And do it all without the slightest evidence that adding those relationships will bring anybody back to our catalogs. Why shouldn’t people question? It is logical to assume that adding the relationships may have as much of an impact on the public as did those hundreds of thousands of updates to the authorized headings that occurred recently, where the cataloging abbreviations were spelled out.
We know what kind of impact that had: that was a “shot not heard around the world”.
But I guess there is the assumption that, as you say, “… in emphasizing the power of free text searching in your blog post, why would not that same power be brought to bear on the topic at hand–retrospective conversion?” Technology will come in as a deus ex machina, and save RDA and FRBR. And not in 15 or 20 years. True, it may, but it is just as probable that after all those relationships are created, it will still not make a difference to the public. After all, there is no evidence for it one way or the other.
But few seem to care about any of that.
I just wish that we could declare victory for FRBR now because modern computing has allowed people to do them right now, as I have shown often enough with using the facets in Worldcat. I don’t have to demonstrate it again because I have tired of it, and besides, anybody can do it on their own. Declare victory and then we can start moving ahead. For instance, we could then experiment with the identifiers and see where they may be strong and where they are weak. Maybe it won’t make much of a difference (as I have already discussed–in a podcast). And maybe we could get the catalogs to start working half-way correctly again?