Posting to Autocat
Hal Cain wrote, in a later part of his message:
But Google is free to make it up as it goes along. We, because of the weight of our past, are not: coherence and consistency are defeated by constant change. Change, if it is to be successful, has to be ushered in by reflection, invention and rigorous testing. I don’t believe that change of everything already recorded in our bibliographic databases is reasonable or economically possible — the markup applied in MARC fielding and subfielding are not sufficiently granular, nor is all the required data present, but change for the future along with considered change in some of our historic data may be. That supposes of course that cataloguers can agree what they want (in pursuit of serving the users they have already) which seems to me rather unlikely.
This I agree with completely, and is one of the main reasons why I do not believe that FRBR or RDA are the solutions we need. I still maintain that our records at basis, provide some very important functions that exist nowhere else, and that people want and need our records. Yet, in my opinion, it is not logical to equate this with FRBR/RDA without the requisite “reflection, invention and rigorous testing,” as you put it so well. I am not alone in this stance and it is the same as the Working Group concluded. Still, deciding that FRBR/RDA is the wrong way to go does not mean that we can’t move forward and introduce change in other ways: I think there are a lot of other possibilities for our records without having to change everything to RDA/FRBR, by instead concentrating on trying to repurpose our records in all kinds of innovative ways without changing the rules and methods we currently use. At least, we don’t change them yet.
This would be the equivalent of a company “repackaging” its products: changing the package and trying to sell them in new places, but not changing the actual contents until they have more information. For a business, this is the quickest and easiest way to try to expand the reach of their products while they gain time and information to figure out how to change their products in the future, if it turns out that they need to change their products at all.
So, while I believe we need change, and it may turn out that the contents of our records will need substantial change, it is still far too soon to know how they will need to be changed. But in the meantime, we can repackage our records and have them appear in new ways in new places to see how the public reacts. As we see what happens, we can start to change in more substantial ways. What can we do now? Let our records out for the public to actually work with, as Tim Berners-Lee has suggested with all data, and linking whatever can be linked now. This means linking the authority-type information. For example, names could go to WorldCat Identities, and/or VIAF, which could be very useful for people. Subjects need to be rethought anyway (as I mentioned before) but our records could be incorporated in all kinds of interesting cooperative tools such as dbpedia. Creating APIs with our systems and moving toward XML would make a good beginning. Somehow, I think Wikipedia must be included because that is where people tend to go today. Other people out there will have lots of other ideas.
All of these things will need to be done eventually anyway and all can be done automatically, without any changes from our current practices, much less the major changes to FRBR and RDA. Why not change now? The public could start using our records and interoperating with them in new ways right now, and we could begin to see what really needs to be changed in our rules and practices.