On 27/11/2012 20:22, Kevin M Randall wrote:
You seem to be making the assumption that the existing metadata in “the older records” are not going to be augmented in any way. There will certainly be mechanisms for creating and maintaining links between, e.g., records in WorldCat and and data in other databases such as IMDb. Most of this will be done by algorithm, not by catalogers keying in links record-by-record. This won’t by any means be “complete” upon the Day One implementation of any system by any library. It will be an ongoing process. But our catalogs have *never* been complete. They have always been growing and evolving. If we don’t take any steps, we won’t get anywhere.
Certainly RDA/FRBR have never discussed the retrospective conversion of every single record in our catalogs–what that actually means to library staff and to the public, and how much it will cost. For films, we may be lucky: the IMDB exists but is only one database–and only a theoretical possibility at that. Where are the databases for editors and all of the other relations? Plus the relationships among the WEMI: the parodies and supplements and so on. Nobody has talked about such matters because it would be a part of the business case that RDA/FRBR has never wanted any part of. The complexities and costs will be phenomenal and it will take some time to edit every, single record.
Catalogers must think about that for a moment: every, single record.
In 90+% of the cases, it must be done manually, probably by higher level staff, unless somebody can demonstrate that these things can be done through automated means. I have worked with these kinds of algorithms, and have had some successes, but I will state plainly that the very idea of doing it for relations is incredible. Creating automatic means to spell out cataloging abbreviations would be child’s play compared to this, and that has already been considered to be too much. Somebody should at least demonstrate there there is a realistic possibility before assuming that a “black box” can do it magically.
The alternative is to decide who will add the relationships: higher-level staff, lower-level (lower-paid) staff or perhaps even through some form of crowdsourcing, which I think we can all agree would lead to total disaster. The FRBR relations are really complex. Also, it is necessary to show that such a huge project can be completed in some reasonable amount of time–let’s say, two years? Five years? Ten years would leave us ‘way too far behind what the technology will be then. If it’s 30 or 40 years, we can forget it.
And apologies, our catalogs have been complete, that is, as complete as any standards can offer: they are complete within certain, definable limits and if people are following the standards. No amount of explaining will deter people from being angry after getting obviously incorrect results for searches on “film directors” or “editors”. So, we consider that with the relationships, how long will it be before a half-way decent result can be possible? Before that can be answered, the term “half-way decent” needs to be defined. 40%? 50%? 80%? Who should decide? What would a researcher reply? Of course, even 40% of the millions of records we have would demand a long time and enormous resources.
The public will simply not believe that decreased access is part of “evolving” and “getting anywhere”. Such thinking is just too Orwellian for my taste and too reminiscent of “War is Peace, Freedom is Slavery, Ignorance is Strength”. I love language and I believe it should mean something. Evolution means adapting to the changes in the environment and FRBR/RDA are based on dogma declared quite some time in the past. (Whether in the 19th-century or in the 1990s is an academic matter at this point) Decreased access and higher costs are not “getting anywhere”, but rather devolution and going around in circles–if not heading backwards. Certainly there is no concern for the public or the catalogers in any of this.
I realize that many who read this will conclude that I just don’t understand and I am retrograde in my attitudes. I reply that considering the records we have now as “legacy data” is an IT form of thinking, especially IT in the business community. For business IT, the information from 10 or 15 years ago is pretty much obsolete and that may well be true for them. It can safely be archived off-line or stored in a zip file and placed on the web if somebody wants it badly enough.
Libraries are completely different. Do we really want to say that the books and other materials acquired in the 1940s or 1950s are less important than those we acquire today? Or the ones bought in the late 19th century? This is what it means to break the consistency with the records created back then–it would essentially wall them off from being found by the public. I tried to show this in my last podcast http://blog.jweinheimer.net/2012/09/cataloging-matters-no-16-catalogs-consistency-and-the-future.html
There will be many more costs to RDA and FRBR–that is one thing I believe we can all agree on. The question is: will they be worth it?