As to records, the notion *will* break down. Each property will have
greater value on its own, and we will create ontological proxies to
bind their context together. We're already seeing this debate in the
RDFa world, where the URI of the page the RDF statements are in, and
what significance we should apply to that. It seems, rather
annoyingly, that nothing can stand on its own two feet. :)
I can imagine this statement being rather alarming to many practicing catalogers. Everything we have done is based on the *catalog record,* and I know I would be concerned. Therefore I have tried writing something in a question and answer format that may address this. Perhaps others have done this already, but...
What are the practical implications of a loss, not only of the traditional unit record, but of the bibliographic record as a separate entity?
In other words, how and how much will my work change?
It does not have to change very much. There will be far less manual typing and copying and pasting for example. Catalogers will still be adding names, titles, subjects, publication information just as today, *except* if the name, title, publication information etc. exists already as a reference somewhere, the record will use a URI. Many cataloging modules have some of this functionality right now using internal relational databases, where there is one authority record for Shakespeare, and the bibliographic records just display the 100 field of that record. Therefore, you are not copying and pasting, and you do not have to type in over and over again
100 1\$aShakespeare, William,$d1564-1616
In the new world using URIs, you will be able to link to and import information from different places, e.g. dbpedia, e.g. http://dbpedia.org/page/William_Shakespeare. from a library authority file, or from elsewhere.
If a record doesn't appear in the chosen one, then another one could be used, if available, e.g. http://authorities.loc.gov/cgi-bin/Pwebrecon.cgi?DB=local&Search_Arg=weinheimer%20james&Search_Code=SHED_&CNT=20+records+per+page (this is not a URI)
If the information is nowhere to be found, you will make a new one in one of the systems (even if it's local, but it would be much more useful to share it) and input whatever is necessary there. Also, you will be able to update the information, make references and so on. Just like today. So what will change?
As I mentioned, very little needs to change in this regard. New initiatives can take place on other levels where people can link related data together, e.g. the VIAF http://viaf.org/viaf/76320393 for Montaigne, which brings together the forms of names used in different national library systems. But these could be mutually linked in all kinds of ways to other initiatives, e.g. dbpedia.
Let's examine the record for Montaigne at dbpedia http://dbpedia.org/resource/Michel_de_Montaigne.
This is not a single "record" either, but consists of many links. (By the way, this dbpedia page could look completely different if they wanted) Just for one example, look at his birthdate, which is complete. Now, I notice an error: "birthplace" actually contains the links for his birthdate. If you click on the year, you will go to another page for 1533. Although this record appears to be of little use, at the bottom, you can go to the wikipedia page for 1533. Perhaps useful. Perhaps very useful, or in any case, it opens up new avenues for exploration. No matter what, this is definitely much more useful to the public than our current authority records, which have completely different purposes in mind.
What will these things that replace bibliographic records look like?
For the public display, they can look however you want. The coding behind the scenes will be completely different, but they could display to the public almost exactly as they do today. The links will go not only into the local catalog, but also into the wider information world (i.e. what is *really* available to people). (This is what I have tried to achieve with the Extend Search function in my catalog using less sophisticated technology)
Add to this the user interactivity (Web2.0) from expert researchers to the general public, and the sky is quite literally the limit. This is how the semantic web could work, but it is still in its infancy, and as we can see, it can use some tender, loving care. I submit, this is a place where all types of librarians could easily find a great number of very important and influential roles to play. Wikipedia's place could also become extremely important.
So why don't we do it now?
Our information structures do not allow it. For one thing, we have MARC "records" that you can probably (I hope!) now see need to come apart in the new environment because they are too limiting. Another major impediment is the MARC format itself, which can't be used at all for the functions described. Still, this does not mean that anyone's day-to-day cataloging interface will need to change drastically, if at all. It could probably look and work exactly as it does today with the current fields and subfields, if you want, but I am sure lots of people could come up with better displays and functions. You will still be able to discuss the 100 $c if you want. You will just have more places to take information from. Behind the scenes though, it will change a lot. For example, you will not *really* be taking information from these other places, but *linking to* that information. And concerning what the users see, this should be more interesting and useful for them than ever before. Authority work can take on an additional layer of importance in this new world.
RDA and FRBR
RDA and FRBR attempt to restructure our data by breaking up the data into WEMI, and each part has specified attributes, e.g. attributes of a work:
title of the work
form of work
date of the work
other distinguishing characteristic
context for the work
medium of performance (musical work)
numeric designation (musical work)
key (musical work)
coordinates (cartographic work)
equinox (cartographic work)
Concerning works, in our current uniform title *records,* (that word should always make us take notice) this information does not exist per se, e.g. date and form of a work is normally (if ever?) there. The dates, added only when necessary to differentiate editions (130/240 $f), applies to expressions (i.e. date of publication or printing if necessary), but still, a lot of the information related to the attributes listed above resides in various places within the separate MARC bibliographic records. The same goes for all the rest of the attributes, some of which are quite different conceptually from what we have always done, e.g. the summary note comes under the FRBR definition of *expression*:
"4.3.9 Summarization of Content
A summarization of the content of an expression is an abstract, summary, synopsis, etc., or a list of chapter headings, songs, parts, etc. included in the expression."
My own opinion
I have tried my best to sum up the current situation as fairly as I can and with a minimum of errors, although please correct me if I am substantially wrong somewhere. (I'm sure you will!) The primary reasons I am against RDA is that it does not change substantially the rules for input (i.e. titles are still the same, publication information, description, even headings and single main entry!), but it changes the organization of the rules into this information Weltanschauung of WEMI, which is definitely based on the traditional library-view of the world as it was in the mid-1990s (in-turn stemming directly from the 19th century and before) and is dubious today. As a result, I submit that you would have a *lot of problems* convincing other semantic web agencies to accept such a structure. Some have argued that even if the model is dubious, it is still necessary that we adopt it because we need some kind of model in order to enter into the Semantic Web. I would agree if implementing cost us nothing, but there will be major costs involved, and therefore, I believe we should be d***ed sure we get it right, because it could be extremely detrimental to the entire library world if, after substantial costs at a very delicate time, it ends up either almost completely wrong, or of no practical benefit to the public.
I believe there are two main impediments to our entering this new world of the semantic web: 1) the problems in our obsolete library formats and 2) the fact that we have not wanted to share our data. We should be putting our efforts into sharing the data that we have right now, and adding linked data whereever we can, i.e. linking into the authority files, which still are not up on the web in a useful format in their entirety, and this leaves developers no choice except to link into non-library tools such as dbpedia. To the question of: should we put our materials on the web in a comprehensible format for development even though they may not be as fully linked as they could be?, my answer is: So what? At least our information will be available to the community for reuse, and they will recreate it for their own purposes anyway.
Throughout all of this, I do not see that the actual rules for determining titles (which ones to choose and how they should appear) will have to change; the rules for publication dates will not have to change; the rules for explanatory notes will not have to change and so on and so on. It is the underlying structures that absolutely must change and certain administrative decisions must change as well, to open the data. Finally, the catalogers along with other librarians must fully accept and understand the benefits and responsibilities of working in the "global cataloging world" (as Alex put it). So, if the rules do not have to change drastically, what is the reason for everyone to be retrained to use RDA?
And that's why I initiated the Cooperative Cataloging Rules Wiki, so that we can continue to use the current rules, and hopefully, to open up their development with other metadata providers. In this entire discussion, there has been little mention of bibliographic standards, their importance and use, and what they mean in the new world. I think they are important.