Wednesday, May 07, 2008

LC statement on RDA

I've long been on the fence with regards to the development of RDA - is it a transformative event or total folly? I think I've finally come to the opinion that RDA is overall a positive thing, and that it represents a necessary (although of course not perfect) step forward in the ongoing evolution of libraries.

What got me thinking about these issues again was a recent letter from Deanna Marcum at LC explaining why LC was issuing a joint statement with the National Library of Medicine and the National Agricultural Library outlining a testing and decision-making plan for determining whether or not to fully implement RDA. The letter and statement essentially say that wide participation in RDA development is a Good Thing (tm), yet so is substantive evaluation of it. Not much to argue with there. (Well, we always do find something to argue about, don't we?)

The stated goals of RDA, as well as its scope and underlying principles, speak to me strongly. I like the idea of a content standard written with FRBR principles in mind. The goal of making library description interoperate better in the current information environment outside of libraries is of course a laudable one. In this way, just by clearly stating these and a handful of others as the rationale behind the work being done, we've made a significant step forward. We're responding to the world as it exists around us today.

The world is changing, though. The environment today won't be the environment tomorrow. There's no indication, and perhaps even no real hope, that what we decide today will be right in a year, three years, ten. That's a reality we have to face, and I've decided I'm in the camp that says we have to move forward anyways, analyzing the risk but not being afraid of it. Looking at RDA through this lens, will it meet the goals it has outlined? Probably not. I see much in the current drafts that don't demonstrate the overall goals well. But we've never done this before, at least not in this way. We're learning. We're going to make mistakes. The stakes are admittedly high, but they're also high if we don't act. RDA has already evolved from community input, and I suspect it will continue to do so. Maybe it doesn't even stick around that long - maybe we learn enough from writing and trying to implement it that another round is warranted with some key needed improvements. We've investing many resources in this, but that's part of life as well. Many things don't pan out, and that's certainly not unique to the library world. I realize our resources are scarce, but they're going to be zero soon if we don't think creatively. I think RDA is an attempt to do that.

I'm still concerned that RDA as a content standard is stepping too far in the direction of a structure standard for my taste. It's explicitly defining "elements" whereas for content standards I like to think of "classes of elements" to help us remember that instructions in a content standard aren't necessarily a 1:1 match with fields in a data record - this is what enables us to mix and match content and structure standards as we see fit. But I'm the first to admit that the distinction between a structure and a content standard is an artificial one, and that any given standard can blur the line a bit. My concern still lingers, however - the RDA Scope & Structure document uses "elements" and "properties" interchangeably, but I believe these terms, even in the context given here, have very different connotations. We'll see, I suppose, whether my concerns are valid. Maybe I'm just being pedantic about terminology. Or maybe there's a fundamental conceptual problem here. I'm a pragmatist - I realize the only way we're going to find out is to try it.

7 comments:

Irvin Flack said...

I want RDA to work but I've decided I'm going to wait for the full final draft before I try to read any more of it. I become too frustrated and confused. I can't afford to lose any more hair! I find myself wondering: why on Earth did they write that rule that way?

My main problem is that RDA up to now lacks a single, clear, consistent conceptual model. Instead, it has multiple personality disorder: the International Cat. principles, AACR2, FRBR, DCAM, indecs, ONIX, etc. all in the mix and getting in each others way.

The showstopper for me is the term 'resource'. How does it relate to the FRBR entities in the RDA universe? Is it 'manifestation', 'manifestation/item', any of the four, all four? Call me pedantic, but I just can't read the drafts until I know the answer. The latest draft of the Int Cat Principles sticks its neck out and says 'manifestation/item'. Yes!

And 'access'? 'Access' presumes access to something, but what? 'The bibliographic and authority records!', we cataloguers cry. But RDA doesn't want to talk about records and so 'access point' is left out there, embarrassed and looking silly.

Jonathan Rochkind said...

I agree in general, but I also agree with Irvin that I find RDA rather unreadable at present.

I'm curious though, can you expand on your comment that properties and data elements are different things? This is something that I don't understand either (to me they are the same thing), but I'm interested in understanding your alternate perspective here. Any readings to reccommend?

Jenn Riley said...

That's a great point about RDA being a bit all over the place with regards to its conceptual model. As FRBR is a specific interest of mine, I tend to pay attention to that one and mostly ignore the rest. But of course that's not a very good approach and you've raised an interesting issue, which makes me wonder how many of those things do you need to understand in order to understand RDA? I'm still struggling to some degree with the terminology in the DCAM - can I really "get" RDA until I sort that out? I dunno.

Jenn Riley said...

Jonathan,

I think the way RDA is defining element and property (the latter from the DCAM as I recall) they are the same thing. But in metadata conversations for years we've been talking about metadata "element sets" - where elements mean fields in a data record. The conventional wisdom on content standards (if there is such a thing) says that they don't define elements, they tell you how to structure a *type* of element whenever it appears, e.g., titles or names look like this whenever you need to record them, regardless of what the specific field/element is. I think we're just facing some growing pains here, and terminology is getting in our way a bit.

Jenn

Irvin Flack said...

This is how I'd put it, using the terms of the DC Singapore Framework [1]: RDA has moved from just specifying 'usage guidelines' (ie a content standard) to also specifying a 'metadata vocabulary' (ie an element set) and a 'description set profile' (eg specifying core elements).

Is this a good thing or not? I think the JSC was almost forced into this because a content standard on its own is not much use to a cataloguer when s/he sits down to catalogue. If cataloguing is going to be truly made anew, it needs more than just a new content standard, it needs a rigorous domain (ie conceptual) model (FRBR is filling this role), a new metadata vocabulary/element set and a new syntax to eventually replace MARC.

However, I don't think this was the original intention: at first it was going to just be a updated content standard, a new AACR. With every new draft, however, there are two trends I can see: i) an increasing reliance on FRBR as the domain model and ii) a widening of RDA's scope to include the development of a metadata vocabulary. This vocabulary is important because it forms the basis of the DC/RDA work, which will produce a machine-readable syntax to succeed MARC.

That is why the success of RDA is critical (IMHO) to the future of cataloguing. The stakes have quietly and steadily been raised with every new RDA draft.

[1] http://dublincore.org/architecturewiki/SingaporeFramework/

- Irvin

Laura said...

The difficulties in getting old-school catalogers to grasp the concept of separating content standard from display standard was one of the many frustrating things about discussing RDA at CC:DA. I also seem to recall some sort of elements/properties discussion when the Scope statement came out.

I think the folks at CC:DA do recognize the problems inherent in RDA as currently drafted and keep calling on the JSC to tweak. You're right in that we have no choice but to accept RDA as a transitional thing as we learn to sort it all out.

I have reached RDA burnout though. I've resigned from CC:DA due to the volume of work and frustration level with the slowness of change. I've got other life priorities at the moment.

Anonymous said...

Laura said...
The difficulties in getting old-school catalogers to grasp the concept of separating content standard from display standard was one of the many frustrating things about discussing RDA at CC:DA.

------

How are you referring to "old -school catalogers" above ?

:-[ ??????

never heard that term in quite a while... just wondering
and don't usually even read this blog.

kinda strange to read. -?