ALA TechSource Logo
 
curve Home spacer Publications spacer Subscribe spacer Blog spacer About  
    

Continuing the Conversation: RDA Vocabularies in the Semantic Web

Submitted by Daniel A. Freeman on November 17, 2010 - 5:52pm

Earlier today, we held the second session of our three-part workshop, Using RDA: Moving into the Metadata Future. The session, RDA Vocabularies in the Semantic Web with Diane Hillmann, completed the successful series. The following are resources and questions for discussion or explanation. Diane Hillmann will be chiming in via comments--please join her.

The following resources or examples were referenced during the presentation.

Property and value vocabularies are registered in the Open Metadata Registry (formerly the NSDL Registry)

For an example of  element in the Open Metadata Registry, see  RDA Group 1 Elements: Book Format (Manifestation) http://RDVocab.info/Elements/bookFormatManifestation

D-Lib Magazine. January/February 2010 RDA Vocabularies: Process, Outcome, Use

DCMI/RDA Task Group Wiki

Below are a few question that Diane will address in comments.

You said we would be free of the tyranny of records.  It seems that the Semantic Web is pulling info from any number of sources--how does it decide which sources to use and reduce the sources to a manageable level?

How do you envision URIs being assigned to entities? And do you see them being assigned to works or manifestations, for example?

There is a lot of anxiety around whether frontline catalogers need to understand these abstract concepts. Knowing that the ultimate goal is that we’ll be working in some kind of cataloging tool, what are the most important concepts for catalogers to understand to participate in the process now? For example, what do we need to know when we’re talking with vendors?

Does anyone know of any vendors actually working right now to build any applications using RDA as expressed in RDF?

Are these vocabularies describing fields, like "tag" names, which can be used to carry content and related to each other as well as to to specific items being described?  Please give a little more context for the vocabularies and their use.

Could a diagram or chart or something possibly be drawn illustrating the relationships between RDA, RDF, XML, FRBR, FRAD,.  It would be helpful to visualize these new concepts and how they relate to each other.

How is Dublin Core related to RDA/FRBR? Is DCMI doing its own DC-RDA application profile?



Comments (10)

How is Dublin Core related to

How is Dublin Core related to RDA/FRBR? Is DCMI doing its own DC-RDA application profile?

If you’re talking about the Dublin Core Metadata Initiative, the relationship is as I described it in the Webinar. The short version was that the RDA development effort was struggling a bit, not quite sure whether it was looking forward or backward, and because of the many people connected with both those efforts, ALA Publishing decided to get both groups together, at which time the DCMI group asked a simple question: “How can we help?”

DCMI is primarily a volunteer organization, and it’s interests can be best expressed as being more about enabling communities to expose their data using web standards, than about co-opting efforts of those communities. The library world can benefit from the work that the DCMI Community has done with Application Profiles but we shouldn’t expect DCMI to do the RDA AP work itself.

Could a diagram or chart or

Could a diagram or chart or something possibly be drawn illustrating the relationships between RDA, RDF, XML, FRBR, FRAD,.  It would be helpful to visualize these new concepts and how they relate to each other.

I think we’d like to do such a picture, but I think we’re not quite there yet. I think it’s clear that we’ve got a way to go before we get really good at explaining this stuff, but we are trying hard!

Are these vocabularies

Are these vocabularies describing fields, like "tag" names, which can be used to carry content and related to each other as well as to to specific items being described?  Please give a little more context for the vocabularies and their use.

Yes, at a basic level this is true. And if you’re used to working with MARC, you’ll know that the MARC documentation includes definitions of all those fields and subfields, as well as textual descriptions of the relationships with other field tags and subfields. The difference is that while MARC documentation is entirely oriented towards humans, we are moving towards a more machine-actionable world (instead of just a machine-readable one). To do that, we need to build vocabularies in a way that many more of our routine data management activities can be accomplished by machines, and not just by humans. The way we’ve built the RDA Vocabularies allows machines to work with the relationships we define, do more with those semantics, and not just display text that a cataloger has typed into a form.

Does anyone know of any

Does anyone know of any vendors actually working right now to build any applications using RDA as expressed in RDF?

I know some are, but others are waiting in the wings for the path ahead to be come clearer, and not incidentally for their customers to weigh in. What that means is that asking me is not anywhere near as helpful as asking the vendors. Please! If you go to ALA or regional/state meetings where the vendors exhibit, go and talk to them, ask them what they’re planning. Ask them all, even if you don’t think your institution is interested in a particular vendor product. If you work in an institution using a particular product, and you or someone in your institution attends user group meetings, make sure they go armed with questions.

There is a lot of anxiety

There is a lot of anxiety around whether frontline catalogers need to understand these abstract concepts. Knowing that the ultimate goal is that we’ll be working in some kind of cataloging tool, what are the most important concepts for catalogers to understand to participate in the process now? For example, what do we need to know when we’re talking with vendors?

This is something I’ve been talking about for a long time with catalogers (and other librarians). I’m not sure that the ultimate ‘goal’ is to be working in some kind of cataloging tool, but rather it is likely to be the ultimate reality, and it may not be the catalogers of today doing that, but perhaps folks at a different level, most likely support staff. We need to recognize that the technology changes we’re talking about also exist in a financial environment for libraries where the squeeze we’ve been seeing for some time is likely to continue. It seems to me unrealistic to expect that the shifts in how cataloging is accomplished (and by who) will not accelerate as the technology changes kick in and investments are needed to accomplish to keep libraries viable.

What that means for today’s professional catalogers is that they should be looking at these changes, and the new learning that’s involved, as an opportunity to upgrade their skills and knowledge to ensure that they are seen as a critical piece of their institution’s planning for changes in the environment of data management. Insofar as I am no longer anyone’s boss, I can’t mandate that anyone actually do this, but I have spent many hours trying to help catalogers figure out how to retrain, and am likely to continue to do that for the next few years. And I freely admit to getting a bit irritated when I hear catalogers whining about this necessity, and about having to take personal responsibility for their own professional development. If your institution isn’t stepping up to the plate to help you with this, by all means organize yourselves to make it happen!

How do you envision URIs

How do you envision URIs being assigned to entities? And do you see them being assigned to works or manifestations, for example?

This is a really interesting (and important!) question and one a lot of people are asking and talking about. There may well be several layers of identification, some identifiers used to manage data within systems, and others used to identify data across systems. Our current systems, because of their ties to a central node, have tended to provide shortcuts for our identification needs, so that within them we tend to use record IDs or other identifiers (ISSNs or ISBNs, which are publisher IDs) to identify resources, instead of having to identify them specifically as ‘resources’ for our purposes. We’re not going to be able to do this in a more open, less silo-ed world, so it’s an important discussion to have, sooner rather than later.

But yes, I think we will have to figure out how to assign them to any data we’ll want to share, like data related to the FRBR WEMI entities.

Folks, I'm going to try and

Folks,

I'm going to try and answer the questions left over in comment 'chunks' and as is good practice for f2f environments, I'll repeat the question before I answer it!

You said we would be free of the tyranny of records.  It seems that the Semantic Web is pulling info from any number of sources--how does it decide which sources to use and reduce the sources to a manageable level?

‘IT’ (the Semantic Web) doesn’t decide anything for you—what sources of data you trust and want to use is entirely up to you. Statements should come with provenance—information that tells you something about who’s data it is, where it came from, and how old it is (among other things)—which should help you automate the process of selection and aggregation. As you can see, this is an entirely different way of managing data than we’re used to in LibraryLand, but it’s not rocket science, and it’s one of the capabilities you should be looking for when you talk to your vendors.

The archive is being

The archive is being processed, and you'll receive an e-mail tomorrow when it is ready.

This is perhaps a dumb

This is perhaps a dumb question, but how do I access the archived version when it is posted? I had to miss today's session.

This is probably a dumb

This is probably a dumb question that was already answered elsewhere, but I had to miss today's session -- how do I access the archived version when it is posted?