Dragging old data forward: finding yourself an RDA Helper
Download as PPTX, PDF6 likes20,239 views
The document discusses the RDA Helper, a proof of concept tool to help libraries migrate legacy MARC data to RDA. The RDA Helper can generate many new RDA fields by extracting data from existing MARC fields like 008, 260, 300. It aims to demonstrate that much of the migration process can be automated and help catalogers and vendors more easily move records to RDA. The RDA Helper focuses on fields like 336-338, 344-347, and 380-381 and processing issues like abbreviation expansion and generating GMDs from existing RDA data.
Dragging old data forward: finding yourself an RDA Helper
- 1. Dragging old data forward: finding yourself an RDA Helper Terry Reese, Gray Family Chair for Innovative Library Services Email: terry.reese@oregonstate.edu
- 2. Vehicle for Research -- MarcEdit • MarcEdit • http://people.oregonstate.edu/~reeset/marcedit 1 January 28, 2013
- 4. Common Questions I hear • What about the GMD? • We code all our data in RDA, how do we deal with other peoples? • What do we do with bulk data loads? Vendor data? • Do we care about Legacy Data? • My library has been encoding records with RDA fields for over a year and now they are incomplete. I have thousands – what can I do? • WHAT ABOUT THE GMD? 3 January 28, 2013
- 5. So what is the RDA Helper? • It’s a proof of concept to demonstrate that: 1. Most current RDA fields can be derived from existing data 2. Migration paths for legacy/bulk data can and should exist 3. Abbreviation expansion maybe isn’t as straightforward as we would like 4. GMD data can be automatically generating from existing RDA data 5. Vehicle for experimentation 4 January 28, 2013
- 6. Scope of the project • RDA helper has been limited to looking at practical implementation of RDA elements into MARC • Looking specifically at: • 336/337/338 field groups • 344/345/346/347 field groups • 380/381 field groups • Evaluating the 260 • Processing Abbreviation Expansion • GMD processing • Determine how easy 3rd-party development/engagement with the RDA standard/metadata community will be going forward. 5 January 28, 2013
- 8. Hitting a brick wall http://www.flickr.com/photos/camknows/8374910613/ 7 January 28, 2013
- 9. Mining the Data • Does the data already exist in MARC records? • Yes and no – while much of the data can be extrapolated, the generation of many new RDA specific fields requires evaluation of multiple data points. • The most important data points? • LDR/007/008 – with these three data points, you can generate most RDA specific field data. • GMD • 856 • 300 • 130 • 240 • 730 • 740 8 January 28, 2013
- 10. Mining the Data • Abbreviation Expansion is challenging • Real-world data is simply real-world crazy • Simple Example: =300 $a1 v. =300 $a1 vol. =300 $aOne v. =300 $a1 vols. =300 $aV. =300 $av. =300 $a12 v. 9 January 28, 2013
- 11. So how does this thing work? • RDA Helper • http://www.youtube.com/watch?v=cqLMPp9vZVM&feature=player_embedded 10 January 28, 2013
- 12. So why create something like this at all? • Admittedly, most of the promise behind RDA isn’t going to be found in these first baby steps in MARC, but… • To demonstrate that much of this initial work can be done automagically and that much of the data in our existing hybrid environments can be moved forward. • To provide a testable implementation for catalogers who are still uncomfortable with what these changes mean. • To support public libraries, many of which utilizing ILS systems that rely on data that that is going away like the GMD to create more user-friendly interfaces. • To support vendors that provide MARC records and offer a simplified path for moving their data forward. 11 January 28, 2013
- 13. Going forward http://www.flickr.com/photos/jannem/2079422115/sizes/z/in/photostream/ 12 January 28, 2013
- 14. Thank you Contact Information: Terry Reese Email: terry.reese@oregonstate.edu Work: 541.737.6384 Getting MarcEdit: http://people.oregonstate.edu/~reeset/marcedit 13 January 28, 2013
Editor's Notes
- #3: I’ve found over the past couple years giving workshops on metadata processing, that talking about RDA is like talking about Religion and Politics. It can really bring out the crazy.
- #4: I wish I was kidding about the GMD
- #5: Experimentation – treating specific fields as objects for purposes of validation.
- #6: RDA Helper was designed for practical usage. Now, there are a lot of concepts related to RDA that exist outside of MARC. The RDA Helper is definitely concerned with how these concepts are related into MARC.
- #8: OSU gives me a lot of indirect support when it comes to my work around MarcEdit. Because of that – I usually find that I spend close to 2-3k a year to access ISO standards documents. These are international standards documents and as a developer, I don’t like it, but I think of it as the cost of doing business. However, I was unprepared to have to do the same to access what should be an open library standard. The library community is going to have to deal with RDA in some form – but I do worry that this specification will be dead on arrival for communities outside the library if we insist on keeping it behind a paywall.
- #9: Is the data already there?You can use other data elements, but as you move down the tree, the ability to extrapolate data correctly becomes more difficult.
- #10: You can us the expansion lists as a guide, but in testing, people create their own abbreviations, they are applied unevenly,
- #12: OSU is in this boat – our primary cataloger is on sabbatical and our technicians haven’t been formally trained. This tool gives them the ability to look process records and start seeing what the data might look like