Okay, so I've been talking about FRBR for a very long time. FRBR, or Functional Requirements of Bibliographic Records, sounds scary and over-whelming. In a nutshell, it provides the logic to connect multiple editions of a work making for easier customer access to materials. In other words, search a library catalog for the classic "Scarlet Letter." You'll receive dozens and dozens of results, and if looking for an available copy will have to check each item.
Now search Amazon for "Scarlet Letter." Select one of the book entries, and there is a nifty box under the basic title info that links you to other editions of the same title.
Library vendors have struggled with the concept for a long time, and haven't found a good way to implement the logic to make this happen. The wait is over for libraries! Library Thing for Libraries offers an catalog add-on option that will do exactly what Amazon has been doing for a very long time. Check out the Salt Lake County Library catalog for one example.
Now that the logic is in place, the next step would be to add availability notations for other editions, making it easy to know which one to click on... much the same way that Amazon lists the price info. Oh, and integration is good for classic webpacs, but lacking with the newer discovery platforms (Encore, Aquabrowser, etc.)... I'm sure that will unfold as well.
Thanks to Tim Spalding and his team over at Library Thing for this little breakthrough.
Friday, February 1, 2008
Subscribe to:
Post Comments (Atom)
4 comments:
Hi,
Please check out how we do FRBR in AquaBrowser at Carroll County. Search for Scarlet Letter in their catalog and see how the results are already rolled up to show 1 record for 8 versions editions.
Jasper
Hi Jasper,
Very nice!! and Hooray! I'm thrilled to see an implementation of FRBR in a search discovery platform.
The display is clean and easy to follow. Can I assume that you are using one of the existing OCLC algorithms to make the connections?
Also, I'm curious to know if you have looked into displaying availability on the initial screen. I know these info-rich screens quickly become cluttered, but that is the missing piece of information to help guide customers to the right edition to click on. So often, especially with our Scarlet Letter example, the customer simply wants what is available. Displaying this info on the initial screen provide the tip on which edition to start with, opposed to a closed-door link.
Kudos to your team at MediaLab, Aquabrowser and Carrol Co Library for becoming a forerunner on a real FRBR implementation!
j
Hi Jessi,
Thank you.
Actually, we do not use the OCLC algorithms. Instead we implemented Syndetics ICE: Indexed Content Enrichment. That product allows you to add extra data to the index that AquaBrowser uses. For example fiction profiles and table of contents data help users to find titles without having to load this information in the MARC records.
One package in Syndetics ICE is title link. Title link is what you see working in the example of Scarlet Letter. The source data for FRBR is the data from Books in Print. Various editions of titles are linked to each other in that database already! If a new isbn is issued for at title, it is linked to the other editions immediately. AquaBrowser leverages this data.
The availability in the initial results screen is a feature we are working on. However, AquaBrowser has to fetch the real time availability from the ILS, which is not always too fast. Oklahoma State University shows the availability on the initial screen, but it is not real-time and therefore not 100% guaranteed accurate.
Regards, Jasper
Library vendors have struggled with the concept for a long time, and haven't found a good way to implement the logic to make this happen.
Well, not all vendors. For what it's worth, Evergreen has been grouping records using a algorithm that can be locally adjusted since, well, before it had an OPAC -- we call them metarecords, and you can place holds across them as well as see them as results. And, not that it's a competition (well, it sort of is, right?), but we had implemented an ISBN cross-referencing web service (the LibraryThing analog of which is the basis of the "formats and editions" functionality in LTFL) earlier than LibraryThing's thingISBN.
All that being said, there's nothing better than seeing more products use these types of algorithms (and mechanical turks, in the case of LTFL) to enhance patron service. I find it very gratifying when we implement something that others then implement, whether copied or simply simultaneously invented. It's an indicator that we're on the right track. We all stand on the shoulders of giants, and I like to think that Evergreen sometimes plays the role of one. :)
--
Mike Rylander
| VP, Research and Design
| Equinox Software, Inc. / The Evergreen Experts
| phone: 1-877-OPEN-ILS (673-6457)
| email: miker@esilibrary.com
| web: http://www.esilibrary.com
Post a Comment