On 7/15/2015 1:50 PM, McDonald, Stephen wrote:
> Bibframe has a relationship to RDA similar to the relationship > between MARC and AACR2. Bibframe and MARC are methods for storing > metadata. RDA and AACR2 are standards for deciding what metadata to > store. RDA replaced AACR2; Bibframe is intended to replace MARC. >
> Bibframe is compatible with RDA, and is intended to serve as a > storage framework for RDA with linked data capabilities. The plan is > that Bibframe will be able to convert to and from MARC, XML, and > other metadata standards, allowing existing metadata (from libraries, > publishers, and web services) to be used as a basis for future > cataloging, and allowing library data to be used in non-library > environments.
Bibframe and MARC are designed not for storage but for communication. Just as most library catalogs that use MARC records store them in relational database formats, Bibframe will allow catalog information to be communicated in RDF triples. These must be stored in a database that can store those RDF triples and can respond to special queries. There are many, many options for achieving that. Some options may use a RDBMS, or they may be completely different kinds of databases. There will undoubtedly be many more in the future. In addition, each of these databases can/will be structured in completely different ways, depending on the local needs. (Perhaps the best and “simplest” explanation I have seen is at http://www.dataversity.net/introduction-to-triplestores/)
So today, when you download a record using Z39.50 into a local library catalog, if you have a relational database (which is almost everybody), the system actually reworks everything that is in the Z39.50 record and places each bit of information into the correct table and cell. I know the most about the Koha catalog structure because it is open. You can see their table structure at http://schema.koha-community.org/. By going through the list, you can see that there are 165 total tables. Also, that the title (245a) goes into the table “biblio/title” but the publisher statement goes into another table “biblioitems/publishercode” and the biblioitems table also includes a complete version of the ISO2709 version of the record “biblioitems/marc” and also has another complete version in MARCXML “biblioitems/marcxml”. Of course, the library catalog you downloaded the original record from almost certainly has internal structures that are completely different from what is in Koha.
So, storing information can be done in a whole variety of ways, each determined by local needs. What is important is that the information can be queried in consistent and reliable ways, and then shared (communicated) in equally consistent and reliable ways. ISO2709 allowed all of that, but RDF uses different methods. Bibframe is attempting to come up with ways to do the queries and communicating using those RDF methods. There is nothing wrong with that, but it also doesn’t mean that internal workings of our library catalogs will have to take on the structures of Bibframe and be completely retooled–quite the opposite. Probably a few extras tweaks may be needed here and there, but nothing major.
When looked at only in this way, RDF offers nothing substantially new from earlier methods except a different structure. What is really new and exciting with RDF is the inclusion of the URI link and how developers can use the information found at the end of those links.
How it will all work in a library, and more importantly: whether the public will like it or not, is still unclear. There is certainly a lot of promise, but our “web technology junkyards” are overflowing with projects that were “promising”. So far, the Semantic Web has promised a lot but given very little that has been impressive. The much-touted Google Knowledge Graphs are pretty much useless. This doesn’t mean that the Semantic Web cannot work, but its successes so far have been limited.
Although I don’t agree with everything in this controversial article, it does at least provide a different viewpoint from what we normally get. https://gigaom.com/2013/11/03/three-reasons-why-the-semantic-web-has-failed/
James Weinheimer email@example.com
First Thus http://blog.jweinheimer.net
First Thus Facebook Page https://www.facebook.com/FirstThus
Personal Facebook Page https://www.facebook.com/james.weinheimer.35 Google+ https://plus.google.com/u/0/+JamesWeinheimer
Cooperative Cataloging Rules http://sites.google.com/site/opencatalogingrules/
Cataloging Matters Podcasts http://blog.jweinheimer.net/cataloging-matters-podcasts The Library Herald http://libnews.jweinheimer.net/