Beyond the goo-cloud: give us your new, your raw, your unpublished resources yearning to be retrieved…
Posted by Lynette Maldonado on Wed, Aug 24, 2011 @ 09:17 AM
By: Jill Moen, MLS, Technical Writer, EOS International
After the latest round of RDA webcasts, I now believe in the vision of a single-access, cloud-based, FRBR-compliant future (Bob Maxwell). In the bubble over my head, a huge, white, fluffy cloud containing all of the world’s resources floats along in a shared atmosphere for everyone’s use. I see it as having the expansiveness of Amazon, the ingenuity of Google, and the consistency of OCLC – so I will call it Amagoocle, or the goo-cloud for short.
In the goo-cloud, our FRBR-uture materializes as all new and existing works are harvested and cataloged to produce a single, unique record for each work. In 2005, OCLC reported over 93,000 bibliographic records for the Bible (held by over 796,800 libraries). In the goo-cloud, there would be one record for the Bible; Mother Goose, The Iliad, The Adventures of Huckleberry Finn would also be represented by one record for each work. Each record would then be linked to records indicating the expressions of the work – English, French, Bulgarian, text, performance, adaption, etc. Each expression would then be linked to records noting the physical manifestations of that expression – an edition, imprint, physical description, etc. As libraries acquire resources, no longer would they copy the details about the works over and over again into their own, respective databases. They would simply get in their EOS-based item cloud, scoot up next to the goo-cloud, and link their item records to the appropriate manifestations. While the item level information would continue to be maintained by and specific to each library, the information about the particular manifestation/expression/work is available to a patron, without duplication or further maintenance in the library’s database, whether that patron is searching the holdings of a library in Boise or Timbuktu.
This vision is exciting but what I like about the goo-cloud is that it reduces redundancy and allows us to focus on new collections.
In addition to standard content, research libraries have been harnessing the power of raw data for years by creating databases of scientific notes and pieces of information uncovered during the process of research itself. Academic libraries are following the lead of their research counterparts by building institutional repositories to capture dissertations, theses, and other scholarly and unique works generated by their institution. Neighboring Cal State San Marcos is considering unpublished undergraduate work for their repository to allow access to what is basically unmediated research which can be used as a basis for further study. In a webcast presented by Darren Poley at this year’s ACRL conference, Poley pushes the envelope even further by suggesting libraries move beyond institutional boundaries to reach into, and partner with, surrounding communities to create or bring to life previously hidden collections.
Working with special libraries as we do here at EOS, we see new content springing up everywhere, databases dedicated to image archives, tape libraries, internal manuals, reports, corporate communication, scientific data, textiles, and artifacts as well as specialty databases that reflect parts of a whole – such as articles, elements of court proceedings, bits of research data – and other forms of corporate and competitive intelligence custom cataloged for specific patron communities. It’s exciting!
EOS client, San Diego Air & Space Museum, recently created a group of custom databases including one to integrate over one thousand oral histories directly into their MARC-based collection. Using Excel to bypass the obligation of having to grant dozens of volunteers access to their ILS, Head Archivist Katrina Pescador and Library Consultant Kathy Wright have been able to define the format for each of the new databases, establish guidelines for content (e.g., COSATI for Technical Reports), and review work done in Excel before importing the data into the Museum’s online catalog. The Museum is currently partnering with two other local museums to catalog more than 3000 audio-visual items. Materials such as technical drawings and biographical files are also being prepared for upload. “We have a volunteer who has just started to edit more than 6,000 drawing records,” says Wright.
These are the exciting collections originating from and showcasing the work being done at the Museum to give form and access to the technical output and knowledge of the Museum’s communities – something the goo-cloud isn’t planned to deliver.
So let us begin cataloging all that is buried, wild, and scattered around us to turn information in its rawest form into searchable stores of knowledge for relevant retrieval and exchange.
And really, until the folks at Apple figure out a way to put infinite knowledge into an i-orb that attaches to our foreheads, reads our minds, and transmit to us what we need to know (while, of course, implanting product endorsements and extracting our deepest, darkest secrets), libraries will have to continue to shoulder the responsibility of mediating access between patrons and all that is known whether the information exists in the goo-cloud or surfaces through the cataloging of new, raw, unpublished resources yearning to be retrieved!
Quick plug: In EOS.Web, metadata-friendly templates can be created for use in your library’s database by adding new rule sets, tags, mappings, indexes, etc. or using KnowledgeBuilder. Both methods allow bibliographic templates to be created with custom entry fields specific to the information being cataloged. For example, in newer databases, we have added KB and MARC record formats based on the Dublin Core guidelines to the list of default rules sets available in EOS.Web. (The DC templates were created based on the DC element set. LOC’s Dublin Core to MARC crosswalk was used to determine field labels and mapping. The import/export format is MARC for a MARC-based record format and CSV (Text or Excel) for KB-based record formats.)