quarta-feira, 18 de julho de 2007
Fim dos velhos catálogos
Pensava que o assunto já estava definitivamente na "prateleira" quando hoje recebo este artigo publicado no Library Journal que, pelo seu interesse, transcrevo.
Demise of the Local Catalog
By Roy Tennant -- Library Journal, 7/15/2007
Over four years ago I wrote that the library catalog should be pushed into the back room where it belonged and that a unified finding tool be put in its place (LJ 2/15/03, p. 28, and LJ 6/15/03, p. 28). I still believe our users want to find everything they can on a topic in one place, with one search, rather than going from place to place puzzling out each new interface. Well, now finally that's happening in a couple of interesting ways.
Most of the major ILS vendors, including Innovative Interfaces, Inc. (III), Ex Libris, and SirsiDynix, are either working on a product in this space or already have one. This new kind of finding tool goes beyond federated searching. Metasearching has always been a technique of last resort, so these new tools are all about locally aggregating everything you can and then metasearching what you can't (i.e., sending out parallel searches to multiple remote sources).
Our library catalog information is part of what can be centrally aggregated. That is, library catalog records may be imported into this new system and indexed along with other information. For example, metadata records from institutional repositories and museums can be gathered together by using the Open Archives Initiative Protocol for Metadata Harvesting (OAI-PMH). Article indexes can be locally indexed along with catalog records. Web pages from selected web sites can also be prefetched and indexed.
Ex Libris's Primo, for which several institutions have served as development partners, and SirsiDynix's Enterprise Portal Solution provide unified searches. In a similar fashion, about a dozen institutions agreed to work with III to develop its Encore product. All these portal solutions implement a variety of technologies, from focused web crawling and indexing to OpenURL resolution, metasearching, and indexing of local catalog holdings.
Some systems even promise FRBR (Functional Requirements for Bibliographic Records) to unify the display or related records, faceted browsing, and robust relevance ranking. Time will tell just how many of these promises are delivered and how effectively they are implemented. All these products minimize the time and trouble our users must take to find what they need.
The role of the ILS
Clearly, the typical integrated library system (ILS) will no longer be the centerpiece of a library's online offerings. Rather, a centralized finding tool will be the focus. The ILS will not go away, since it is still required to perform such essential tasks as acquisitions, inventory control, and circulation. But it will be relegated to the back room where it has always belonged. That isn't all. The ILS must provide local holdings information to these unified finding tools for batch indexing, as well as circulation information on demand.
This last bit is the tricky part. It's possible to export records from virtually any ILS, but not every ILS offers an application program interface (API, or a way for different software programs to share information) for providing on-demand circulation information. This may improve, but it isn't yet clear how quickly.
All in one
The University of Washington Libraries is using a tailored version of WorldCat.org, dubbed WorldCat Local, a new option from OCLC. (Disclosure: I am now employed by OCLC.) Essentially, it is WorldCat with the institution's own brand and search results ranked by location as well as relevance. For any search, locally held items will appear first, then those from a regional cooperative, if appropriate, then all other WorldCat holdings. Persistent users would eventually see all the items available to them locally as well as through interlibrary loan.
The move to central finding tools is a huge step forward for individual library users. People generally want to find everything they can on a topic, ranked by relevance and displayed in ways that make it easy to narrow in on their goal. Increasingly, there is no reason why we can't start giving it to them.
Roy Tennant (firstname.lastname@example.org) is Senior Program Manager, OCLC Programs and Research, Mountain View, CA. He is author of Managing the Digital Library (Reed Business Pr., dist. by Neal-Schuman