Authority Control
RFP's and Tests
Of the several hundred authority control projects that LTI performs each year, relatively few involve responding to a Request for Proposal (RFP), Request for Quotation (RFQ), Request for Information (RFI), etc. Creation of an RFP by the library, as well as the vendor's response to it, are both time-consuming and expensive to all parties. Too often libraries tend to "borrow" heavily from other library RFPs or a suggested set of specifications prepared by an authority control vendor. In some cases LTI is asked to meet specifications that have not been current in years or, even worse, to meet a competitor's limitations that have been cast as "desirable" specifications. Vendors may respond with their own stock responses lifted from their service program literature.
In some ways purchasing authority control services is similar to selecting a local system. Vendors are not able to re-write their software for each new customer based on how that library thinks authority control should or should not be done.
The recommended approach to selecting an authority control vendor is to read carefully each vendor's documentation and then clarify issues and questions via telephone or email. For specific questions, email has the great advantage of allowing the vendor to make a thoughtful response, and gives both parties a written record of the vendor's responses. Of course speaking with the library's local system vendor, as well as other libraries that have used the authority control vendor's services, always adds valuable insights.
Should a library submit a test database to several vendors and compare the results? Asking vendors to run a test on several thousand sample bibliographic records prior to selecting a vendor can be a useful tool in deciding which vendor to use. Assuming the sample records are of sufficient number and drawn randomly, such a process permits one to make meaningful comparisons of how different vendors performed, as opposed to their marketing talents.
To be able to draw useful conclusions it is important that test records be selected in a random fashion-e.g., every Nth record from the database. If the library's systems people are unable to do this easily the library can FTP its entire database and request that the vendor select every Nth record. Once test records are extracted staff may want to include special interest records to see how vendors handle problematic headings. Keep in mind that, up to a point, the larger the database, the larger the sample should be. To ask an authority vendor to authorize a small test file of one thousand bibliographic records for a library having a database of one million records is largely a waste of time. Another aspect of sample size is that the library must have the resources to analyze and evaluate the work returned from the various authority control providers.
A complicating issue with giving vendors a small group of records to authorize is that the vendors are not going to treat the test as just another job. Since each vendor would like to win the job, test records are going to be processed with more scrutiny than the records from the entire database will receive. In fact, given a couple weeks to complete the test, a talented cataloger without any resources other than access to the LC authority records, could do a very credible job on a small database. One needs to remember that 70% or more of the headings will probably link to authorized heading without the vendor doing anything. If a library believes a test is necessary prior to awarding a contract, make it a requirement that the vendor complete the processing in as short a time as possible, e.g., certainly no longer than one week after the library's records have been received by vendors. That reduces, but not eliminates, the opportunity of the vendor to devote extraordinary attention to the test database records.
Pre-authority control tests can be useful in other ways. They can determine if: a) the library is able to extract its bibliographic records in MARC format; b) the local system ID or control number appears in every record (this will be the necessary match or overlay point when the bibliographic records are re-loaded after authority control); c) the library can send and receive files of bibliographic and authority records via FTP; and, d) the necessary load tables are in place that will allow the library to re-load the post-processed bibliographic and authority records back into the library's local system.
This last issue is extremely important since it may come as a surprise to the library that its local system vendor requires purchase of a new "load table&qupt; that, for library databases under 50,000 records, can cost more than the authority control processing. Libraries considering an authority control project should begin the process by first contacting their local system vendor to find out if there are any special requirements or additional charges applied by the local system vendor.
Caution should be exercised prior to adopting authorization specifications that have been lifted from an authority vendor's model RFP. Incorporating vendor-created boilerplate into the library's specifications may seem innocuous but it distracts attention from evaluating specs that are genuinely important, versus those that are fluff and self-serving.
An inane, but always mystifying, specification that turns up regularly in many library RFPs is that the vendor should: "Supply only authority records from the LCSH, LCNA and MeSH authority files. "Provisional" records are not acceptable. If vendor cannot limit output of authority records to authoritative sources, it must be so stated." One can only guess what the intent of this requirement is. LTI has been offering authority control services for two decades and, during that period, for LC name and subject authority control, we are not aware of any authority control vendor that has ever provided other than LC name and subject authority records. Upon request, as an option, some authority control providers, including LTI, are asked to generate so-called "provisional" authority records--but these by definition do not conflict with LC authority records. They are in fact identical to what the library's local system would create--i.e., a skeleton authority record containing only the 1XX for which there is no LC authority record distributed.
Another common suggested vendor requirement is to provide a report listing headings that match two or more authorized headings, as well as indicating the cost and format of the report.
Such a report relates to one's perception of how an authority control provider goes about its work of linking accurately as many library headings to authority records as possible, without making bad links. From LTI's perspective, by definition "two or more authorized headings" for the same unique heading cannot exist. Following authority control, a heading is either linked to an authority record or it is not linked. In fact these so-called "multiple matches" reports are a construct of the vendor's inability to deal with the necessity to "block" ambiguous and identical cross-references prior to linking headings.
Admittedly, there are some instances where vendor-blocked cross-references prevent a link. In these cases LTI editors either review manually the high frequency unlinked headings, or such headings remain unchanged and appear in the unlinked headings report. On rare occasions the Library of Congress distributes multiple authority records for the same heading. These can be the same heading or a variant heading appearing in the 1XX of the authority records. When this occurs, we notify LC in one of our weekly reports and LC resolves the conflict, often in the next weekly distribution of LC authority records. In summary, it does not seem reasonable that a specific vendor's limitations should become a "mandatory requirement" for another vendor that resolves the occurrence of multiple identical cross-references in authority records in a more rational and effective manner. Deluging a library with yet another set of reports simply removes the burden from the authority control provider and places it on library staff who must examine the reports and make their own decisions.
Computers excel at churning out reports but they are seldom a blessing. Unless a report presents summary data, its very existence implies that someone on the library's staff is going to have to review the information, either as a requirement to fix ambiguous or unlinked headings, or because the library believes it necessary to quality check the vendor's processing. Regardless, library staff are being asked to do work which the institution contracted with the authority vendor to do. Perhaps more troubling is accepting a vendor's "Highly Desirable Requirements" on the assumption that the identified requirements define the playing field. Not to diminish the efficacy of consultative marketing, but a company cannot sell what it does not have. For example, LTI alone among authority vendors offers a comprehensive continuing authority control service that updates not only the new and revised authority records but one that also returns the library's updated catalog records for overlay, thereby removing this thankless and time-consuming task from the shoulders of library staff. But for inexplicable reasons this essential component of any authority control update service goes unmentioned, as does the more common sense need for some level of editor review in backfile authorizations, under one-hour turnaround for authorizing new cataloging records, etc. There is a price to be paid for confining the library's requirements to what is available on one provider's menu.