ماتریس E ؛ انتخابی برای رشد سیستم مدیریت منابع الکترونیکی

Even with commercial options now available, the North Carolina State University (NCSU) Libraries has chosen to continue developing a homegrown electronic resource management (ERM) system, E-Matrix. This contributor provides a brief description of E-Matrix and outlines the reasons NCSU chose not to go with a commercial product. He also discusses changes to technical services workflows and presents lessons learned by NCSU during the development process. Meyer offers recommendations to consider when implementing either a homegrown or commercial ERM system.

At its core, E-Matrix, a homegrown electronic resource management system, is a staff management tool for serials, journals, and electronic resources. The system brings together the various data sets about the North Carolina State University (NCSU) Libraries' collection which were previously stored in separate information silos. Information in E-Matrix includes data from NCSU's catalog, link resolver knowledge base, and evaluative data collections stored in small Access databases and Excel spreadsheets. Bringing together the various information serves to create a centralized repository for managing information about serial resources, such as databases and journals. The system is accessible through a Web interface for staff to look up information and perform tasks for library materials.

The NCSU Libraries has learned many lessons during the development of E-Matrix. The first lessons involved the state of the data we were pulling into the system. Nonstandard data, such as the kinds found in our knowledge base or various acquisitions and serials data sets, turned out to be much more reliable than some of the standardized MARC data that libraries rely upon so heavily. Part of the reason for this is that these data sets were designed with modern systems in mind. One of the primary tasks of an ERM system is to reveal the relationship between an aggregator database and the journals to which it provides full-text access. These nonstandard data sets were designed to accomplish exactly this task. On the other hand, the MARC record provides very few ways to make explicit the link between the same two records for a journal and the database. While MARC records have been modified and adapted organically over the last couple of years to display the relationships in note fields or 856 link fields, the MARC record does not contain the links between its records that a machine needs to efficiently process and reveal the relationships. For any library thinking about incorporating a new ERM system into their environment, there are many things to keep in mind: ! Patience. These are new systems with no histories, no track records, and no time-tested best practices. ! People. The technology behind these systems is only one half, if that much, of the equation. Be mindful of the people factor. Many of the users of ERM systems will feel as though they have new jobs ! Planning. The successful adoption of a new technology will likely require an equal period of planning before and beyond the mere implementation. ! Research. There is a wealth of literature that exists on ERM. Chief among this body is the work done by the Digital Library Federation’s Electronic Resource Management Initiative (http://www.diglib.org/standards/ dlf-erm02.htm). For those thinking about designing their own systems, there are many other recommendations that our experience has taught us. In terms of staffing, you will probably need a dedicated programmer and a dedicated librarian to be successful in the project. Both will need a strong understanding of relational databases and strong communication skills. Know where your data currently reside, what format it is in, and what hooks are available to link data from separate systems (e.g., an ISSN). Avoid premature feature creep at all costs.1 If your system is designed well, it will be extensible. In conclusion, we have been very fortunate at the NCSU Libraries to be able to devote the necessary staff resources to E-Matrix. Through the development process, we have gained a better understanding of our collections and work processes. It is important to note that, for NC State, ERM is not simply about creating the systems for data management but also about understanding patron needs and the institution’s goals. Effective e-resource management is about translating all of these layers into a usable system for both library staff and patrons.