When I was a biomedical library director, people frequently asked me why libraries had reaped so few savings from the introduction of electronic journals. I often answered by explaining that we were maintaining parallel systems: We had to keep processing print, because it still represented a substantial portion of our collections, while at the same time we provided access to electronic content. The cost of running two systems more than compensated for any savings from moving to digital delivery. Using a term from the business world, like “parallel systems,” seemed to make this rather mundane observation more persuasive. Of course, it probably helped that the answer also gave the welcome impression that sooner or later we would drop print and see some savings. The current issue of the Journal of the Medical Library Association (JMLA) provides a different perspective on the challenge of maintaining parallel systems. This issue confirms that health sciences librarians are definitely experimenting with new initiatives that make use of our ever more “high-tech” environment to improve access to information. At the same time, we still pursue the “high-touch” concept of librarianship that encouraged many of us to enter the profession in the first place. Elizabeth Workman et al.'s article on biomedical text summarization explains how new web-based applications can allow librarians to develop secondary databases in an automated fashion. Painstakingly curated lists of faculty publications and laboriously assembled bibliographies of current literature on a narrowly defined topic once were common in special libraries. They became impractical as the literature expanded, but with these new applications, we may be able to once again create specialized databases for our users. Equally new possibilities for health sciences information delivery are suggested in the brief communication by Valerie Lawrence et al., which describes the efforts in Washington to deliver electronic content to health practitioners throughout that state. Should this pioneering effort gain traction elsewhere, it opens up a whole new set of opportunities for health sciences librarians to reach health practitioners. At the same time, the issue contains two reports of activities in a very traditional area of librarianship, collection development. Ana Ugaz et al. report on the results of the latest update to the “Basic List of Veterinary Medical Serials,” while Amber Burtis and Mary Taylor have updated a 1997 study mapping the literature of health education. Statistics from PubMed Central indicate that the mapping studies for nursing are some of the most frequently downloaded papers from the JMLA, confirming that this traditional aspect of our profession is still very important to us and to our users. Similar evidence of the importance of traditional library functions can be found in Felicia Barrett's brief communication analyzing the decline in reference services at a regional academic health sciences library, and Sandra De Groote and Felicia Barrett's brief communication examining the use of print collections. Barrett concludes that although user groups may be changing, demand for reference services from our core clientele remains important, while De Groote suggests that in some circumstances there is still heavy use of onsite print collections. It is not easy to operate parallel systems, whether we are referring to the maintenance of print and electronic or the introduction of new roles while we still perform our old ones. I hope this issue, however, will help us to meet the challenge.