reason, there is room for a wide variety of approaches in developing our future engineering and science data liaisons. Libraries that wish to develop data liaison competence face big challenges as they respond to the staggering opportunities ahead. First and foremost, underlying systems and organizational structures must be in place in their libraries and universities, so that the library system can offer a strong suite of services to the community that are backed up by a robust and continually improving infrastructure. Leaders in the organization need to seek out and foster existing liaisons’ interest in data services by bringing like-minded individuals together, introducing them to faculty and community members involved in data work, and providing funding for directed skill development and conference attendance. In addition, experimentation, proposals, and constant investigation of the disciplinary landscape must be the order of the day. Library organizations must also seek to recruit liaisons with strong analytical skills or with domain-specific knowledge. These activities, in concert with strong staff effort will produce outstanding data liaisons on whom to base a program of data services. At first glance it may appear that adding data liaison services on top of an already full service portfolio at a university library system is too much to ask right now. But it is important to realize that services like this are a major component of libraries’ future. Ensuring that the complex output of the research enterprise is collected and is reusable by others is central to the ongoing mission of research libraries. Liaison roles are already undergoing an evolution that has reduced their role in traditional areas such as on-site reference service and acquisition of collection content. Researchers’ needs are evolving and data liaison services have great potential to offer a new model for subject liaisons. Where subject liaisons have created value and enjoyed great intellectual satisfaction from their work carefully building subject collections or tracking down difficult-to-locate literature, future data liaisons will find rich and varied intellectual challenges in tackling the data needs of our science and engineering researchers. Many thanks are due to a host of people who wrote about these issues, talked with me and provided inspiration and the insights contained here: MacKenzie Smith, Anna Gold, Katherine McNeill, Anne Graham, Amy Stout, Erja Kajosalo, Patsy Baudoin, Katherine Clopeck, and Howard Silver. RLI 265 20 The Last Mile: Liaison Roles in Curating Science and Engineering Research Data ( C O N T I N U E D ) AUGUST 2009 RESEARCH LIBRARY ISSUES: A BIMONTHLY REPORT FROM ARL, CNI, AND SPARC