A number of studies have shown that paper holds several advantages over computers for reading tasks. However, these studies were carried out several years ago, and since that time computerized reading technology has advanced in many areas. We revisit the issue of reading in the workplace, comparing paper use to state-of-the-art hardware and software. In particular, we studied how knowledge workers perform reading tasks in four conditions: (1) using paper, (2) using a dual-monitor desktop system, (3) using a pen- enabled horizontal display surface, and (4) using multiple tablet computers. We discuss our findings, noting the strengths and shortcomings of each configuration. Based on these findings, we propose design guidelines for hybrid horizontal + vertical systems that support active reading tasks.