Are the responsibilities of women and men in public policies addressing HIV/AIDS represented differently in policy language? Could these distinctions correlate with different outcomes for female beneficiaries of HIV/AIDS policy programs? Using interpretive policy analysis, this paper examines gendered responsibility narratives in public policy. We performed a mixed methods analysis of the publicly available documents from the President's Emergency Plan for AIDS Relief (PEPFAR), to determine differences in description of responsibility language around HIV according to gendered categories. We found that gendered responsibility narratives are pervasive across the body of PEPFAR documents and that two discourses concerning women's responsibility-women as problems and women as vulnerable-and two dominant discourses about male roles-men as violent aggressors and men as absentee-dominate the policy narratives. Our intent in this analysis is to deepen understanding of the role of gendered narratives as fostering differences in meaning that may or may not contribute to disparities in HIV/AIDS policy outcomes.

First of all, before reading this book, you must realize there is no magic bullet to counteract spam. As the authors point out, "The problem is largely based upon your perspective." One person's spam is another's legitimate mass mailing. Until society-and its laws, regulations, and technology-decides to address the issue, spam will be a fact of life. In this book, the authors set out to advise system administrators on how to combat this adversary. The authors have set up a Web site (http://www.slammingspam.com) with updates, including URLs, errata, and other information. This is to be commended, as the fi ght will be played out on a moving battlefi eld. The book focuses on the server side and has a decided Linux/Unix bias. Readers are not expected to be "Unix geeks," however, since in most cases, step-by-step instructions are given with accompanying explanations. The explanations should also be of interest to those who are not working with Linux/Unix. The book is organized into 12 chapters and seven appendices. Chapter 1 sets the scene and tempo of the work. Chapters 2 through 6 cover Procmail, SpamAssassin, native MTA (message transfer agent) anti-spam features, SMTP issues, and distributed checksum fi ltering. Chapters 7 and 8 discuss Bayesian fi ltering and how this technique is used in the fi ght against spam. Although Chapter 7 provides a good introduction to Bayesian analysis, and is probably suffi cient for this work, a dedicated reader might wish to consult a probability textbook. For those who don't care to understand the theory, the authors kindly grant permission to skip the discussion entirely and focus on the way Bayesian fi ltering applies to the problem. Chapter 8 describes, in turn, three fi ltering packages. Chapters 9 through 11 cover aspects of Microsoft Outlook, Microsoft Outlook Express, Microsoft Exchange (including a description of McAfee SpamKiller for Exchange 2.1.1), and Lotus Domino and Lotus Notes. Chapter 12 describes a collection of lesser-known open source products. The appendices provide extra detail;