Elizabeth, great question. I have thought about this often, especially when people post UNIX-based solutions to questions on this list. Someone mentioned that it may depend on your research questions and types of methodological approaches to your questions, I think this is likely correct in most cases. I don't think it's reasonable to assume that social scientists will have all sorts of technical skills nor must they depending on their research. For a group (ours here) that is supposedly in tune with inequalities and differences in access and use of IT, it's often interesting to observe how some people ignore the fact that differences in abilities are likely present among our members as well. And let's not simply blame the students (or profs). These issues are often structural. After all, most graduate programs are not equipped to train people well in both the social sciences and computer science. I'll take this opportunity to make a pitch for our interdisciplinary PhD program in the School of Communication at Northwestern. In our PhD program on Media, Technology and Society (especially the Technology and Social Behavior track) we offer training in both. In traditional social science PhD programs a student will usually not get course credit for taking a course in a CS dept nor will s/he know where to go (e.g. with which course to start) to get some such training. We approach this differently. A few years ago when I was running the computer orientation session for the incoming cohort in a Sociology dept almost none of the students had ever logged on to UNIX. So while it may seem like a joke to run a computer orientation program these days, it's not depending on what you cover and what may be helpful for students. I suspect this lack of experience with UNIX is quite representative of other cohorts in social science programs nowadays. So to tell such people to go use wget or perl is not going to be very helpful. Numerous universities don't even support UNIX access to general users anymore and you have to jump through hoops to get such an account. But how to get such an account when you don't even really know what it is? So I think it would be helpful if people were sensitive to these differences when suggesting solutions to questions. I think it's great that people suggest all sorts of approaches, but this "Mine is better than yours" approach based on operating systems doesn't necessarily seem productive or helpful to the person posing the query yet hopefully the goal (or at least one goal) of responding to questions on the list is to help out the person asking the question. And by the way, there is no reason to have to switch away from Windows even if you do use UNIX to some extent. If you do have a UNIX account, you can easily access it using a program such as SSH Secure Shell. Eszter PS. In case you're curious, I'm a former Mac user, currently on Windows, still using Pine for some email (a comment that will be meaningful to other Pine users and pretty meaningless to others, I realize), and running some programs in UNIX including some scripts I wrote - with help from others - in perl.