Computery grad class

I should be thinking about the summer capstone seminar for the Masters in Informatics program I’ll be teaching in May, or either my undergraduate class (Media in the Information Age: hands-on version) or graduate (Com Theory) courses for the fall. Instead, I am trying to figure out what I’ll be teaching in the Spring. I don’t want my normal problem: having to drum up people to take the course! So I’m staying away from “topics” courses, and I am trying not to make it super intimidating. I am also hoping to appeal to some of the techier MI students next year, since the communication grad students seem relatively tech-averse, for some reason.

Right now, there are three possibilities and these are not at all mutually exclusive:

1. Programming for Informaticists. This would be roughly what Uta Priss taught at Indiana. In fact, when I found her syllabus, I was blown away: I wouldn’t have taught the course any differently. It would include basic scripting in Python for the standard sorts of tasks we do: surveys, data collection from the web, and the like.

2. Computer Applications in the Social Sciences. This might include a touch of python, but it would likely take a look at some basic data analysis systems, including software for content analysis, note taking, visualization, and the like. We might even push toward use of tools like R and ggobi (which, obviously, would benefit from a grounding in Python, though certainly does not necessitate it). Finally, we could talk about agent-based modelling, using swarm and other systems. That should only take a few decades.

3. Social Sciences Visualization. Another alternative is to look more directly at issues of visualization, especially of network data: something akin to this course.

Ideally, I guess, I would like to do all of the above. But I generally am overambitious in my course planning and end up trimming way back once the semester begins. A saner approach might be to focus on #1, and spend an inspirational segment of each class looking at “what you can do with your pythonic prowess.”

(As an aside, anyone who has seen my code knows that I am not a programmer. I am, rather, a code bricoleur. I hope to pass this on to students. I don’t expect them to go on to be application programmers, but I want them to be able to bang out a few lines of python when they need to manipulate a large text file, survey data, or the like. I want to create bricoleur scriptors rather than “real” programmers. Later, they can curse the bad habits I taught them!)

This entry was posted in Uncategorized and tagged . Bookmark the permalink. Trackbacks are closed, but you can post a comment.

2 Comments

  1. Cory
    Posted 3/13/2003 at 9:52 am | Permalink

    I’m definately not the best student in the world. Actually, up untill Fall ’01 I was probably the worst student ever, making my chance at higher education than a BA very hard. Plus I’m too poor, and really need a job come May. Whether I have credibility or not who knows. But, I always have an opinion! I read a few books by Tufte this past semester regarding the display of visual information (writing’s dry, but helpful). I would definately lean towards that, and find that most helpful for myself. I think as technology and media keeps converging, and the possible thought of the “do it all” machine, the ability to make the best out of any given “real estate” becomes invaluable. To me, in all sense, any kind of information can be visually displayed. That ability to visually display information at the most understandable way will always be a desired trait. I think of how the GUI and Windows made the information display of PC applications soar. Imagine doing what we do now in ASCI art. By the way, do you think that if I went ahead and learned all those programs and languages listed it would kind of be like I took a few graduate courses for free? Mmmmm free…

  2. alex
    Posted 3/13/2003 at 11:03 am | Permalink

    I went to one of Tufte’s seminars when I was in Seattle. I was pretty taken by his books, so I thought I’d get more out of the seminar. I’m not sure I did. What I like about the Tufte books is that he collects some very cool visualizations. But I don’t think I agree with all of the conclusions he draws from them. The “get as much data in per square inch as possible” mantra, for example, simply does not cut it as far as I am concerned. Sometimes, less really is more.

    Anyway, this visualization class would be less that and more scientific visualization. That is, the Tufte sort of design visualization–although, as you note, it applies to UI design–is less what this would be about. It’s more about displaying very complex, very high-dimensional data in a way in which the social scientist can find patterns. Think of a stock broker wanting to represent visually all of the data, including historical data, available on the stock exchange and from related sources. How can you get as much of that information to the stock broker as possible. Clearly, it has to be visually, but there are a lot of ways to think about presenting that data and making it interactive.

    As far as learning the programming stuff on your own, that’s really the only way to do it. This class will set up milestones and provide students with the impetus to go out and learn to program, but, in my humble opinion, this isn’t something you can really “teach.” (I’m not sure there is anything that isn’t this way, but that’s another posting.) If you ask most programmers how they learned, they will tell you they went out and got a book, or bugged someone they knew, and just played around for a while until they got the swing of things. This course would just provide a little structure as a kind of safety net for those who are not as comfortable playing around with computing systems.

    Would it be like taking a grad class, if you did it on your own? This would be an unusually “hands-on” class for an academic graduate school. Honestly, I think one of the primary reasons for graduate school is the chance to hang out with other smart people. This might sound a bit callow, but I think it’s true. Certainly you learn things at the time, but one of the reasons you learn things is that you can bounce them off of your classmates and your profs in a way you don’t get a chance to at the undergraduate level. So yes, in terms of the material there really isn’t much you can’t do on your own. But you may find yourself starving for the kind of interactions that are helpful to your intellectual development. Certainly in the case of a course like this, which is skills-based, most of the students will really be learning it on their own anyway. I only hope I can provide a supportive environment for them to do this effectively.

    Finally, I think most studies show that graduate school pays for itself in the long run. That is far less true in, say, English or History than it is in say Business, Engineering, or Law, but it’s true of graduate school in general. Perhaps more importantly, you only live once, and I thought it was a worthwhile experience, personally. Many people, however, do not find it a worthwhile experience.

    One nice path for this is to work youself into a job where they will pay for your graduate school. You would be surprised at how many of these are out there. I think that working while you go to graduate school is a waste of time: grad school needs your full attention to be rewarding. I had to work during the first years of grad school, and I know I didn’t get as much out of it. But it lets you try before you buy, and do so on someone else’s dime. You take a course or two, decide if you like the environment, and decide if it’s worth the financial and other costs.

    In any case, I always thing that those who go out into the working world after the BA and then return to grad school do better than those who try to go straight through.

Post a Comment

Your email is never published nor shared. Required fields are marked *

*
*

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>

  • Tweets

  • Archives