I thought that it was very appropriate to read this article right after talking with Phil and Leslie about 17 18. The basic complaint of the article was that students often get caught up in the syntax and semantics of a particular language and don't learn how to problem solve efficiently. Phil and Leslie attempt to counter this same problem.

In discussion today we talked about how their plans for 17 18 could be thought of as a breadth first approach. They intend to cover three languages and many different aspects of cs. While 15 covers OOP intensely, 17 18 will touch on OOP but move into parsing, algorithms and other things that I don't know about because I haven't taken anything beyond 15.

To be perfectly honest, I think that their approach might be better. Or maybe I should say that they sure sold it well.

Their course does not intend to have the glitz and glamour that 15 has grown. Instead, Leslie said that she expects people with more of a mathematical focus.

I think, in fact I'm nearly certain, that given the chance, I would have taken 17 instead. Hopefully it would have prepared me for later courses better.

But, just like this article states, you cannot teach all the language constructs and variability in a single semester. This could potentially put the students that are planning to take 17 18 in a disadvantaged spot. They might not know the code to race with those that took 15 16. But instead, what code they know, they might know better. I think that even to prepare me for 16 I might have been better prepared if I had taken 17. My visions for it are great. I see a course that introduces you to many facets of cs instead of the OOP facet, or the game/visual aspect of it. I think that this is a more honest approach.

To be honest, I see 15 as a dying course. Albeit amazing, I think that 17 18 is going to prepare students better for industry and for later courses. A data structures course is going to need to be taught at some point. I'm assuming that eventually eighteen will put them into it's curriculum.

I think that it will take several years but I'm expecting to see a 17 type course as the intro and a 16 type course as following. My reasons are long and complex but beleive me, they are just. Maybe this could be my research paper...what I intend, or would like to see as the first two courses in any CS dept. Whaddayathink?

I think that this could make a very good paper, but will require a lot of work. You need to fully understand the pedagogical goals of 15 and 16, plus the intended pedagogical goals of 17 and 18. However, a detailed evaluation would be incredibly useful to the department, and the CS education community in general.

I am not completely convinced that a BFS approach teaches CS better. I think that there are things to say about either a DFS or a BFS for teaching, and that different people will be attracted to one or the other. It seems to me that if the people are ones who are very interested by (and learn best with) a broad overview, in which they know how things fit together globally, then this would suit them. However, I think some people really prefer getting in-depth with one topic, and learning it really well, then learning a different topic well, and eventually synthesizing them all.

However, I think presentation of subject material is just as important (if not more so) than the material itself. So I'm really looking forward to see how 17 and 18 go, and hope that the people who take it get a lot out of it.

One of the most important things to remember is that people think differently and therefore learn differently. Although I firmly believe that CS17 will be very effective for some people, I do not believe that it will be the panacea for the department. Unlike some subjects, there is so many different aspects to computer science. It is important to introduce people to what will interest them. I don't believe that CS15 is a dying art. There are many aspects to it that attract a group of people that scheme interpreters never will. But the variety is only helpful!

I agree that 17/18 may provide a better way to teach CS to certain people, but not everyone. Just as you would like a math-based course, there are many other students who would prefer a 15-type approach. As Phil and Leslie said, 17/18 provides a different track for those students who feel that they might like a different approach to CS; I did not get the idea that they were making 15's replacement. On the contrary, I got the idea that two would run together. Maybe I'm wrong, or maybe it was the Robitussin...

I think the main point we should get out of this discussion is that there is no right way to do anything. Making general overarching statements like "I think that switching from blah-blah-blah to whatever is the better thing we can do" i think ignores the real issue at hand that we all seem to recognize but haven't really grasped yet. And that is exactly what Danah said, that we all learn differently. As soon as a system becomes diverse, like this system is, there can no longer be any one true Truth which we can all aspire to. We have to accept everyone's personal truths as a form of Truth and attempt to allow everyone to find their truths in the Truths we give them. Got it? I think it is pretty simple. :)

I'm not saying that it woule be the right thing to do, just is what I would predict...