What’s Expedient vs what’s good - curriculum vs teachers
If we don't do anything then only schools like Stuy will have CS.
I heard that the other day. It was a comment relating to a number of efforts, both in and out of NY to get CS into the classroom.
These programs abound and up front let me say that they're all well meaning and have the best of intentions.
In NYC, programs receiving money, publicity, and support, include:
- The Software Engineering Pilot (SEP)
- Google CS First afterschool
- The NSF grant bring the Beauty and Joy of Computing class to NYC (BGP)
Terrific – we'll scale up CS ed in NY from nothing to everyone in no time. It's fast, it scales. But is it good.
I'd argue that maybe it's not. These programs all share a few things in common.
First, they're all expedient. You can implement them in an election cycle. Roll out a pre-fab curriculum with scripted lessons, give the teachers a summer and some weekend workshops and we're good to go. No CS knowledge needed.
Next, as far as I can tell, none were developed by experienced high school teachers with a deep knowledge of CS.
The SEP was originally developed by someone, who, according to his linked in profile never coded and never taught.
I could be wrong about BGP. It was developed by Dan Garcia - an educator whom I like and respect, but as far as I know, his experience is at the college level.
And Google's CS First? I've been very impressed by Google's engineers, in fact, many are my former students. It's high school education team? Not so much.
Finally, these efforts buy into the myth that anyone can teach and that anyone can teach computer science. Even worse, the Google CS First claims people running their program don't need to know a thing about computer science.
Anyone who knows teaching knows that you can't learn a subject well enough to reall teach it in such a short period of time. Of course, it does align itself with education "reform" where the true master teacher is no longer desirable and we should all be drones that teach lessons from scripts and all testing shall be standardized.
I rememer the first time I saw a progarm like this - it was the CISCO Networking Academy. Teachers trained for a few weeks in the summer and the course was basically run via computer based tutorials. I can't speak for it nationally but in my experience, what the students learned was superficial - very test oriented and there was no value added by the teacher - even after years of "teaching" the classes. There's a reason why the few networking lessons in our System Programming class are called "CISCO in an hour." The kids learn more in those couple of lessons than in a year of the CISCO curriculum.
Who was to blame? Cisco? They were trying to do something good. The teacher? Partially, after time they should have learned something of networking but they were very much out of their element having started the process from zero tech experience.
The fact is, that got a "Computer Networking" class into the schools. It might have been meant as a stop-gap until we could get real Computer Networking teachers hired or trained but it ended up being good enough. The end result – bad classes and students not learning what they could.
So here we are back again at the "let's take non CS teachers, give them some PD and we'll be good." Rather than take the time to solve the CS Education problem, we're going to throw down a curriculum, do some PD and say that it will buy us time. The truth is, however, that we're going to say "look, we have CS in the schools" and it will be good enough.
I'd argue, though that for CS, good enough is particularly dangerous. It's very easy to work through a scripted and coached program and end up with kids thinking they know some real CS and have a real product yet having not really learned the subject. This won't be exposed until later, when they're gone and in college. I predict that unless something changes, these efforts and similar ones will drive up the number of people who after this exposure will say they love CS but once they get hit over the head with "funcitonal programming in HASKELL" or whatever real CS they see in college, they'll drop like flies.
What's the answer? We've got to stop doing what's expedient and do what's good and what's right.
Part 2
A couple of other blog itches I want to scratch are the many after school, summer, and drop in programs to go with the there I mentioned above. Hopefully I'll get to rant about them soon.