Yesterday was the last day on our first course for teacher certification. A programming course similar to a college CS1 - think APCS-A. We're now moving to a data structures course.

There are a few reasons for this. First, it's depth of knowledge. The most advanced class a high school student will normally take would be APCS-A. Data structures is the next course. A teacher should have studied a topic to a greater depth than the students. We fill in the breadth in our topics course. Now, this means that our elementary school candidates need data structures and I'm the first one who'll agree that elementary school teachers don't need the same depth of knowledge as a high school teacher but New York State in it's infinite wisdom decreed that their CS Certification would be K12 - same thing for all teachers from Kindergarten through the end of high school.

A second reason, and this is a big one, is that studying data structures is a computer scientist or programmers first opportunity to understand what's going on under the hood. Why is this important? Well, to me, it's one of the differences between what, for lack of a better word I'll call a coder or programmer with a computer scientist.

I'm not talking about a computer scientist as a college employed researcher, I'm using the term as someone who problem solves and creates solutions using the tools and techniques of computer science. For a coder / programmer, I mean someone who creates working programs but doesn't think deeply about the solution but rather uses built in constructs, libraries and packages blindly.

This is not to say that someone employed as a coder can't be or isn't by my definition, a computer scientist, they certainly can.

Back to the point. We closed out the programming class learning about classes and implementing something that we called a SuperArray, basically a cut down Java Arraylist.

On day one of data structures we went over the ArrayList. This is our usual strategy - implement something and then you learn about and can use the built in one.

A poorly trained programmer will just learn about the ArrayList and then use it blindly. But by first exploring something like the SuperArray, the computer scientist has deeper knowledge.

An ArrayList has a number of nice features. You can keep adding to them paying little attention to size, you can delete from the middle, insert anywhere and more. All these operations take time. If you try to remove an element from an ArrayList, the ArrayList has to shift all the elements past the element in question down. If you want to insert at the front, you've got to shift everything down to make room. Adding something to the end, however, takes no time at all unless the underlying array is full, in which case, more work is to be done.

When writing the SuperArray, students had to write the code necessary to do all of these operations so we can discuss the good, the bad, and the ugly.

If you are using an ArrayList and you blindly do all your insertions at the front, you're going to have horrible performance but if you always add to the end, things will be much faster. The coder doesn't understand this but the computer scientist does.

When we do LinkedLists next week we can dig deeper on these ideas and even more when we get to sorting and searching.

The bottom line is that even if one doesn't write an entire ArrayList, exploring under the hood can arm a coder (or teacher) so that they make wise choices instead of treating every problem like a nail.

A side added bonus on this specific SuperArary / ArrayList project is that when you're done you can actually look at the Java ArrayList code. The class won't understand all of it but they should be able to get the gist and they'll see that the Java code is very similar to what the students wrote. This can be very empowering.

After this data structures course, the teachers may never again write the code for a linked list, implement a hash table or tree or manually sort a data set but the underlying knowledge will help them understand and teach all sorts of computer science concepts. A web page is represented in a a tree, a network is a graph. Understanding hash tables and search trees are a gateway to databases.

Sometimes the specific implementations one studies are never used again but for teachers, the underlying concepts can be game changes for their students.