Transcription

1 Introduction to Computer Programming Concepts in StarLogo TNG Overview In this class, we will use the StarLogo TNG language to introduce basic concepts used in computer programs. Learning Objectives Over the course of this class you will learn What is a computer program Abstraction & decomposition of a problem Algorithmic thinking Instructions Conditional instructions Evaluating expressions Repeating behavior How to create a control structure incorporating the above. About Computer Programming Computer programming is the process of planning and creating a sequence of steps for a computer to follow. In general, this process will help us solve a problem that is either too tedious or too difficult to work out by hand. About StarLogo TNG StarLogo TNG ( The Next Generation ) is the newest branch of the StarLogo language. It was created to 1. Lower the barrier to entry for programming by making programming easier. 2. Entice more young people into programming through tools that facilitate making games. 3. Create compelling 3D worlds that encompass rich games and simulations. StarLogo TNG provides two significant advances over the previous version, StarLogo 2.2. First, the programming is now done with programming blocks instead of text commands. The primary advantage of this innovation is that the program has moved from the abstract to the visual. As many students are already familiar with using a computer by dragging and dropping objects, StarLogo TNG comes more naturally to them. The programming blocks are arranged by color and shape based on their function, and it enables students to associate similar programming blocks with each other. Since the programming blocks are puzzle piece shaped, only blocks can fit together only in syntactically sensible ways, which eliminates many sources of problems for students. StarLogo TNG also provides visual organization capabilities, such as breed drawers, which provide similar functionality to object oriented programming. StarLogo TNG s second significant advance is a 3D representation of the world. This in turn presents several new opportunities. These include the ability to model new kinds of physical phenomena, and a feature that allows students to take the perspective of an individual agent in the environment, which helps them bridge the description of individual behaviors with system level outcomes.

2 Necessary tools StarLogo TNG is available for free download at StarLogo TNG requires a 1GigaHertz (GHz) or better processor and a graphics card that supports OpenGL. We recommend 512mb of RAM or more, but it may run with less. If you are running StarLogo TNG on a Windows System, we recommend a Pentium 4/Celeron/Athlon 1.8GHz or better or Pentium M 1.4GHz or better for the procesor and a nvidia (GeForce 5200 or better) or ATI (Radeon 7000 or better) graphics card. DirectX is required. If you are running on a Macintosh System, any newer 1 ghz or better imac, emac, Powerbook, or ibook will do. The Macintosh version is a Universal Binary so it runs natively on Intel and PPC Macs. You will need Mac OS or higher along with Java 5 release 4 or later installed. If you don't have Java 5 release 4 or later installed, it is available via Software Update or directly from Apple for PPC or Intel machines. Our approach The approach we use in this course is to allow students to construct on-screen computer games in which the concepts of agents with behaviors and environment is made tangible and visible. Onscreen agents (or objects) in StarLogo TNG populate a three-dimensional environment. Students create computer programs by dragging and dropping program blocks into place on the block canvas. We will introduce basic programming elements (if/then statements, loops, variables, expressions) using the TNG blocks and give students an early exposure to event handling. Web references StarLogo TNG documentation and video tutorials are available at Computer Programming basics What is a computer program? A computer program is a set of instructions that tell the computer what to do. Computer programming entails decomposing the problem into a sequence of steps and specifying them sufficiently, precisely, unambiguously, and primitively such that the computer interpreting the program can effectively realize the intended solution. Abstraction and decomposition of a problem By abstraction, we mean learning how to communicate complex ideas simply and to decompose a problem logically. For example, think how you might tell a robot how to bake a cake. You might break the problem down into smaller pieces: make the batter, bake the cake, decorate the cake. Each of these can in turn be broken down in to smaller pieces: make the batter becomes beat the eggs and the butter, stir in the flour and sugar, then pour batter into a baking pan. This process is called problem deconstruction and it simply means breaking a problem down into a list of simpler tasks. Algorithmic thinking Algorithmic thinking means thinking in terms of instructions that the computer can understand. Once a problem is decomposed into smaller pieces, we can think about how to tell the computer a set of instructions to carry out each of those smaller pieces. Algorithmic thinking is thinking about arranging a sequence of instructions to carry out a task.

3 Instructions An instruction is the simplest kind of direction to give a computer. In StarLogo TNG, a simple instruction (also called a command) you might give is go forward two steps. The command in TNG would look like this Another simple command is We say an instruction is executed when it is run by the computer. Arguments Some commands like the ones above need arguments (in these cases, numbers) as part of the instructions. These numbers tell the computer how much to go forward or how many degrees to turn. Here s an example of an instruction that takes two arguments, an x position and a y position. To complete this command, we need to add the arguments that fit to complete the command, for example, sets the turtle s position to 3,8. When modeling complex systems often we want some degree of randomness in an agents behavior. Suppose that instead of always turning 90 degrees to the right, we want to turn some random number of degrees to the right. To do this we would use the random function This function returns a value between 0 and 89. When used as the argument to an instruction, the function provides a new random value as an argument to the instruction each time the instruction is executed.

4 Conditional instructions More complicated instructions are used to tell a computer what to do under certain circumstances. For example, we might want to say go forward three steps if nothing is blocking the way. For this we need a conditional instruction. Programming with conditional instructions allow for steps to be carried out only in specific situations. This is what the if/then block looks like: To use it, we need to insert a test condition and tell the computer what to do if the test is true. Evaluating Expressions Expressions are questions that get resolved to a value. Conditional statements like the one above make use of expressions to check a current condition. Often checking this condition requires a comparison. For example, the test condition could be the if the turtle is green. The expression turtle is green gets evaluated as true or false. When used in a conditional statement, the test condition is the expression. If expression evaluates to true (because the turtle is in fact green) then the statement(s) in the then portion are executed. In this example, if the turtle is green, then it turns 180 degrees to the right. Repeating behavior Computer programs use repeat loops that carry out the set of instructions with in the loop a number of times. Here s what the Repeat block looks like:

5 To use it, we need to insert how many times to repeat and tell the computer the statements to repeat. For example, makes the turtle take one step backwards ten times in a row. In TNG often we want the turtles to do something continuously, or forever, until turned off. For this we use the control structure that looks like this brown block. When the switch in the bottom left corner is flipped (by clicking upon it), the blocks that are being executed are highlighted in green. Can you guess what this block does? User-driven events In creating a game, user input is vital. StarLogo TNG has an easy way to check for userdriven events. Using the event control blocks, we can check to see if the player is pressing a key or an arrow button. To react to the user event, such as the user pressing a key or an arrow button, we use a conditional statement that tests whether the key or arrow is being pressed and reacts with an action.

6 What do you think happens when the up arrow gets pressed? NEW SECTION: Explorations with a model of Fish and Plankton The TNG Interface: Explore the canvas, the drawers, the blocks, spaceland, the cameras, the terrain editor. Describe the organization of the drawers and the difference between the factory and my blocks. Open the Fish and Plankton model Walk through the code. Describe what is going on in each block area. Ask students what they think different pieces of code are doing Point out where simple variables, initial # of fish and plankton, can be changed. Point out that fish have a variable called energy, How do you think fish gain energy? How do you think fish lose energy? Before running the model, ask for predictions, what will this simple code do? Run the model. Ask students for predictions, Show how easy it is to restart model with new parameters. Questions: If I run the model again with the same parameters, will the same thing happen? Why do we need to run a model over and over? What can we learn from a model? How do we run experiments with a model? Give students time to play with model and ask for some observations. NEW SECTION: Create the Litterbug model together.