• none

• none

• Setup your computer for the semester
• Learn how to take screenshots.
• Introduce command line control of your computer.
  • Meet the Terminal
    • Put it in your dock
    • Customize it
• Learn basic commands and terminology for using the Terminal in Mac OS X.
• Introduce the MATLAB programming environment.
• Introduce basic MATLAB programs and programming syntax.

• Mac OS X Terminal
• MATLAB
• BBEdit

Let's make sure the software we'll use most often is easily available by putting shortcuts in our dock (the row of applications at the bottom of your screen). For now, don't worry about knowing what these programs are or what they do.

1. Confirm that you have the icons for MATLAB and BBEdit in your dock. Let me know if you can't find one or both.

2. Put a shortcut to the Terminal app into your dock.
2a. Find the application by going to Applications –> Utilities. You should see the Terminal application.

2b. Drag and drop the icon onto your dock.

Scientific computing is often a frustrating uphill slog punctuated by moments of epiphany and success. It's sort of like fighting a losing war, just hoping to win some battles along the way. This is particularly true for beginners, but can also be true for intermediate and advanced users because, regardless of skill level, we are all generally operating at the edge of our knowledge and pushing the boundaries of what we can achieve. I am sharing this perspective (admittedly, exaggerated for comic effect) of scientific computing because one of the skills I want you to acquire in this course is problem solving. It is tempting to think that problem-solving in science refers to problems with a capital “P”; big ideas, big questions, and big theories. But the fact of the matter is that the daily work of science is most often about problem-solving the “small” practical problems that stand as barriers to achieving our larger goals. The first step in technical problem solving is often the same first step you use to get the phone number of the local Chinese restaurant; Google.

To that end, here are some tips for self-sufficiency. Also, there are tons of useful OS X Terminal and Unix tutorials online. I've selected a few here.

The Mac operating system is a Unix-based graphical user interface. You can therefore interact with your Mac computer using many (most) of the same commands used in Unix/Linux systems. Many of these commands achieve the same things that you can achieve using 'point and click' methods.

We “talk” to the computer by typing instructions on the command-line in the Terminal.app.

Most things we want the computer to do can be achieved several ways. For example, let's say we want to create a new directory on our desktop.

On our Mac, we can create a new folder on our desktop by clicking on the desktop, selecting File New Folder. Alternatively, you could right-click on the desktop and select New Folder.

But you can also create a new folder from the command-line. Let's give it a shot.

1. Open Terminal. You should see a window like the one below. (the colors of yours will look different, but we'll come back to that in a bit)

2. Type the following command and then press return

cd ~/Desktop

3. Type the following command and then press return

mkdir ./mynewfolder

Voila! You should see a new folder on your Desktop named “mynewfolder”. What just happened?

• cd ~/Desktop told the computer to change your directory to the Desktop (we'll talk about the ~/ part in a bit.
• mkdir tells the computer to make a directory.
  • ./ is telling the computer that you want that new folder right here.
  • mynewfolder is telling the computer what to name the new folder

At this point you are undoubtedly thinking “It was way easier to just use the mouse/trackpad and point-and-click, so why would I ever want to use the command-line!?” Well …

• the command-line gives you direct control over many aspects of your computer that would be hard (or impossible) to access via the graphical interface.
• we can write our commands into a text file (called a “script”) and then tell the computer to read the commands from that file. This is great because it means that we can give the computer lots of step-by-step instructions.
• if we need to run the task 100 times (e.g., we want to create 100 new folders), we can just tell the computer to read the script 100 times. That sounds a lot easier than right-clicking and selecting New Folder 100 times.
• among other great things, this ensures that whatever task we've “scripted” is being done that we're exactly the same way every time we do it. This might be kind of important when we're, say, analyzing data from many subjects.

The majority of what we do in this course will rely on controlling your computer using the command-line (either in Terminal or Matlab (a program that will be introduced later in this lab). You should therefore dedicate a sufficient amount of time to getting comfortable with basic Terminal/Unix commands and behaviors. We will spend some time tonight learning a few basic commands. A more comprehensive list of commands can be found here.

Let's get started learning a bit about the Terminal!

If you're so inclined, you can change the way your Terminal looks. In that same settings window, click on Profiles to see many different default layouts and color patterns. You can learn more about how to customize it here.

File paths are simply the directions to a particular file or directory on your computer. You already have an intuitive understanding of these paths, though you might not have ever referred to them as such. Most likely, you think of your file system in terms of “folders” and “sub-folders” arranged in a hierarchy (e.g., my term paper is inside the folder named “psyc class” which is inside the folder “kenyon stuff”). The path is simply directions for how to navigate this folder hierarchy.

1. Open your Mac Finder by clicking on the finder icon on your dock.

2. Navigate to your Downloads directory by double clicking on it. You should now see the contents of the Downloads directory. I assume that this type of interaction with the file-system is very familiar to all of you.

3. To see the filepath for this location select View –> Show Path Bar (if you only see an option for Hide Path Bar then your computer is already set up the way we'd like it to be and you can skip this and continue reading.)

You should now see a list of folders displayed at the bottom of the window. This is the filepath! If we wanted to use the command line to refer to this location we would type /Users/hnl/Downloads

4. Let's try getting to that same location using the command line. Open your Terminal program by selecting the Terminal icon on the doc.

5. In the terminal window type the following command followed by return (from now on, pressing return after each line is always implied).

pwd

pwd tells the system to “print working directory”. You should see /Users/hnl, which is your default home directory.

6. Before we leave this location, let's see what directories and files are in our current location.

ls

ls tells the system to list everything in the directory.

7. So we know where we are and what's in this directory, but now we need to get into the Downloads directory. To do that we need to use the cd command (change directory). The command below tells our system to change the directory to the Downloads folder using the full path. Go ahead and enter the command below into your terminal window.

cd /Users/hnl/Downloads

8. Finally, make sure that you've successfully navigated to the Downloads directory by using the pwd command.

An absolute path is a full set of instructions. It tells the computer how to get someplace starting from the very beginning (i.e., the top of your directory hierarchy). Let's say your parents were coming to campus and planned to meet you in the Crawford Computer lab, but wanted to check out the sculpture in the science quad along the way. You would give them directions from their house to the science quad.

So far, so good. But what directions would you give them to get from the science quad to the computer lab? You could give them directions from their home to the computer lab. This is similar to using an absolute path. You ignore the current location and it's proximity to the desired location and simply give the entire path.

However, I doubt that any of you would actually do this. Instead, you would tell your parents “when you're done checking out the sculpture, come in the back entrance of Sam Mather, come up the stairs, etc.”. In other words, you would tell them how to get to their desired location from their current location. This is a relative path.

When we are using the command line to navigate our directory hierarchy .. means 'the directory above'. So what do you think the following code would do?

cd ..

Correct! It would move us up one level in our directory hierarchy. If we were in /Users/hnl/Desktop it would move us to /Users/hnl

And if we wanted to move up two levels we could use

cd ../..

If we were in /Users/hnl/Desktop it would move us to /Users

So now let's think about this in terms of relative paths.

9. Let's first go back to our home directory using what we've learned. We can type

cd /Users/hnl

or simply type

cd

or simply type

cd ~

10. We know we can get to the Downloads folder from our current location by using the absolute path and typing cd /Users/hnl/Downloads that we used in Step #7. But let's use a relative path instead.

If we do not specify a path, the computer will assume we are starting from wherever we currently reside. Therefore, the following command will work because computer assumes we are starting from our current location, which happens to be /User/hnl, which is one level up from Downloads.

# change to the Downloads directory
cd Downloads

The following command would have worked exactly the same, but it is more explicit that we are starting from our current location. Whereas .. refers to the directory above our location, . refers to our current directory.

# Change to the Downloads directory
cd ./Downloads

cd ~
cd Downloads
cd ./Downloads

11. So now we are in /Users/hnl/Downloads. What command should we use if we want to navigate to /Users/hnl/Documents. You now know we've got two options.

# Absolute path
cd /Users/hnl/Documents
 
# Relative path
# The Documents directory is parallel to the Downloads directory in our directory hierarchy. So to get there, we
# need to go up one level, then down into Documents.
cd ../Documents

You've already learned a couple of important terminal commands; cd, pwd, and ls. In this section we'll learn a few more that can be used to create and delete, move, copy, and rename files and directories. And remember you can find a list of the most used commands here.

Over the next series of steps we'll create a new folder and some new files and then play with moving, copying, renaming, and deleting them.

1. To create a new directory we use the mkdir command (make directory). Let's use that command to create a new directory named “foo” on our Desktop.

mkdir /Users/hnl/Desktop/foo

You should now see a folder named foo on your Mac Desktop. Kinda cool, no?

2. Delete the folder you just created. To do so we'll use the rmdir command (remove directory).

rmdir /Users/hnl/Desktop/foo

Voila. Keep in mind that rmdir will only work if the directory is empty.

3. Now we'll recreate our “foo” directory and also create a sub-directory named “subfoo” inside of “foo”. Try this

mkdir /Users/hnl/Desktop/foo/subfoo

Oops! That doesn't work and we get an error. This is because the “foo”directory does not exist, so the computer is confused about where to put the “subfoo” directory. Does this mean we have to create the “foo” directory and then create the “subfoo” directory? Not at all. We just need to give a little extra info at the command line. Try this

mkdir -p /Users/hnl/Desktop/foo/subfoo

Success! The -p part of our code is called a flag. It is an option that modifies the mkdir command. It tells the computer to create the directory “subfoo” and create any other directories in the path that do not yet exist.

4. Now that I think about it, “subfoo” is not a great name. I think I'd prefer to name that directory “babyfoo”. To rename files or directories we use the mv command (move). Not surprisingly, we also use this command to move files and directories. But here we'll use it to rename our “subfoo” directory.

mv /Users/hnl/Desktop/foo/subfoo /Users/hnl/Desktop/foo/babyfoo

5. Now we'll create some files inside of our directories using the touch command. The following code will create a couple of text files in the babyfoo directory and a Word document file in the foo directory. After running the code, see for yourself that the files were created. You can show off and use the command line, or just double-click the folder on your Desktop.

touch /Users/hnl/Desktop/foo/babyfoo/file1.txt
touch /Users/hnl/Desktop/foo/babyfoo/file2.txt
touch /Users/hnl/Desktop/foo/file3.doc

6. We've already seen mv used to rename a directory. Here we'll use it to move the file3.doc from the foo directory to the babyfoo directory.

mv /Users/hnl/Desktop/foo/file3.doc /Users/hnl/Desktop/foo/babyfoo/file3.doc

7. Now all three files are in the babyfoo directory. D'oh! I actually need to have file3.doc in both directories. So let's use the cp command (copy) to make a copy of the file in the foo directory.

cp /Users/hnl/Desktop/foo/babyfoo/file3.doc /Users/hnl/Desktop/foo/file3.doc

8. Actually, it'd be best if there were copies of all three files in both directories. We could use the cp command to copy each file, but let's be a little more efficient. We'll use the wildcard symbol * to make a copy of all files that end with .txt.

cp /Users/hnl/Desktop/foo/babyfoo/*.txt /Users/hnl/Desktop/foo

9. Okay okay okay, I now realize that I don't need the file3.doc file in the foo directory after all. We've already seen that rmdir will remove (i.e. delete) a directory, but here we'll use rm to remove a file.

rm /Users/hnl/Desktop/foo/file3.doc

So far we've used the command line interactively. That is, we wrote a command, the computer carried out our command, we wrote another command, the computer carried out that command, and so on. This, however, is not the only way to have the computer carry out our marching orders. We can instead write out all of the things that we want the computer to do and save them into a shell script.

In the last section we created, renamed, moved, copied, and deleted some files and directories in a long series of interactive steps. The script below would accomplish the same thing without us needing to write each command interactively. Instead the computer would simply read our list of commands and carry out each one in order.

1. Delete the “foo” directory from your Desktop (you can test your command line skills or just drag and drop it into the trash).

2. Download the following script to your computer by clicking on the tab that reads “myFirstScript.sh”.

myFirstScript.sh

#!/bin/zsh
 
# create the foo directory
mkdir /Users/engella/Desktop/foo
 
# remove the directory we just created
rmdir /Users/engella/Desktop/foo
 
# Do not run this next line of code
# mkdir /Users/engella/Desktop/foo/subfoo
 
# make the foo directory and subfoo sub-directory
mkdir -p /Users/engella/Desktop/foo/subfoo
 
# rename subfoo to babyfoo
mv /Users/engella/Desktop/foo/subfoo /Users/engella/Desktop/foo/babyfoo
 
# create two text files in the babyfoo directory
touch /Users/engella/Desktop/foo/babyfoo/file1.txt
touch /Users/engella/Desktop/foo/babyfoo/file2.txt
 
# create on Word doc in the foo directory
touch /Users/engella/Desktop/foo/file3.doc
 
# move the Word doc from the foo to the babyfoo directory
mv /Users/engella/Desktop/foo/file3.doc /Users/engella/Desktop/foo/babyfoo/file3.doc
 
# copy the Word doc in the babyfoo directory into the foo directory
cp /Users/engella/Desktop/foo/babyfoo/file3.doc /Users/engella/Desktop/foo/file3.doc
 
# delete the Word doc from the foo directory
rm /Users/engella/Desktop/foo/file3.doc

3. Navigate to your Downloads directory in your Terminal and run the script using the following command.

source ./myFirstScript.sh

This should have carried out all of the steps we previously did interactively. Obviously, there's not much value in this series of silly steps. But scripting can be an extremely powerful tool that makes repeating tasks easy, reliable, and leaves a document of what has been done.

MATLAB is a multi-platform (that means it runs on Linux/Mac/PC) programming language that is designed to excel at numerical computing, particularly numerical matrix manipulation (indeed, it's name means matrix laboratory). It is therefore the preferred language for many scientific, engineering, and economics applications. Many of the EEG and fMRI programs and applications that we will use throughout the semester are written in MATLAB.

To open MATLAB click on the icon in your dock.

The window that opens will look like the one below (without the shaded boxes). Note that there are several sub-windows.

• The PURPLE shaded area shows you the path to your current directory.
• The RED shaded area is your command window. This is similar to the command line that you used in Terminal in the previous section.
• The GREEN shaded area is your workspace.
  • All of the defined variables are shown in this window (it's empty because we haven't yet defined any variables).

Some of the path navigation commands you learned in the Terminal will also work in the MATLAB command window. Try using cd, pwd, and ls. Other file management commands will not work. For instance, if you wanted to create a text file named “testfile.txt” and typed touch testfile.txt in the MATALAB command window you would get an error (whereas typing this in the terminal would successfully create the file.)

% The code below will not work because the Terminal file management commands don't work in MATLAB
touch /Users/hnl/Desktop/testfile.txt
 
% The code below will work because putting the ! at the start of the command tells MATLAB 
% to use the bash shell (i.e. Terminal) to carry out the command rather than MATLAB
! touch /Users/hnl/Desktop/testfile.txt

As with Unix/zsh/OSX Terminal, there are many(!) MATLAB tutorials available online. Let's take a look at one offered by MathWorks (the company that makes MATLAB)

1. Go to this MATLAB page.

2. Select the MATLAB Onramp class. (you'll probably need to enter your login credentials again)

3. Complete the Commands and MATLAB Desktop and Editor modules.

• These two are the only modules you are required to complete, but I strongly recommend that you continue to use this, or other resources, to strengthen your MATLAB skills.

4. When you've completed both modules take a screenshot of your progress.

• click the hamburger icon (blue box with three blue stripes) in the upper left corner.
• capture the entire screen or, at least, the portion that shows your progress and your name. (see example below)
• If you've already closed the window, you can just log back into the course to access your progress.

I would like you to watch a couple of 'Introduction to Matlab' videos before our next class. I also strongly recommend that you spend some time going through some online tutorials on your own time. For the most part though, I expect that you will learn MATLAB as we use it.

Please watch the following videos.