So far everything we've done has been just
in the interface tab

and the problem is that when we write all
those commands in the command centre

and then we close NetLogo, as you may have
found out, when you come back in

it's no longer there and you no longer
have those commands.

So what we need to do is start editing in
the <i>code</i> tab,

which is where we can then save the
results into a model file

so the code that we're writing will be
preserved...

for when we want to modify it again in the
future.

So let's click on the <i>code tab</i>.

The code tab is where, as you might
expect, we actually write the code.

What we do is define <i>procedures</i>.

Procedures are like new commands for the
NetLogo model.

There are two procedures that we almost
always define right off the bat,

they are the same as the two buttons you
see in almost every NetLogo model,

<i>setup</i> and <i>go</i>.

You define a procedure by starting with
the word 'to'

and then you write the name of the
procedure (<i>'to setup'</i>)

and then when you're done with the
procedure you write the word 'end'.

So I'm going to go ahead and put stubs in
for our <i>setup</i> and <i>go</i> procedures.

And as you can see, NetLogo automatically
indents you a couple of spaces,

this is standard NetLogo style to indent a
little bit each procedure

or sub-piece of code. That way it makes it
easier to see when the format is changing.

One of the first things you always write
in the <i>setup</i> procedure...

is the 'clear-all' command, or you can 
abbreviate it to 'ca' if you want,

and what that does is destroy the model
state as we talked about,

and start the model all over again.

And then, once you have the model state
changed back to the initial state,

now I can go ahead and start writing my
model, because I know the world is blank.

So I can do that, for instance, by
creating a group of new turtles,

'crt 100'

and one thing that we could do within
that, that we haven't talked about before,

is immediately ask those turtles to do
something.

So I can put these square brackets right
after the 'create-turtles' command

and that's as if you're putting an
implicit 'ask' to the turtles I just created.

So, for instance, I could ask them to set
their xy coordinates to random values

'setxy random-xcor random-ycor'

I could also ask them to change their
color, for instance, to blue

'set color blue'

and I can have them change their shape.
We haven't talked about this yet,

but NetLogo comes with a wide variety of
shapes that you can use for your turtles,

the standard triangle you've been seeing
is the default shape,

but if you go up to the 'tools' menu, and
then go down to 'turtle shapes' editor

you can see a bunch of other shapes that
exist within the system.

One that I often use is the 'person'
shape, so you can set the shape to 'person'

'set shape "person"'

And now that I've written this code, I can
hit the 'check' tab

and what the 'check' tab will do is check
to see if there are any syntax errors.

Right now there are not... but let's say I
forgot the hyphen in 'random-xcor',

it would tell me that the command,
'RANDOMXCOR', doesn't exist,

and then I might be 'oh yes, there needs
to be a hyphen there',

and once I put it back in, it's OK.

So that allows me now to create my basic
first command,

and if I go back to the interface tab now
I can type 'setup'...

and there you go - I have 100 little
people scattered throughout the world.

Now unlike the triangles shapes, the
<i>people</i> do not have a particular direction,

so they're always facing up, even though
they <i>are</i> actually facing different directions

it's just the objects themselves are what
NetLogo calls <i>non-rotatable</i>,

they just stay upright in this particular
context.

That could be fine, it's up to you as to
what kind of shapes you want,

if you wanted to <i>change</i> that for some reason,
you can go to the turtle shapes editor,

go down to "person"

hit 'edit', and make them rotatable.

That will allow them to take on a
particular heading.

But let's leave it as it is for now.

So we have a <i>setup</i> procedure, let's write
our <i>go</i> procedure,

and we'll just write a simple command, ask
turtles forward 5 right random 90 degrees

'ask turtles [ fd 5 rt random 90 ]'

and now, if we go back to the interface
tab, we can type 'setup'...

and then 'go', 'go', 'go', and we can see
them moving.

If you want to see them moving a little
bit, you can run the speed slider down...

and you can see them actually moving
around a little bit more.

Now once we've created these two commands,
<i>setup</i> and <i>go</i>

we probably want to interact with them
more directly through the interface tab,

rather than typing 'setup' and 'go' all
the time,

and we can do that by creating two new
buttons in the interface tab,

which we talked about before.

When we create a button we have to specify
what command it is going to execute.

So in this case we want one button that is
going to execute <i>setup</i>,

and we want one button that executes <i>go</i>.

So now if we hit <i>setup</i> and <i>go</i> we keep
running the commands.

Now one small note. You can right-click or
control-click to pull up 'edit' on a button,

I could give the button a different <i>display
name</i> from the command that's executed,

so this could be 'My cool go', and that's
what it will say,

even though it's still executing the
command 'go'.

If you don't put anything for the <i>display
name</i> it will just use the command as the name.

There's also a button here called <i>Forever</i>

What <i>Forever</i> does is say 'keep executing
this command until I push the button again'

So if I hit <i>setup</i> now, and then I hit <i>go</i>
you'll see it just keeps running,

and then when I hit <i>go</i> again it stops.

At this point you probably want to <i>save</i>
your model,

so go up to the 'file' menu and pick
'save',

and then you can give your model a name.

So I'm going to go to my desktop, and call
the model 'my-first-model.nlogo'.

And you might also want to start updating
the info tab at the same time,

to explain what going on,

so this is 'my first model to explore
Netlogo'.

The reason why you want to update the info
tab is because you want to make sure...

that you always have documentation
that matches up with the model.

One other small note, as you continue to
do this,

you might start to do things like 'my
first model to explore NetLogo version 1'

and then when you make some major changes

you might want to save the new model as
'my-first-model-1' or '...-2',

this allows you to keep old versions that
you have around,

of course you could also use a more
sophisticated version control system

like svn or git or something like that as
well.

That's a basic introduction to how to
write code in NetLogo.

In the next talk we're going to discuss how
to define <i>properties</i> within the code tab

that you can then use within your model.