Think about what you want the kids to get. Squeak can be an introduction to

programming, or a way to explore another subject such as math, or a way to

have fun making cool things, or a way to explore and experiment. Some of

each of these things can happen, but you and the kids will probably be

happier, and your activities will be more effective, if you know what you

want to accomplish and what you want the kids to accomplish.

One way of teaching that has worked well: at the beginning of a session do

a short demonstration for all the kids, showing them the activity before

you turn them loose to do it. That way if there are any unfamiliar skills

or concepts needed for success, you can introduce them while giving

everybody a feel for the activity.

Don't be shy about having them try something off the computer in order to

understand a concept. For example, kids were trying to figure out how to

get an object to move so it would draw a square. They had a surprising

amount of trouble with this until BJ asked everybody to watch while one kid

stood and walked a square on the floor. Then everybody had to think about

what instructions to give that kid if he didn't know how to walk a square.

They wrote the instructions on slips of paper and handed them to the walker

one by one. It was easy to see right away when there was a bug (as long as

the walker only followed the instructions and didn't use intuition!) This

activity seemed to help the kids immensely. Sometimes seemingly small leaps

of understanding--especially about abstractions like heading--are actually

quite large leaps for kids. Putting it into their bodies really helps.

Iit's good to have plenty of help available for the kids--especially at the

beginning. So that means you should have a small group or some assistants.

Any computer activity involves confusions and missteps, and Squeak is a

research system so there are even more possible confusions and blind alleys.

Working in pairs is very fruitful. Conversation and collaboration help to

make the learning more explicit, and of course the kids can help each other.

In thinking about the concepts that I needed to present my children

with, prior to teaching them a new idea or concept in Squeak, I think about

my math standards. In the case of the car scenario, I used what they had

learned about mean, mode, median and range, and how I could incorporate this

concept into what I wanted them to create in Squeak. I needed them to

understand the concept of random and random numbers and what a "variable"

was. This is some pretty sofisticated stuff and Squeak gave the children a

means to use some abstract concepts in a meaningful way

Scott Wallace and I recently taught 3 classes of 33 kids at a local elemantary

school's "Discovery Day". The first two classes were fifth graders, the last

class was sixth graders. We worked throught the "Drive-a-car" example.

Our experiences strongly supports John Steinmetz's observations of the

Open School classes. In particular:

Re:

>One way of teaching that has worked well: at the beginning of a session do

>a short demonstration for all the kids, showing them the activity before

>you turn them loose to do it. That way if there are any unfamiliar skills

>or concepts needed for success, you can introduce them while giving

>everybody a feel for the activity.

We actually taught one of the first two classes with an up-front demonstration

and one without it. Even though the up-front demo takes an extra five minutes

(out of 40 minutes), the class with the demo got further. We decided to

teach the final group with the demo and that class also got further. One

practical thing about an up-front lecture/demo: that's the only time you really

have the full attention of everyone in the class. After they start their projects,

some of them will always be distracted when you ask for their attention.

In fact, we asked them to not even start up Squeak until we'd finished the

initial demo and introduction, and that was a good idea.

Re:

>Iit's good to have plenty of help available for the kids--especially at the

>beginning. So that means you should have a small group or some assistants.

>Any computer activity involves confusions and missteps, and Squeak is a

>research system so there are even more possible confusions and blind alleys.

Scott and I were only two "teachers" for 33 kids who had never seen Squeak.

I thought it would be a chaos. Actually, it worked better than I expected, in

part because we only had 18 computers, so kids worked in pairs. That

meant you could help two kids at once, and often one of the two would

understand your suggestions quickly. In contrast, we recently taught 14

kids who had never seen Squeak at Disney's "Bring your child to work day"

and we had about seven teachers. In that situation progress was very fast,

because kids who were stuck got immediate attention. However, I don't

believe that many teachers is necessary. I thank that if Scott and I had just

one more assistant, it would have been optimal: one teacher for every six pairs.

(Actually, one might say it is the computer/teacher ratio that matters! You

want that ratio to be under six for maximum progress.)

Your original posting said there were six PC's in the lab. I think that's

about the max for a single teacher, but there should be no problem

with pairing up two kids per computer. If you do this, I'd limit it

to 10 kids on 5 computers, at least for your first time. That would

also leave one machine available as your "demo" machine. You also

said it would be open to kids from 5th-8th grade. We had

that same span for the "Bring your child to work day". In that case,

the eighth grader was noticably faster and more self-sufficient than

the youngest kid. If that happens in your class, you could recruit the

fastest kids as teachers.