Scott,

Here's a project file that demonstrates the problem. I'm working in the most recent ver. of Squeak from www.squeak.org (just downloaded a few days ago) .... 3.9-final-7067

When I activate the script with the object controlled by a joystick, the first few times it hits the top of the world and it's silent - all is well. But at some point, after 3 or 4 or 5 collisions, it's no longer silent.

All help much appreciated and thanks very much in advance for looking at the project (made by a student in the class I'm teaching).

Ab,

Mark

Scott Wallace wrote:

Hi, polishookm,

I cannot reproduce the phenomenon you describe in any image I have close to hand -- not in Squeak 3.9, not in Squeakland3.8-05, and not in the OLPC etoys image. (Can anyone else?)

What version of Squeak are you using? If it's 3.10, I'll leave it for other 3.10 users to track this down, since I don't have a copy of 3.10 handy.

Otherwise, perhaps you could email me a copy of a project of yours which exhibits this symptom...

Cheers,

-- Scott

On Feb 20, 2008, at 6:38 PM, polishookm wrote:

...

I"m finding that with

[Rectangle forward by 5] [Rectangle bounce silence]

the rectangle eventually collides with the top of the world and bounces off in the opposite direction - with no sound ... the bounce is silent.

However, this script

[Rectangle forward by JoyStick's upDown * 8] [Rectangle bounce silence]

doesn't allow the Rectangle bounce silently off the edge of the world. In this case, when the rectangle reaches the top of the world, it makes a bouncing sound. The exact sound can be selected in the [Rectangle bounce silence] tile. But selecting 'silence' as the sound doesn't seem to work.

Am I missing something obvious? Shouldn't it be possible to silence the bounce?

---

Beginners mailing list Beginners@lists.squeakfoundation.org http://lists.squeakfoundation.org/mailman/listinfo/beginners

Hi Herbert,

Thanks for trying the project. It's a first attempt (by a student) to connect behavior of two objects in this case, a paddle and a joystick mechanism. Like all first attempts, it will likely morph and correct and revise and clarify itself into something else as the student acquires more familiarity with tile scripting.

Ab,

Mark

I'm confused by the fact, that the reaction of the paddle to the joystick reverses after it has bounced.

It even starts reverse when I drag the project onto a fresh 3.9 final.

No experience with etoys but curious.

Cheers,

Herbert mailto:herbertkoenig@gmx.net

---

Beginners mailing list Beginners@lists.squeakfoundation.org http://lists.squeakfoundation.org/mailman/listinfo/beginners

Hi Herbert,

Herbert König wrote:

Hi Mark,

p> mechanism. Like all first attempts, it will likely morph and correct and p> revise and clarify itself into something else as the student acquires p> more familiarity with tile scripting.

doesn't he know the rule "Never do anything for the first time" :-))

Well stated :) .... actually, perfectly stated :)

Actually I couldn't find a fault in the script (remember I never used etoys myself) so I think this is more something counter intuitive in etoys than a problem of your student.

That's an interesting observation. Did you, perchance, play with it to see what a more intuitive (for you) solution would look like? I had the same thought when I first examined the project. But because the student was satisfied with the joystick but not at all with the sound, the (the sound) became the issue that required a solution.

All suggestions welcome ...

Cheers

Herbert mailto:herbertkoenig@gmx.net

"Bounce" is particularly unsuited to this example, because it reverses the direction the paddle is "heading" and hence reverses the motion brought about by "forward" commands. This makes it hard to retain an intuitive relationship between the joystick and the motion of the paddle.

FWIW here's a script giving good, or at least intuitive, joystick/ paddle action:

Cheers,

-- Scott

On Feb 23, 2008, at 5:30 AM, Herbert König wrote:

Hi Mark,

p> That's an interesting observation. Did you, perchance, play with it to p> see what a more intuitive (for you) solution would look like? I had the

thanks for asking, that made me try :-))

p> All suggestions welcome ...

Nearly everything i know about Etoys is from the Squeak video DVD. So I opened the viewer of the paddle and it has a tile "PongPaddleA's heading" which jumps between 0 and 180 when bouncing.

So the simple solution would be to add a "PongPaddleA's heading = 0" to the script. I thought I'd just have to drop the tile onto the script. But this doesn't work, the script doesn't accept the tile.

This reminded me of several posts on squeak dev that 3.9 is not good for etoys. So I guess the Thing to do is follow Scott's advice and use a Squeakland image.

Dropping the Project onto a 3.8 image doesn't work, this is also known.

So I redid it from Ground up in my 3.8 image using an arrow as the pong paddle (guess why). It made the sound but didn't turn on bounce. There I learned that one has to drag a tile by the arrow to get it into a script.

Then I tried this in 3.9 and in 3.9 the arrow also didn't turn and the the trick with fixing the heading of the sketch also worked.

So I proved my accusations on 3.9 unfounded, Sorry 3.9 :-))

Now drop the one who did (against better knowledge) something the first time and let the project leader speak up.

What I did with fixing the heading of the paddle in the script was "an ugly hack". The price to pay is more uglyness if we ever want to use "Paddle turn by Joysticks leftRigt" by having to simulate differential geometry by "absolute geometry".

So the way to go is find out what makes the Sketch (the inspector shows the paddle is a SketchMorph) misbehave on bounce. Maybe there is a reason and then you should use a wide line as your paddle.

Finally:

• Attached is my changed 3.9 Project. --> Not to the list but private

mail

• Maybe this is off topic here, at least you might get better advice

on the squeakland mailing list (where I'm not subscribed).

• I'd be curious and willing to comment on further development of the

project, maybe by private mail unless somebody here also wants to follow this.

Cheers

Herbert mailto:herbertkoenig@gmx.net

---

Beginners mailing list Beginners@lists.squeakfoundation.org http://lists.squeakfoundation.org/mailman/listinfo/beginners

Scott, thanks very much for both solutions (and that great explanation about the fence).

The solution you've proposed here is pretty much what the student began with. Except his scaling factor was slightly higher.

I'll see if I can post the project once it's finished.

Ab,

Mark

Scott Wallace wrote:

And here's a very different, and perhaps nicer, one-line solution.

---

Cheers,

-- Scott

On Feb 23, 2008, at 2:10 PM, Scott Wallace wrote:

...

"Bounce" is particularly unsuited to this example, because it reverses the direction the paddle is "heading" and hence reverses the motion brought about by "forward" commands. This makes it hard to retain an intuitive relationship between the joystick and the motion of the paddle.

FWIW here's a script giving good, or at least intuitive, joystick/paddle action:

<positionPaddle.jpg>

Cheers,

-- Scott

On Feb 23, 2008, at 5:30 AM, Herbert König wrote:

...

Hi Mark,

p> That's an interesting observation. Did you, perchance, play with it to p> see what a more intuitive (for you) solution would look like? I had the

thanks for asking, that made me try :-))

p> All suggestions welcome ...

Nearly everything i know about Etoys is from the Squeak video DVD. So I opened the viewer of the paddle and it has a tile "PongPaddleA's heading" which jumps between 0 and 180 when bouncing.

So the simple solution would be to add a "PongPaddleA's heading = 0" to the script. I thought I'd just have to drop the tile onto the script. But this doesn't work, the script doesn't accept the tile.

This reminded me of several posts on squeak dev that 3.9 is not good for etoys. So I guess the Thing to do is follow Scott's advice and use a Squeakland image.

Dropping the Project onto a 3.8 image doesn't work, this is also known.

So I redid it from Ground up in my 3.8 image using an arrow as the pong paddle (guess why). It made the sound but didn't turn on bounce. There I learned that one has to drag a tile by the arrow to get it into a script.

Then I tried this in 3.9 and in 3.9 the arrow also didn't turn and the the trick with fixing the heading of the sketch also worked.

So I proved my accusations on 3.9 unfounded, Sorry 3.9 :-))

Now drop the one who did (against better knowledge) something the first time and let the project leader speak up.

What I did with fixing the heading of the paddle in the script was "an ugly hack". The price to pay is more uglyness if we ever want to use "Paddle turn by Joysticks leftRigt" by having to simulate differential geometry by "absolute geometry".

So the way to go is find out what makes the Sketch (the inspector shows the paddle is a SketchMorph) misbehave on bounce. Maybe there is a reason and then you should use a wide line as your paddle.

Finally:

• Attached is my changed 3.9 Project. --> Not to the list but private

mail

• Maybe this is off topic here, at least you might get better advice

on the squeakland mailing list (where I'm not subscribed).

• I'd be curious and willing to comment on further development of the

project, maybe by private mail unless somebody here also wants to follow this.

Cheers

Herbert mailto:herbertkoenig@gmx.net

---

Beginners mailing list Beginners@lists.squeakfoundation.org http://lists.squeakfoundation.org/mailman/listinfo/beginners

---

Beginners mailing list Beginners@lists.squeakfoundation.org http://lists.squeakfoundation.org/mailman/listinfo/beginners

---

---

Beginners mailing list Beginners@lists.squeakfoundation.org http://lists.squeakfoundation.org/mailman/listinfo/beginners

Hi Scott,

Scott Wallace wrote:

Of course it's the "fence" that allows this second solution to work. If your student didn't know about the "fence", I suppose he/she felt an urge to use "bounce" in an attempt to contend with the situation at the boundary, and that led to the complications we've seen.

This basically describe the situation. Is there a place somewhere where the fence is described (other than an email thread)? I know (now) that the balloon pops up from the playfield option in the World menu ....

This is probably a case where trying things out very early in the authoring process would have been helpful. Trying out the simple "forward" script would have revealed that it already worked, with the only task remaining being to experiment with different values for the scaling factor to get the best "feel," and hence the siren song of "bounce" might have been avoided.

I agree with you on this. But at the same time, in the early process of working with Squeak/E-toys - or really almost any development environment, there are significant forces other than logical progression/procedure that come into play. Having said that, what turned out to be helpful in class was discussion that ensued about the fence issue. For example: (1) how quickly an explanation for the fence was available on the list, (2) how flexible features in a development environment are very different from fixed menu choices in an application (3) how close to the surface the solution actually was, eg, a menu choice in playfield, (4) soon to surface shortly - the difference between Squeak and Squeakland, (5), and discussion of the tradeoffs that features sometimes bring.

.... One more question that seems to have come up. The student found today that saving the project in 3.9 (as a PR) doesn't seem to be reliable. At least that's the report I have. The symptom specifically, is some of the tiles that trigger on and off scripts don't seem to work after being saved into a PR file (and then being reloaded). I have no idea at the moment is this is the inexperience of someone new to Squeak or if its something else. From the students' experience, saving the project to a PR file is disabling some scripts whereas saving it in an image seems to be just fine.

I'm going to follow up on this and see what I can learn ....

Ab,

Mark

Cheers

-- Scott

On Feb 23, 2008, at 4:56 PM, polishookm wrote:

...

Scott, thanks very much for both solutions (and that great explanation about the fence).

The solution you've proposed here is pretty much what the student began with. Except his scaling factor was slightly higher.

I'll see if I can post the project once it's finished.

Ab,

Mark

Scott Wallace wrote:

...

And here's a very different, and perhaps nicer, one-line solution.

---

Cheers,

-- Scott

On Feb 23, 2008, at 2:10 PM, Scott Wallace wrote:

...

"Bounce" is particularly unsuited to this example, because it reverses the direction the paddle is "heading" and hence reverses the motion brought about by "forward" commands. This makes it hard to retain an intuitive relationship between the joystick and the motion of the paddle.

FWIW here's a script giving good, or at least intuitive, joystick/paddle action:

<positionPaddle.jpg>

Cheers,

-- Scott

On Feb 23, 2008, at 5:30 AM, Herbert König wrote:

...

Hi Mark,

p> That's an interesting observation. Did you, perchance, play with it to p> see what a more intuitive (for you) solution would look like? I had the

thanks for asking, that made me try :-))

p> All suggestions welcome ...

Nearly everything i know about Etoys is from the Squeak video DVD. So I opened the viewer of the paddle and it has a tile "PongPaddleA's heading" which jumps between 0 and 180 when bouncing.

So the simple solution would be to add a "PongPaddleA's heading = 0" to the script. I thought I'd just have to drop the tile onto the script. But this doesn't work, the script doesn't accept the tile.

This reminded me of several posts on squeak dev that 3.9 is not good for etoys. So I guess the Thing to do is follow Scott's advice and use a Squeakland image.

Dropping the Project onto a 3.8 image doesn't work, this is also known.

So I redid it from Ground up in my 3.8 image using an arrow as the pong paddle (guess why). It made the sound but didn't turn on bounce. There I learned that one has to drag a tile by the arrow to get it into a script.

Then I tried this in 3.9 and in 3.9 the arrow also didn't turn and the the trick with fixing the heading of the sketch also worked.

So I proved my accusations on 3.9 unfounded, Sorry 3.9 :-))

Now drop the one who did (against better knowledge) something the first time and let the project leader speak up.

What I did with fixing the heading of the paddle in the script was "an ugly hack". The price to pay is more uglyness if we ever want to use "Paddle turn by Joysticks leftRigt" by having to simulate differential geometry by "absolute geometry".

So the way to go is find out what makes the Sketch (the inspector shows the paddle is a SketchMorph) misbehave on bounce. Maybe there is a reason and then you should use a wide line as your paddle.

Finally:

• Attached is my changed 3.9 Project. --> Not to the list but private

mail

• Maybe this is off topic here, at least you might get better advice

on the squeakland mailing list (where I'm not subscribed).

• I'd be curious and willing to comment on further development of the

project, maybe by private mail unless somebody here also wants to follow this.

Cheers

Herbert mailto:herbertkoenig@gmx.net

---

Beginners mailing list Beginners@lists.squeakfoundation.org http://lists.squeakfoundation.org/mailman/listinfo/beginners

---

Beginners mailing list Beginners@lists.squeakfoundation.org http://lists.squeakfoundation.org/mailman/listinfo/beginners

---

---

Beginners mailing list Beginners@lists.squeakfoundation.org http://lists.squeakfoundation.org/mailman/listinfo/beginners

-- Mark Polishook, D.M.A. Technology Coordinator Morehead Hall, Rm 132 Center for Academic Advising and Adult Learning Montclair State University Montclair, NJ 07043

(973)-655-7114 http://www.montclair.edu/AcademicAdvising/html polishookm@mail.montclair.edu

---

Beginners mailing list Beginners@lists.squeakfoundation.org http://lists.squeakfoundation.org/mailman/listinfo/beginners

---

Beginners mailing list Beginners@lists.squeakfoundation.org http://lists.squeakfoundation.org/mailman/listinfo/beginners

Scott Wallace wrote:

On Feb 23, 2008, at 11:03 PM, polishookm wrote:

...

.... One more question that seems to have come up. The student found today that saving the project in 3.9 (as a PR) doesn't seem to be reliable.

maintained, i.e. Squeakland and olpc/etoys. It would be interesting to hear if your students encounter the same problems using one of those systems -- if so, please send along a project illustrating the problem.

Yes, I will follow up on this.

Finally, one last time, if you want a modern, up-to-date, documented, *supported* etoys experience, you'd be well advised to use the olpc/etoys squeak image (or the forthcoming Squeakland release derived from that).

Ok. Thanks for your persistence. Last night I was exploring the Squeakland image. A few minutes ago, I downloaded

http://etoys.laptop.org/src/etoys-image-and-pr.zip

and, just to be sure, this is what you're suggesting that I work with, yes?

Or should I be doing this?

Building etoys.image from developer's image

Developer's image (etoys-dev.image) is included in the svn repository http://etoys.laptop.org/svn/trunk/etoys/

* Press 'load code updates' button. * Save the image.

* Place project files for 'Gallery', 'Tutorials', and 'EtoysActivity' in the image directory. * Set screen size to 1200@900 ("Display inspect" helps).

* Save as 'etoys.image'. * And execute:

ReleaseBuilderSqueakland new prepareReleaseImageForOLPC.

Thanks (again) in advance.

There are two mailing lists devoted to uses of Squeak in education -- see:

http://squeakland.org/mailman/listinfo/squeakland

http://lists.laptop.org/listinfo/etoys

Thanks I joined squeakland last night. Will do the same for etoys later today ....

Cheers,

-- Scott _______________________________________________ Beginners mailing list Beginners@lists.squeakfoundation.org http://lists.squeakfoundation.org/mailman/listinfo/beginners

On Feb 24, 2008, at 12:07 PM, polishookm wrote:

Just to simplify the question of my last email - and this is really a generic question that maybe anyone on the list can answer -

...

Finally, one last time, if you want a modern, up-to-date, documented, *supported* etoys experience, you'd be well advised to use the olpc/etoys squeak image (or the forthcoming Squeakland release derived from that).

Where exactly is the dev version?

There turned out to be a glitch in the condense-sources process used to create the etoys3.0 release, so at this point it's best to wait just a little while longer...

Cheers,

-- Scott

On Feb 23, 2008, at 5:30 AM, Herbert König wrote:

Nearly everything i know about Etoys is from the Squeak video DVD.

The "Squeak Video" is a splendid thing, but expecting to learn how to use etoys from it would be like expecting to learn how to drive a car by watching an elegant Mercedes Benz commercial...

Cheers,

-- Scott