Blocks' digital geometry

[moon]

Dear all,

I'm starting a project that is NOT about programming although it has to do with

The context is the following.

In our small country-side villages most of schools don't have enough computers to start teaching the basics for kinder (<10 y.o), which is since last year mandatory (included in the official programs)

It came to my mind that we can teach using puzzles i.e having piece of woods that have the same shape/colours as those of ai2 and use that to teach without having the kids facing screens for a long time and as a way to overcome the low-level of equipments. When a teacher considers that the solution is good enough, kids can move by small groups to implement their solution.

Of course there is a way to scan etc, but it's gonna take ages...

I don't know if any started anything like that.

Do you think it's possible to get the geometry of the blocks so that I can think about asking a voluntary parent to manufacture some.

Actually no way of asking for public money to buy any potentially existing product.

Any comment/suggestion is welcome.

M

[Chris Ward]

Hi moon

It's an interesting idea - but the cost of each set would most likely exceed the price of a computer + screen (or Tablet).

You could possibly get the costs down by not having jigsaw shapes, just laser printed magnetic blocks similar to those used in Industry for organising projects/workflows on metal whiteboards - but I fear that even that would be too costly.

If you can find materials and a way to make the blocks that is significantly cheaper than buying more computers, then your project would probably qualify for financial backing via Kickstarter. Here is a nice example:

https://www.kickstarter.com/projects/publiclab/the-community-microscope-kit

[moon]

Thx Chris,

You're right and this is why I'm basically considering volunteers parents (two are already happy to produce that for free, not only because they've got kids....)

On the top of this, as in, probably, many countries, parents are against having their kids front of screens for long time.

it's not "wood" against "computers" rather a mix that can satisfy the pedagogical need and the parents' concerns.

In other words, all is matter to get the kids learning and complying with the contradictory constraints from parents/admin etc.

In addition, for pedagogical reasons, I consider that it's better to not create an additional  gap between what the kids do to solve a problem and the "real" programming. We're talking about <10 y.o kids (starting at 6 y.o in real life), this is why I want it to be exactly the same shape. 

Thx so much for the idea about eligibility to kickstarter, it didn't came to my mind at first. 

M

[SteveJG]

Why make this difficult.  What you are discussing might be a useful tool.

Here is the Block documentation for components and  and Blocks  summarized.

• Built-in Blocks Documentation
• Component Documentation

Enlarge your PC display scrreen.

Print the blocks

Make copies of the blocks.

Paste them on cardboard

Cut out the Blocks

Some blocks have the ability to mutate (create additional puzzle slots); you will have to produce mutated images.

Before you do anything that proves to be tedious or impractical, try the simple solution and see whether it makes sense with students.

Make a fancy set (glue a small piece of magnetized material on the back of the BLOCKS.... now these blocks can be moved around on a metal sheet.

Let us know how this goes.

Also look at something called MakerCards by Per Mathijessen

Good luck.

Regards,

Steve

[SteveJG]

@ moon   If you do not already participate in this MIT Forum  http://teach.appinventor.mit.edu/  you might consider joining and post your idea/proposal there moon.

Posting here is OK where mostly technical issues are discussed, however educational issues are discussed at the above link.  Posting there might result in additional ideas/input from educators.

[moon]

Oups, Thought I did so !

Sorry about that.

M

[moon]

Hi Steve,

It's not about making things difficult. the idea is so simple that I thought may be someone has already the digital data that provides the geometry. Matter of saving time.

The cutting tools (used by the manufacturer happy to help me) use something like (x,y) data. I just need to set a little script to get the format acceptable by these machines.

Your option is definitely the one I should consider if none did it before. The challenge is to ensure that the geometry around those areas that may receive other blocks is precise enough to make it working as a puzzle.

Thx again.

M

BTW anyway to move this thread to the right place ? as apparently It didn't went to the place I wanted.

[Chris Ward]

Hi moon

Yeah, you would think that the blocks are initially defined in a graphics program - It's basically Google's Blockly behind the scenes. Given how the blocks can be mutated, I think it is highly likely that it's all done by code, no images required. I have written a Windows Desktop app which makes a preview of graphics and that is all drawn in code too.

In any event, for cutting the pieces out of a flat sheet material, the items need to be nested such that material waste is minimised. So, let's say the sheet material is A1 size, you would probably cut multiple copies of one jigsaw piece from one sheet. If using something like vinyl, the sheet could already be the correct colour (near enough) and it would be possible to write on it with "white board" pens - an essential part of the equation.

So, the pieces need to be defined in a 2D CAD program. Even better if that had a rudimentary CAM module to output g-code for the cutting machine. Being part of an Educational Institution, you should be able to take your pick from some of the most capable CAD-CAM apps on a free educational license.

[moon]

Okay Chris,

I'll take it from here.

Being also a 3D-CAD Professor (that's at the University) , I'll use the idea from Steve (Silly me I was thing borrowing a 3D scanner) which means using a graphic, will model the blocks  using Bezier curves or even B-Splines (in case those little bubbles are not really circular). This being said, considering pictures is not that easy because I don't master the scale on a display. All the blocks should be displayed with the same scale otherwise there will be no "puzzle effect" ... unless displaying all the existing blocks at once to ensure everything is scaled with the same ratio.

In order to minimise the lost material (as you mention it), I'm thinking of plugging several blocks (artificially).

I'll give it a try, this w.e, using my personal cutting tool (definitely NOT a NC one) that I used to use to build customised puzzles for my kids. It's ll be my proof of concept hopefully. The only question is really about the lost material between two plugged blocks.

Hopefully I'll get back to all with pictures of my realisation(s).

M

This is definitely going out of the scope of the forum, sorry about that.... even if it's still about AI2

[moon]

BTW Steve, thx a lot for the blocks. Going to check whether they're scaled with the same ration. As I told Chris this is critical.

M

[Abraham Getzler]

There will be a problem of how to cram successive blocks

into the inner space of the control blocks.

Like a snake, the outer block must stretch its jaw wide to eat

the blocks in it.

You might be able to do this using a sliding clamp construction,

or by using nested vacuum cleaner hose extension segments (very short ones) to

build up a proper length spine.

Considering the need to write names onto the blocks along

with parameters, consider making these out of folded colored construction paper,

using freely available Origami templates.  

With proper folding, the Origami shapes could be given valleys and peaks at their edges to

act as sockets and nipples .

ABG

[Abraham Getzler]

Another 2D approach would be to make rubber flat markable magnets in block shapes,

with extensible blocks coming in two or more parts that overlap on the magnet board.

Maybe magnetic backsplash boards  could be light enough

and cheap enough to lend out?

ABG

[moon]

Hi Abraham,

Interesting thoughts. 

You're right. I didn't explore enough to consider those needs. I'm at the very early stage.

The slider clamp is something easy to consider and definitely a good idea.

M