The Xylophone Maze: Screen-free coding for children
20y.huMy daughter and I play "dadbot" where I'm the robot and she has to give me clear instructions on what I should do. It started after I showed her https://lightbot.com/ but we like the "screen-free" nature of dadbot better.
Eventually she jumps on my back and the game doesn't last much longer because dadbot gets tired.
Sounds like it could devolve similarly to what's being done here: https://www.youtube.com/watch?v=cDA3_5982h8
I sometimes try with my children. Having one on my shoulder and pulling either the left or the right ear to tell me where to go. The way home from kindergarten can then take a while, depending on how attentive they are. Of course, parent stumbling into hedges makes for a good laugh, so that may also sometimes be intentional.
The first (Flash) version is playable here: https://dagobah.net/flash/light-bot.swf
It has a pretty steep learning curve if you're not experienced with building a program step-by-step from another perspective and you're only able to run it. The mobile version is a lot more gentle and has more content, but I think less challenging and less fun.
Yeah she and I prefer the flash version too.
The music is great, and sitting together at a computer is a better collaborating experience than the mobile app.
Unless you're on a roadtrip.
Sounds great. I was playing a lot of Exapunks[1] when my son was little and we started playing a game in real life where I would use the Exapunks commands to tell him what to do. Like
LINK HALLWAY GRAB BALL LINK BEDROOM DROP BALL
etc... Then he would tell me what to do. It was fun.
Fun idea! Clear objectives and simple tools. It's also a remote control example.
I find that "algorithmic thinking" is too artificial for young kids. They are very versatile in a richer set of methods, which often are mutually conflicting yet lead to desired result.
Once we played with a programmable mouse that needs to find cheese. Very predictably, the most used button was "do a trick", which makes mouse make funny noise without moving further... Sweet times for the kid, doomed moments for the teacher.
I would argue young kids are very good at logic and reasoning, but they lack self moderation controls and the context of The Real World.
For example:
Parent: "We're not going to have ice cream today."
3yo: "If we don't have ice cream today, can we have ice cream tomorrow?"
Or, more commonly:
3yo: "What if we have ice cream today and no ice cream tomorrow?"
Or, even more commonly:
3yo: "What if we have ice cream today and ice cream tomorrow too?". Because small children know better than to constrain themselves to artificially restricted choices offered by parents.
Source: experience from navigating such negotiations for the past 1.5 year with my now almost-5 daughter.
I wish I could find the source, but I'll relay it less eloquently. Children have a remarkable ability to ask for things they know they can't have. Their innocence gives them the audacity to ask for the impossible.
This always stuck with me. I sometimes catch myself when I self-curate questions to eliminate what I "know" will be impossible.
It also has led me to sometimes just say "yes" when I get those questions, just to impart a bit of expanded possibility into their life.
I wonder if it's because the average parent is not a very sophisticated bureaucrat. We get tired and just say yes because we need to survive. As children get older and interface with more institutions that have teams of people to tell you no (or no process to even ask), we enter a learned helplessness: a no today is almost certainly a no tomorrow.
This is an excellent activity in that it has multiple modalities for tacit learning. Keep in mind that when you are playing with young children, it is the time you are spending together that is most important, so if they want to not follow the rules, make up new games, or just bang away on that xylophone, let them, and enjoy the time you have.
To add: make sure you include unstructured play where the kid leads out without any structure from the parent (rules, etc). Instead, play with and have them lead you.
My first thought as well... it's so unlikely my kid would respect the rules at all.
First off I love this analog programming idea. I have young children who I would love to try this out with.
Maybe missed something in skimming through the blog post but seems like primarily it's simulating doing up/down/left/right and navigating a character through a maze. For some reason this seems to be the most popular approach for apps that teach kids programming.
i.e. https://kodable.com, which one of my kids is into and https://codecombat.com, which has been around for a while now.
I think this paradigm (navigating a character using "move" function invocations) is good but kind of exhausts its usefulness after a while. I question whether my daughter learns coding this way or just is playing a turn based top down platformer. The most code like thing is when you use 'loops' to have characters repeat sequences of moves. I think when kids grok these things these apps become just types of glofiried education flavoured video games. There are a lot of things in kodable for instance that I feel are just basic web games with coding terms slapped on it.
https://scratch.mit.edu/ is more like 'programming' imo, even at the level of the objective -- having a blank canvas to create something. It seems a little advanced for my kids right now though.
I love this idea! I can't wait to try it with my son when he gets a couple months older. By the way, what you have there is a glockenspiel. A xylophone is made with wooden bars, not metal as you have. It's a common mistake!
I'm aware of what the xylo- in xylophone means. But words mean what we use them to mean, not what pedants insist upon, and in vernacular, you're wrong, not him: the meaning of xylophone includes metallic instruments in the same style.
In musicology, sure, these distinctions are useful there. But what I want to stress is that you are wrong in this context. Not technically correct: wrong. The only mistake was you choosing to reply the way you did. It is, to be fair... a common mistake. Around here at least.
I knew both words when I was about 5, as we had both instruments at the back of the classroom.
Your approach leads to calling them all sticks.
> Your approach leads to calling them all sticks.
No, it leads to calling them all xylophones. Or calling the metal ones metal xylophones (which I do only if I remember that "xylophone" is the wrong term) and the wooden ones wooden xylophones (which I don't want to do but am stupid enough to do).
> I knew both words when I was about 5, as we had both instruments at the back of the classroom.
Even so, remembering that different words can exist for different instruments with the same general shape but differing other characteristics (wood vs. metal bar material, in this case) is a waste of mental effort for people who have reliable eyes (like me) and don't practice those instruments (also me).
I call the piano the "piano", even though the more functionally accurate term (also more historically accurate) is pianoforte. Even though I know this, I'm still going to call the piano a piano, because I don't want to bother to remember that the word "pianoforte" exists until someone asks me "why is it called piano?" As for tomato, I've learned to accept that a tomato can be a "fruit" in some contexts and a "vegetable" in some contexts (not mutually exclusive).
No, you are wrong and the GP is right.
Incorrect.
It's an amazing idea! I'm always looking for screen-free activities related to problem solving with my daughter. By the way, if someone happens to have other examples, it would be great to share!
My four-year-old grandson and I enjoy playing with go stones on a go board. The black and white stones and board’s grid lines inspire a lot of counting and geometric-shape activities. Similar things could be done with other types of objects, of course, but I had the stones and board from when I used to play go forty years ago.
My daughter wants me to teach him how to play go at some point, but he’s not quite ready for that. He hasn’t yet learned the concept of winning or losing at games, and we don’t see any need to force that on him early. In the meantime, playing with and talking about the stones seems to have boosted his understanding of numbers.
I think the "Peanut Butter Sandwich Instructions" game could fit into this category as well. For kids who can't write, you can do the actions in real time. For kids who can, have them write down instructions first then "run" the program. Teach them to "debug" or test their program along the way.
Every year I teach a few weeks of "coding" at my kids elementary school. I always start with the Peanut Butter Sandwich and it's a huge hit. For time's sake, i print & cut out 30-40 random instructions like "openTheBag();" or "holdJellyOverBread();". They get in 4 groups, choose which instructions to use, and put them in order.
I have a short collection of ideas here.[0] The biggest thing I've done is tell bedtime stories (often collaboratively) that incorporate topics like functions, negative numbers, and debugging. Here's one where we did FizzBuzz together. [1]
[0]https://twitter.com/CBancos/status/1581662053189574656
[1]https://bancosparenting.wordpress.com/2023/02/04/pickle-and-...
Getting someone else to do something is really hard if they take your instructions as literally as with a programming language: https://www.youtube.com/watch?v=waP5vkMaYjg
Another example in Dutch [1] (with activity sheet here [2])
[1] https://www.youtube.com/watch?v=7BcZBl-BKhQ
[2] https://www.leukekinderactiviteiten.nl/wp-content/uploads/20...
Well, for AI, the "Matchbox Computer"
https://en.m.wikipedia.org/wiki/Matchbox_Educable_Noughts_an...
A similar game but not DIY is https://www.thinkfun.com/products/robot-turtles/
I played Robot Turtles [1] with my kids, a boardgame with a similar idea: The kid can lay out cards that define what the turtle should do, and the adult moves the turtle exactly as instructed. Slightly different because, like in the original Logo, you can say "turn left" or "turn right," and it can be hard for the kids to remember that left and right are from the perspective of the turtle, whichever way they are currently facing.
By about age 7 or 8 it stopped being fun, because the kids could pretty much lay out an entire one-shot sequence of cards that solved the maze. (We never really got into trying to code "functions," it never quite seemed to be intuitive in the context of the game.)
I feel like a next level would be RoboRally. One of my all time favorite “programming” games. I think it may still be in production
One silly, maybe extreme, direction to move to try to explore building functions together is Alligator Eggs: https://worrydream.com/AlligatorEggs/
Thank you for sharing! I tried it out with my 7yo, who was interested. Unfortunately we had some issues with the toddler interfering so our session got interrupted.
I did find an online version here, for anyone who wants to try it without cutting out a bunch of alligators: https://tibordp.github.io/gator-calculus/
I am also on the search for more puzzles/challenges to do with it.
This blog post [0] has a few more examples, and I like how the author's alligator calculator [1] translates from traditional notation into alligators.
[0]https://propella.blogspot.com/2011/09/yet-another-alligator-...
It was one of the first successful boardgames on kickstarter, that really started the trend of boardgames being published by crowdfunding. Found this 2014 article claiming it was "the most backed board game in Kickstarter history": https://coolmomtech.com/2014/06/robot-turtles-game-from-thin...
There is a competitive mode for it, still available on the official site: http://www.robotturtles.com/galapagos/ I have thought of trying that some day. The original game has not been played much since my kids were younger, but they are also not yet all old enough to play RoboRally. On the other hand we also have Colt Express that I think is a very fun lighter almost-RoboRally kind of game (if you squint a bit), that might appeal more to kids too old to still want to play Robot Turtles.
> The robot must not look at what color bar was hit but rather carefully listen to the sound only.
Nice recipe for developing perfect pitch :-)
Hm, is perfect pitch stable across puberty?
One of my observations is that humming a note within my valid vocal range has "feeling" differences even within the hard cutoffs (usually about 2 octaves for most people). But those cutoffs (and presumably the feelings) move between childhood and adulthood.
(As an adult, my perfect pitch is not completely stable - if I don't use/tune it regularly, it can drift up to 2 semitones, but no further regardless of how long between tunings. Unfortunately I never did tests as a child.)
In my mind, all of programming can be reduced to the following 4 concepts.
- variables - conditions - loops - functions
If there was a way to teach those concepts screen-free, even to adults, I would really love to know it.
ThinkFun has a game CodeMaster that teaches loops and conditionals, but my kids found it way to tedious to play. It is better than nothing.
Maybe some HN'er has an idea how to create some game for all these concepts?
(And, if there is any other core concept I am missing, I would love to hear that too.)
>> (And, if there is any other core concept I am missing, I would love to hear that too.)
When you get into parallel programming, I think you'll need more than just those 4. But I'm not sure what the reduction is. Something akin to forking and joining.
Of course! There is another one that is really difficult enough to earn it's own pace on the list:
- naming variables.
See also Turing Tumble, a screenless, mechanical, gravity-powered computer for kids:
https://upperstory.com/turingtumble/
My sons (7 and 9) love it, and both have some grasp of binary thanks to it.
Upper Story’s Turing Tumble and Spintronics are solid favourites in our house too. Weaving a graphic novel into the instruction / tutorial book is a brilliant tactic which really helped my eldest grasp the concepts covered.
Spintronics also helped give me a new perspective on electrical current, current division and capacitance. Seeing and feeling resistance in the chain, how little/no load results in high current (chain link throughput) was more valuable than the “water in a hosepipe” analogies I learnt in my University EE classes. Really looking forward to induction in the expansion set.
I was curious about the company, and discovered that the co-founders are a husband and wife couple who were in engineering and education respectively before starting the company. I’m glad to see they are able to operate profitably without listing on Amazon, and hope they continue to release more excellent educational engineering toys that I can explore with my kids.
My kids also love Turing Tumble. (Me too!) Highly recommended.
I don’t have any kids, but this is a great idea for a D&D puzzle!
I started doing Turing Tumble with my kids. It gets to be quite a bit more advanced than something like the Xylophone Maze, but also zero screens. It's been fun.
This is a great new addition to my collection of low-budget screen-free ways to teach programming concepts to little kids.[0]
We don't have a xylophone but I might just have to get one so we can try out this game - I love it!
How is this programming rather than just learning how to recognize musical notes. Also, would this then start to associate these colors with these notes in any weird ways later in life? I'm not talking synesthesia or anything, but I can remember the music I was listening to at the time of reading a book or think of the book when I hear the music. Unless, that's the programming
Here are my two cents on why this is programming.
1 cent: Every hit is an atomic action that is causing the robot to take a certain action. Furthermore, all points in that maze has a decision (from at most 4 different choices) to make. So, hitting on a note (making a choice) is like writing an if statement. Furthermore, you can ask them to come up with the color combinations to hit before hand and try to run it all at once. If it fails, you do it again.
2 cents: Since this is designed for 2-3 year olds, if statements make a good basis for starting programming or logic in general. As they grow older, we can introduce loops and functions.
Moving on to the next question about starting to associate colors with notes although avoidable by randomly assigning colors to the notes (glue and paper), is possible like you said. However, I would like to claim that it will only stick (no pun intended) if the same colors play the same notes for years, if not months. Which given how two year olds are, is highly unlikely. They are done with a toy in about a week or two, max a month, give or take.
> if the same colors play the same notes for years, if not months.
I wouldn't be so sure. As I stated, I have an association between book<=>music which was made within the days it took to read said book
This is wonderful.
There's a market for a pre-built set like this, with preplanned mazes to produce popular children's songs.
Just make sure your xylophone is actually in tune. Its sad how many cheap kids toys look like musical instruments but don’t sound correct
This is cool, just be aware it could be frustrating if a child has any sort of colorblindness, particularly since the Duplo blocks are different shades than the xylophone. It would help if the blocks had letters on them to match the notes as well.
This is a wonderful idea, and although I don’t have kids, I think I want to have the pieces to put it together for when there’s one to play with.
If there are more than one color-matched blocks, then where the player move to? Just random?
Nice idea for something without screens
I like the sneaky missing block.