Show HN: I'm 12, learning JS, and wrote Wolfram's cellular automaton in Node | Hacker News Reader

Show HN: I'm 12, learning JS, and wrote Wolfram's cellular automaton in Node

bitbucket.org

1,072 pointsliamilan283 comments

Liam, this is incredible!

I thought it might be useful for you to incrementally see how someone in industry would review or change this code, so here's a little code review via video: https://youtu.be/UkVOrcS--04

We very incrementally build up to the final code, which can be found here: https://gist.github.com/stevekrenzel/b490564bf1c7f98e232a6c8...

Hope you find it helpful!

Kudos for taking the time for doing this! It's places like these where the HN community really shines.

Thanks a lot!

Thanks! I really enjoyed making it. HN has been such a huge part of my life over that last ~10 years that it's truly the least I could do.

If you would start a channel where you review code of random public repos I would definitely subscribe. Well done.

Seconded. Do it. :-)

+1, reminds me of Destroy All Software

+1 Please and thank you

Ron Swanson?

Thanks for the kind words! Will definitely consider.

I would absolutely sub to this.

me too i'll subscribe to your channel if it existed

Please consider. This is the rare time I'd actually use some sort of !remindme bot. Most lurking autodidacts, I assume, are hoping that you consider it too.

Come out of the woodwork, let him know! If I can, you can.

I'd love to both participate and lurk :)

Watching your video was like watching Bob Ross help a wine and paint class. Well done.

Let us know here if you are planning to do it :)

Just registered to +1 this

+1 count me in

Drive-by code reviews (and maybe other types of annotated "audits", like for security bugs) would be a nice feature source repo sites might consider adding one day. It seems like a missing practice in open source development, code reviews (if any) are done by active developers once on check-in/merge acceptance. But as we see here code reviews can be a useful way to pay it forward too, not just new code / issue fixes, and done out-of-band there's no pressure to bow to the reviewer just to get something in; you can just have conversations and learn things.

The video format is a nice twist, in my distributed teams sometimes we've done "code review meetings" for some changes that are rather involved to capture some of the benefits of in-person / video code reviews.

Steve, it is outstandingly awesome that you did this and also how you did this! Thanks.

Thank you, much appreciated!

Thanks Steve!

I watched your video and although I don't understand everything, I learned a lot. Separating my program into functions is a good idea. Your video clarified the map function for me.

Through your video, I understood that I was running code multiple times, when it only needs to run once (how I parsed my rules was one of these instances that you went over in your video), so thanks for pointing that out.

Again, thank you so much for taking the time to review my code. I really enjoyed your feedback.

So glad to hear, Liam! Was a pleasure to put it together.

Good luck, and feel free to reach out if you get stuck on anything.

Every learner could do with a mentor like you!

> Again, thank you so much for taking the time to review my code. I really enjoyed your feedback.

Whatever you do, hold on to that. Never be mad about feedback just try your best to understand it, and if it's wrong explain yourself without getting frustrated, this is worth more than a dozen senior developers who explode at feedback (I would argue they're junior developers :) and you will have a wonderful career. Also look for good mentors, and don't be afraid to ask questions about direct feedback, if you didn't understand something he said just ask (this becomes an immediate learning opportunity and you will be better off for it), even if he doesn't get to answer, someone else might. This is a field where learning is endless, you will always learn something new as a software developer, new platforms are made all the time to be programmed on / with.

Having soft skills[0] will be the biggest asset you can have as a software developer, whether you do it for money or just for fun, if you have to collaborate in the open you will need to communicate with others. Nothing breaks a team like terrible communication and someone who rejects feedback out of pride.

[0]: https://en.wikipedia.org/wiki/Soft_skills

This whole interaction is really reaffirming my faith in humanity. You guys are awesome!

Liam -- this response is just as impressive as the code that you wrote. You're obviously very smart, but that isn't good enough on its own. I've seen very-smart people who aren't open to critique, who get defensive rather than take the opportunity to learn new things. They don't get very far in life. On the other hand, the attitude you're showing here is tremendously valuable, and will serve you incredibly well throughout life. Never let yourself grow out of it!

Nice of you Steve.

Liam - whatever Steve suggested are very good pointers but please remember you are getting accolades here because you cared to publish your code. Don't hesitate to publish your initial version without any hesitation in future also.

I'm subscribing to you, just incase you do more video's on code reviews.

Instead of this:

    (rule & (1 << i)) === 0 ? 0 : 1

You could just do:

    (rule >> i) & 1

Otherwise it was a good review! I'd probably also extract parts of this long line so it's not as cumbersome to read:

    return state.map((item, i) =>
        rules[((state[i - 1] || 0) * 4) + (item * 2) + (state[i + 1] || 0)]
    );

    return state.map((item, i) => {
        const prevItem = state[i - 1] || 0;
        const nextItem = state[i + 1] || 0;
        return rules[(prevItem << 2) | (item << 1) | nextItem];
    });

It also uses bitwise operations instead of arithmetic, even though they are equivalent in this context, simply because we are effectively operating on arrays of bits.

I like to break expressions across multiple lines, and line up any patterns that are repeated, so your eye can easily scan up and down and obviously see what's the same, and what's different.

For example, instead of:

    var length = Math.sqrt((x * x) + (y * y));

You can go:

    var length = Math.sqrt((x * x) +
                           (y * y));

That way you can see the similarity of squaring x, and squaring y. But breaking up the multiplications because they repeat the x's and y's would be taking it too far -- it always requires balance and thinking about how can I communicate my intent and the SHAPE of the problem I'm solving.

You can also use indentation and formatting to bring out and make obvious other two-dimensional patterns and regularities, especially with two-dimensional cellular automata and graphics code.

There are some examples of that technique in my 2D cellular automata machine code. (There's a lot of code, so don't be overwhelmed, since I've been working on that since around 1985 or so. So never give up and just stick to it! You don't need to understand the code, which is pretty esoteric, just notice the patterns.)

https://github.com/SimHacker/CAM6/blob/master/javascript/CAM...

                        // Load the right two columns of the 3x3 window.
                        n  = cells[cellIndex - nextCol - nextRow];  ne = cells[cellIndex - nextRow];
                        c  = cells[cellIndex - nextCol          ];  e  = cells[cellIndex          ];
                        s  = cells[cellIndex - nextCol + nextRow];  se = cells[cellIndex + nextRow];

[...]

                            // Scroll the 3x3 window to the right, scrolling the middle and right
                            // columns to the left, then scooping up three new cells from the right
                            // leading edge.
                            nw = n;  n = ne;  ne = cells[cellIndex + nextCol - nextRow];
                            w  = c;  c =  e;  e  = cells[cellIndex + nextCol          ];
                            sw = s;  s = se;  se = cells[cellIndex + nextCol + nextRow];

[...]

                            var sum8 =
                                    (nw & 1) + (n & 1) + (ne & 1) +
                                    (w  & 1) +           (e  & 1) +
                                    (sw & 1) + (s & 1) + (se & 1);

[...]

                                var tableIndex =
                                    (((c         >> plane) & 0x03) <<  0) |
                                    (((e         >> plane) & 0x03) <<  2) |
                                    (((w         >> plane) & 0x03) <<  4) |
                                    (((s         >> plane) & 0x03) <<  6) |
                                    (((n         >> plane) & 0x03) <<  8) |
                                    (((cellX             ) & 0x01) << 10) |
                                    (((cellY             ) & 0x01) << 11) |
                                    (((phaseTime         ) & 0x01) << 12) |
                                    (((plane     >> 1    ) & 0x03) << 13);

[...]

When the sub-expressions won't fit on one line, you can still use indentation to help the reader understand the spatial patterns:

                            error +=
                                (nw * kernelBytes[4 - kernelDown - kernelRight]) +
                                    (n  * kernelBytes[4 - kernelDown]) +
                                        (ne * kernelBytes[4 - kernelDown + kernelRight]) +
                                (w  * kernelBytes[4 - kernelRight]) +
                                    (c  * kernelBytes[4]) +
                                        (e  * kernelBytes[4 + kernelRight]) +
                                (sw * kernelBytes[4 + kernelDown - kernelRight]) +
                                    (s  * kernelBytes[4 + kernelDown]) +
                                        (se * kernelBytes[4 + kernelDown + kernelRight]) +
                                frob;

But that would probably be more readable if I used intermediate variables for the kernelBytes[...] expressions, gave them descriptive names matching their corresponding directions, and lined up all their calculations so you could see the similarities and differences, because right now all the kernleBytes[4 +/- kernelDown +/- kernelRight] expressions are jumbled around horizontally, when they could be lined up nicely by themselves if they weren't interleaved with the n/s/e/w/ne/nw/sw/se multiplications. (I'll leave that improvement as an exercise for the reader. ;)

The point is you want to vertically line up as many of the repeated letters and symbols in nice neat columns as possible, so that your eye can easily see which are the same, and the ones that aren't stand out obviously so you can ignore all the same things and focus on the differences. (Which are typically variations like +/-, sin/cos, x/y/z, -1/0/1, nw/n/ne/w/c/e/sw/s/se, and longer series of names and numbers. Whatever changes should stick out visually, and be lined up and grouped together so your eyes can scan over them!)

That not only helps other people understand your code, but it also helps you be sure that you wrote the right thing, what you actually meant, instead of making an easy to miss typo.

The difficulty of reading you own code, which is HARD even for professional programmers, is that you usually see what you MEANT to write, not what you actually wrote. That's why it's so important to get other people to read your code, and to read other people's code, like pair programming and code reviews, or even explaining it to a duck.

https://en.wikipedia.org/wiki/Rubber_duck_debugging

It takes a lot more effort and concentration to slow down and look at what's actually there, instead of what you want to be there. (That's true for many aspects of life...)

it infuriated me that Google's style guides forbid this kind of formatting. Anytime someone else tried to get them to allow it it was shouted down as ASCII art

If I'm honest, I like that kind of formatting for small 2-3 line, 2-3 part expressions. But I am not a big fan for it in much bigger parts (like that huge index calculation above). I always see them and think that I'd rather see a loop over an array of values, or split into many smaller expressions.

Thanks for the edits! I can't change the video, but did mention that both of those lines are a little clunky. I did update the gist though! Much cleaner.

I'm a long time coder and still liked watching your video just for watching someone else work in emacs. How do you get emacs to highlight all occurrences of map? Was it a keybinding for "highlight-phrase"?

This is vim.

I think I just died.

WHAT YEAR IS IT

Plausible deniability: It could have been Emacs pretending to be vim.

Thanks, glad you enjoyed it! And like another person already responded, this is vim with a split pane where the right pane is a `:terminal` (a relatively new feature of vim... though prior to that I would have just used tmux).

Highlighting searches is enabled with `:set hlsearch`, and you can search for whatever word is under the cursor with ``. So I just type and it highlights all of the matching words. Nothing too magical!

Incredible code review!

I wish this inspires clever people to produce huge amounts of parody-videos "refactoring" the same piece of code. :)

Congratulations!

I showed your project to my 12 year old son who has just started to learn programming (in JS no less!) and his jaw dropped.

In addition to being a programmer, you are now a role-model. Really well done.

Thanks a ton!

Tell your son that it took me a while, and a lot of work, but it feels great to finish something (there are a lot of projects that I don't finish...).

I began learning using Scratch a long time ago. They even featured me last year (the project was a solar system in which I used sine and cosine to calculate the rotation of the planets). I then used blocklike.js to move to JS.

I watch a lot of YouTube videos. I like The Coding Train, Carykh, and Code Bullet, and I get a lot of my ideas from their projects.

> there are a lot of projects that I don't finish...

You and me both. :)

He's one of us :)

Startup Idea: Start a slack/discord group for aspiring students between the ages of [pick your age range, say (8-18)] and start a mentoring/support group with these like-minds...

You'll learn, build relationships and some of you will go off and found companies together.

Starting a Slack group for students is a good idea, but decidedly not a startup idea.

What about a saas to support the companies providing groups for slack.

I've considered building a "coding buddies" site. Something that's been lacking about the whole Codecademy/Hacker Rank/etc sites is the lack of intrinsic motivation. An actual person with whom you can work could potentially provide that.

Yeah thats what I was inferring... Basically - start a slack, find a bunch of people like him - pick a project to build together that suits their need (coding buddies) and build the tool.

Kind of like when I was trying blacksmithing, the teacher makes you make your tools first.

I'm 21 but I wish something like this existed when I was younger!

I'm 47. There were computer camps and clubs when I was a kid. They probably took a downturn, though, as computing went mainstream and turned into consumption of video games and, later, online socializing.

Maybe you were either too late, or too early, depending on perspective. :)

38 here. I was learning Turbo Pascal in around 96. I'm sure there was a lively community around it, but it would have been on IRC and usenet, which was still a little daunting to me. My primary resource was the SWAG archives, and I honestly can't remember if it was a bunch of floppy disks or a website. Either way, I remember many hours lost to running various random bits of Pascal and hacking them together to make what I wanted. Which is pretty much the same way I code today.

I wish I was 21. Or 31 for that matter.

I am 24, and was programming when I was 12. I remember spending days trying to get Visual Studio to download. I began working with Visual Basic Macros in Word. Although it wasn't very well understood in the schools around me, I found the comfort of those online willing to teach, and help others like it seems he is being shown here.

The mentality of those taking boot camps, and the mentality I had when I started coding are vastly different, but I can see the use.

sine and cosine last year? Now I want to know what your parents are feeding you!!! ;)

I also picked up sine and cosine at around that age; it was thanks to programming.

Various example BASIC programs (e.g. Lissajous curve and circle plotters and such) used those functions and from their behavior I kind of picked up the relationship to both undulation and the circle. I found out about the arctangent function, and from that I realized that it could be used for 3D projection. I made a proof-of-concept program that mapped some connected points in 3D (boxes and whatnot) to 2D by calculating the visual angles using arctangents: i.e. I plotted the angle between the line of sight and each point, rather than a simple planar projection. It resulted in the objects appearing in a very nice "fish eye" perspective on the screen, which was very pleasing. From all that I learned about radians also and how there are "two pi" of them in a circle, since the functions accepted radians and not degrees.

You know how they say that necessity is the mother of invention; it's also the mother of learning trigonometric functions and whatever else.

Much better motivator than math classes in school too

Similar age for sin/cos. I wanted to move a game object in a circle that I used to develop amateur content for. I remember being wowed by light displays some scripters put together at the time (there was no particle system at the time).

By the time it came around, I was disappointed to learn about all the "triangle stuff". I want quite there enough to make the link between right angle triangles and a circle but it made that topic so much easier to connect.

Thanks to programming and the thought patterns it drums into you, I was able to do my maths course work in a few pages (far less than anyone else) but still achieve high marks. I remember one top performer in my class complained to my teacher because of how little I'd produced! The teacher said "doesn't matter - he used an approach that let him do that".

I learned how to use sine and cosine for drawing circles when I was probably 9 or 10. It was years before I found out what they actually mean and how they work. I didn't even know what the curves looked like.

I've got a 10 year old who's been making a lot of stuff in Scratch since he was 6. I've been trying to think of how to phase him over to less limited languages and wasn't aware of blocklike.js, so thank you for that idea.

Having a lot of projects that you don't finish is completely normal, so don't worry about that.

>I showed your project to my 12 year old son who has just started to learn programming (in JS no less!)

It would be much better if he would start with Scratch, Processing, Smalltalk, or even Python.

On JS he'll have to lose time coping with complicated build tools, horrible package system, async etc.

Is this serious? I started programming around 9 or 10 on a Commodore 64. I would go to the local library to get programming books for it. Now you can hit F12 in your browser and get a powerful console into JS. JS is fine, encourage the thirst for knowledge.

JS is an objectively bad programming language. It is not fine, it is just annoyingly common and convenient. The main advances in its ecosystem are related to transpiling it from other, saner, languages. If he has fun writing JS, then great for him. He's 12, he'll do it if he wants. But it'd be doing him a disservice not to open his eyes to much wider, greener pastures of the programming world.

JS was my second language (around the same age as he is, which is more than 15 years ago) and it is still my favourite one. You give zero arguments but I won't make the same mistake:

- devtools are unmatched for browser development (and getting better for node)

- absolutely best language for prototyping (dyn. typing, monkey-patching, eval, and a lot of other "evil" things you usually don't need in further phases but it's really great to have them for this exploratory phase)

- I respect that some people prefer static typing and code-completion but I prefer writing tests and having few well-known libraries for which I don't need completion at all

- no threads, no mutexes, no locks

- object literal, rest-spread, destructuring, classes are just ordinary (and mutable) objects, etc...

If you think people are doing JS just because it's the only language in the browser (not true anymore) you should seriously reconsider. There were to-js compilers before and it was always awful (remember gwt? dart?)

Of course, it depends on use-case, JS is not the best language for everything.

Transpilation certainly has been important, but in no way is that the only or even the most important advance. The most significant advances are the massive changes in architecture that have turned it into a modern, fully-featured language. You can use most of those in all modern browsers right now. It might not be the best language, and isn't a great choice for a first language to learn, but it is a good one to learn next.

JS is a bad language but it's the easiest environment to set up.

Back in the day we started with Qbasic not because it is a good language but because that was easy to set up, we all had it and our friends had it and our school computers had it.

And JS is less bad a language than Qbasic.

That's debatable. It's not as if downloading Python and making your first app will take you more than a few seconds, minus download time.

Set up is rarely an issue, once you set up a language even as complex as C++ its completely automated. You press a button and magic happens.

If the biggest selling point of a programming language is that it comes preinstalled with the OS of your choice, its not much of a selling point when the average user downloads and installs GBs of data just to play the latest game hype like Fortnite.

The only people who mind installing stuff are predominately web devs. 99.9% of users out there are fine with it for more than 6 decades now and that wont be changing any time soon.

It does make a difference especially to a beginner or a child. You might not have admin privileges at your school or your friends may not be able to install those tools so you can't show your work.

But you can hack away on any computer with a browser.

there are tons of services only that will allow you to code online using any language eve C , C++ and Assembly. From inside the browser of course. Of course I am not recommending a beginner to start from any of those languages but it shows that JS does not have the monopoly of in browser coding. If you have to stick with browser no matter what.

Ideone is a popular example https://ideone.com/

no man its objectively good language, believe me.

see how that sounds? that's how you js haters sounds each time I see this bs here on this and many other platforms that wouldn't be possible without js.

> no man its objectively good language, believe me.

A language that requires a 21x21 matrix to remember how equivalency works is objectively bad: https://dorey.github.io/JavaScript-Equality-Table/unified/

A language whose default sort behavior is the following is objectively bad:

    [5, 12, 9, 2, 18, 1, 25].sort(); → [1, 12, 18, 2, 25, 5, 9]

I used to be a frontend dev. I have written hundreds of thousands of lines of production JS code for apps that many Hacker News users use on a regular basis. I well understand how much JS has made possible. The web would not be the same without it. But that doesn't mean it's a good programming language.

This kid is practicing calculus at 12 years old. He will be able to handle a package manager.

It's still a waste of his time to deal with it. The brain has only so many ultra-neuroplastic years, why burn them on dicking around with npm when they could be leveraged to focus on more universal concepts?

because "full-stack"

JS is the New York of languages: dirty and gross and hella expensive, but if you can make it there, you can make it anywhere.

Isn't Liam's project proof that you don't need any complicated JS tooling?

A simple `node index.js` can go really far.

Yes amazingly he did this without the need of `left-pad` too. ><

Good job Liam

The only thing you need to learn programming with JS is a text editor and some rudimentary HTML knowledge.

Everything else is unnecessary, at least as far as the programming language itself is concerned.

You need none of those things to get started with JS development

It’s still perfectly possible to just create a html document, shove in a script tag, and start developing

And if you want to use packages from the npm ecosystem you can easily use https://unpkg.com/

Great work! I made a quick Observable notebook out of your code, in case anyone wants to see what it does without downloading or running it on their own machine:

https://beta.observablehq.com/@jrus/wolfram-cellular-automat...

Dang, that's a neat tool. It's different enough from a Node environment, though, including some of the necessary rewrites of stuff, that I wonder if a "more Node-y" sandbox would be more useful for this particular case. An example: https://codesandbox.io/s/r0kwk0o28p (I tweaked the rule number, column width, and output to print to the page, but otherwise it's the same as the original.)

Hi Jacob, I never got to thank you for doing the observableHQ notebook for my quasirandom sequencing. For your information, I have linked to it on many occasions which includes in my post "Unreasonable Effectiveness of Quasirandom Sequences" which is on front page of HN right now! :) https://news.ycombinator.com

Hey looks good :)! Just a few things you might not know about:

Google String.prototype.padEnd. It’s built in to JS and should be able to replace your zero fill function.

And instead of mutating the ruleSet array you should be able to use Array.prototype.map like this:

const ruleSet = zeroFill(8, rules.toString(2)) .split('') .reverse() .map(item => parseInt(item, 10));

And you can clone arrays like this, instead of using loops:

const oldArr = [...arr];

Also arrays are still mutable even if you assign them to const variables.

I’d offer more help but I’m on mobile. Sorry for the poor formatting. Hope this helps anyway. Good luck!

Though, keep in mind that some of this stuff (most notably the [...arr] syntax) won't work in old browsers if you make a web project (IE 11 is a big offender here). See https://caniuse.com/ for a useful source of info on what will work with what for stuff like that.

If you do web stuff (and you don't want to just say "no IE 11" - some people, and even some big companies, do that!), usually the easiest way to handle stuff like that is to use the newest syntax you're comfortable with, plus a tool like Babel that will produce a "time travel" version with older syntax that will work in all browsers.

This project explicitly says Node. But in any case, these days I generally recommend using Babel to transpile the newer JS to browser friendly JS as the productivity and readability improvements in newer versions of JS are worth the effort to me to set up the babel build step.

> This project explicitly says Node.

While that's true, future projects are a lot easier if one doesn't have to keep re-referencing "what works in Node" vs "what works on the web".

current projects are a lot easier if you can use modern syntax that will eventually be supported in future projects, and which can be transpiled in either way

Good tips if you want your code to fail randomly on older devices.

Older devices running node?

Surprisingly there are some embedded developers who use Node. I would assume those devices have a trimmed version of the Node stdlib.

I'm sure when he ports it to Espruino he can handle Babel.

The code would fail to run on older devices without transpiling wnyway.

Not even businesses doesn't (and shouldn't) care for older devices when it doesn't make economic sense -- even less so a fun learning project.

I think it is good he's learning to do the loops by hand first. It's good to have an understanding of algorithms first when learning to code.

I second this. Knowing why something is better to use is most likely learned through doing it a worse way first.

I third this. Speaking from personal experience. I sometimes try to implement the logic of some library or algorithm myself, even if I know there is well-known or canned solution for it. For the same reason as you and your parent comment said.

Interestingly, some people (typically newbies who are looking for rewards without commensurate effort. but others too) do not think that way, and even sometimes don't like it if such suggestions are made. I once made such a suggestion to a poster (an obvious newb) on comp.lang.python. He asked how to do some small task in Python. I suggested that he may learn more if he tried it on his own first. He misunderstood the intention behind the advice, and made some comment to the effect that my idea was not helpful, or something like that. Another regular pitched in and pointed out that my comment was meant in the spirit of helping him learn better, because a) if you try it yourself first and do get a solution, you just got better at something, and it will boost your confidence, and b) even if you do not get a solution, you will learn that the thing is not as easy or trivial as you thought (or why were you asking in the first place, except if wanting others to do your work for you) (which would happen if you just blindly used some library to do it). And will gain a new respect for the depth of your profession and all that it entails.

Related: How To Ask Questions The Smart Way:

http://www.catb.org/esr/faqs/smart-questions.html

It's by Eric Raymond, author of The Cathedral and the Bazaar and The Art of Unix Programming.

There are many interesting points in that article, e.g.:

http://www.catb.org/esr/faqs/smart-questions.html#noprivate

Seen people ask for private replies. IMO some of them do it because either a) they don't want to be seen as not knowing the thing if it was obvious, or b) not wanting others (likely their peers) to get the answer too. Both not good.

>...sometimes don't like it if such suggestions are made

I'll admit to occasionally being frustrated by comments like this. It's important to remember that not everyone is looking to learn. Sometimes, they just need to get something done ASAP.

It's a bit idealistic to assume that everyone has time to learn the canned solution. The real world doesn't always allow for that.

Sure, but it's still a valid advice that might be helpful to others.

True. I did not mean it as an absolute statement. I did say above:

>I sometimes try to implement

It's each person's call, on a case-by-case basis. Had just shared my view with that newb, thinking it might help him.

In fact this "you should never code FizzBuzz yourself, just use util.FizzBuzz(n) to print the first n iterations" mentality probably contributes to the surprising number of software developers who can't code.

And yet, we also have nearly-infinite number of (reinvented) wheels...

Humans are the worst

Good for you young jedi. I started coding around that age as well. I made the mistake of not going through comp-sci but have still been able to work for Google, publish books, lead teams and companies and products. You'll do just fine.

You have the ability to write some code! If you want to get a jumpstart, you should take the princeton algorithms course (for free!): https://www.coursera.org/learn/algorithms-part1

Or read the textbook by sedgewick that accompanies the course. https://algs4.cs.princeton.edu/home/

There is very little math in there - it may take some elbow grease and mentorship to help to convey some of the ideas - but I believe that you could implement and solve most everything in there. That was my path - I had a teacher at the age of 14 who would explain to me how different algorithms worked and I would go and implement them. Drawing a circle (in memory - it was in C!) or sorting a list (we did bubblesort first IIRC!)

I think you could do it! I believe in you! The course material is approachable - much more so than basically every other algorithms/data-structures material I've found. it may take you some time but you'll be soooo far ahead with your thinking about code.

If you ever start working with Object oriented languages like Java, another book that may help you when you've gone down the road a bit is the Head First Design Patterns book. http://shop.oreilly.com/product/9780596007126.do It's very easy to read, mostly pictures. It is made to be very easy to read (all of the books in that series are so look around at them.)

It's helpful to do both - code and also take in some material, but at 12 I imagine some of the material may be a bit daunting. You're doing really well - keep it up.

Liam, brilliant!

I am not saying you don't do this, but please please ensure you always sleep well, follow a good diet and go out and play.

Sometimes, computers make us give up good habits!

Kudos for your great work and keep learning, keep shipping!

Woah - very cool! I'm a part of Hack Club (https://hackclub.com), which is a nonprofit worldwide network of high school coding clubs. Hack Club has an amazing community of young developers - I actually lead a Hack Club at my high school, and I definitely recommend that you join their slack (https://hackclub.com/slack_invite) to connect with other young makers!

By the way, I'm a 17 year-old developer (https://shamdasani.org), but I probably started ~14. My main project is Enlight (https://enlight.nyc) to teach others to code by building projects. If you'd like to contribute, that would be awesome - feel free to shoot me an email :)

https://photos.shamdasani.org/ doesn't work

Hey Samay! To all in the thread, I started Hack Club and am happy to answer any questions about the org.

Liam, this is super cool. I also got started young and wish I had the skills to frontpage at 12 :-). Keep it up!

Would love to have you say hi in the Hack Club Slack. I'm @zrl.

That's awesome. What resources did you use to learn Node and JS? Just looked at your code. Kid. You're a badass. Keep this up, you'll go places.

Thank you! I'm watching a lot of Youtube videos. This is how I got to cellular automata. My dad showed me git last week.

That's so freaking cool. Git rocks, and I remember how psyched I was when I learned about it. What's your next project?

I'm not sure. I'm always messing around with different projects.

I used blocklike.js to jump from Scratch to JavaScript. Then I worked for months on this one game for the app store called Sticky Quest. You can find it here: https://itunes.apple.com/us/app/sticky-quest/id1425679649?mt... This how I learned how to use npm packages such as Cordova and Eslint.

I'm 28 and have been using Javascript for years and still have no idea how to use Eslint in a terminal.

You could try expanding this project beyond elementary CAs.

One option is a 'totalistic' CA which looks at the average value of a given neighborhood, and can have more than two states.

http://mathworld.wolfram.com/TotalisticCellularAutomaton.htm...

Or you could generate CAs with different starting conditions. Some beautiful elementary CAs don't 'show up' with just one initial point.

The Moveable Feast Machine is similar to cellular automata, but different in some important ways, that make it extremely robust and fault tolerant:

It's a "Robust First" asynchronous distributed fault tolerant cellular-automata-like computer architecture.

Robust programs running on massively parallel unreliable hardware can actually tolerate hardware failure and repair themselves. The Demon Hoard Sort algorithm is an inherently robust sorting algorithm for the Moveable Feast Machine.

http://movablefeastmachine.org/

The "Distributed City Generation" video demonstrates a Movable Feast Machine rule that builds a self-healing city that fills all available space with urban sprawl, with cars that drive between buildings, and city streets that adaptively learn how to route the cars to their nearest destinations, and the city even repairs itself after disasters!

https://www.youtube.com/watch?v=XkSXERxucPc

Here's some more info:

https://news.ycombinator.com/item?id=14236973

How did you like Scratch, and how was it going from Scratch to JavaScript?

There's another visual programming language called "Snap!", which is inspired by and similar to Scratch, but has the full power of JavaScript and Scheme, everything's an object, and you can define your own blocks. Plus there are some amazing networking, speech and AI blocks, too! (It has a JavaScript block that makes it very easy to extend and plug in existing JavaScript code.)

I'm interested to know what you think of Snap!, from your perspective of going from Scratch to JavaScript!

https://snap.berkeley.edu

AI extensions of Snap! for the eCraft2Learn project:

https://ecraft2learn.github.io/ai/

>The eCraft2Learn project is developing a set of extensions to the Snap! programming language to enable children (and non-expert programmers) to build AI programs. You can use all the AI blocks after importing this file into Snap! or Snap4Arduino. Or you can see examples of using these blocks inside this Snap! project.

Here's more info:

https://news.ycombinator.com/item?id=17594403

> My dad showed me git

Cheater! :) Seriously though, great work. I started around your age (which I guess means I've been coding for half my life at this point.) Keep challenging yourself with new and interesting projects and you're going to learn a lot and have a lot of fun.

That's pretty quick!

Do you have any plans for future projects? Maybe a simple place where you could chart your progress?

I think one of my favorite uses of cellular automata besides conway's game of life (if you are interested in it) is modeling simple fluid mechanics (water flowing from place to place) in games like dwarf fortress http://www.bay12games.com/dwarves/

> My dad showed me git last week.

Anyone else miss learning this quickly?

In all seriousness, congrats. Keep at it. There is always something else to learn. It's the career that never stops giving.

Looks great. I tried it and it runs fine. Nice syntax and convention.

Just a tip, you don't need to include `npm init` in your instructions. That command only needs to run when you create a new project. It worked just fine for me with just a `node index.js 18`

Great work!

Thanks! You're right, I fixed it.

Great that you start so early, I've did the same. For me it was a great way to gain confidence and build a skill which is very useful. I wouldn't be the same person now if I didn't program so early.

Seeing how things come to life that you've build with your own hands is a very empowering experience. Keep it up! :)

Far better than I could do at that age!

I think the furthest I was getting at that time was building out really terrible looking web pages in Notepad and making awful little animations in BASIC and Turing (learning lang).

Keep it up!

I made a colourful ASCII birthday cake in Pascal that beeped happy birthday at that age. Still one of my proudest programming moments.

writeln(chr(7)) ?

or something more (e.g. playing different frequencies to sound like the happy birthday song)?

Related, and good fun for kids of all ages (first did something like this on my earliest home computer [1]):

Play the piano on your computer with Python:

https://jugad2.blogspot.com/2013/04/play-piano-on-your-compu...

[1] Lissajous hippo, retrocomputing and the IBM PC Jr.:

https://jugad2.blogspot.com/2012/09/lissajous-hippo.html

Western classic music fans may find the comments on that post interesting - stuff about octaves and frequencies.

I was on my Atari 800XL making games at that age. I started hitting problems that I thought should be easy. Like, I wanted to draw circles. But the library was just pixels and lines ( when I got to assembly programming, suddenly lines became a problem too ). There was no googling the answers, but luckily I could go "Daaaad..", who was a physicist, and he happily introduced me to trig and I eventually learned to draw circles....slow drawing circles, but still, circles.

+1. Reading your answer made me think you were talking about me. Except I had a TI99/4A, and "Daaaad" was my real google. My circles were probably slower and more pixelated

Great job! One way to improve this code is to add comments. Use comments to describe what each function does and what the parameters are supposed to be. Then you can tell at a glance what the function does without having to analyze the code later on.

Us old people tend to forget things.

Also, if you use JSDoc a lot of modern editors can use that to automatically provide type hinting, documentation references on mouseover, etc.

Comments in a twelve-year-old's code? For whom? Seven-year-olds? (I'm serious.)

How about for a 32 year old who is just learning JavaScript and comes across this repo? Age has nothing to do with it.

Why not learn good programming practices at an early age?

Perhaps for himself when he's thirteen or fourteen and wants to refactor

This is awesome! Keep up the good work. As a fellow developer who began writing software around your age, I have some advice that is definitely subjective and biased, but which you may indeed find helpful.

1. Keep practicing. Practice makes perfect. Explore different languages, frameworks, and problem spaces (interfaces, servers, distributed systems, statistics, etc.) to see what you find most interesting. You will make mistakes, and that is okay. That is how you learn.

2. Take frequent breaks. Programming is hard work. It can be truly exhausting, and you should not be afraid to step away and go for a walk or something.

3. The most valuable people in the industry are called "T"s. I.e. people with a little bit of knowledge/experience across broad range of topics (the horizontal part of the letter T), but who have deep knowledge/experience in one or a couple particular topics (the vertical shaft of the letter T).

4. Begin to develop the skill of taking feedback and accepting criticism. There's a lot of people out there who will be quick to criticize everything you do if you put it out into the open. It's a whole skill in itself to be able to interpret this feedback and separate the signal from the noise. Try to empathize with the person or group providing feedback and understand that their motives/perspective may be different from yours. Sometimes that's a useful thing. Sometimes it's a distraction.

5. Dn't take things too personally. You are not your code. A criticism of your code is not a criticism of your character. The less you take things personally, the easier it is to work with others. After all, the most impressive systems require immense collaboration.

6. Don't take yourself too seriously. Have some fun with it! Programming is super fun, so you should ask yourself regularly "am I still having fun?". If the answer is "No", maybe find something else interesting to focus on for a bit.

7. Broaden your perspective. Along the lines of #1 and #2, it's important to maintain a broad perspective and push yourself to keep expanding your perspective. This doesn't just apply to technical problem areas or languages or frameworks. The best engineers are good problem solvers because they have a broad perspective not just of the problem space but also more generally of the world they inhabit. Try to learn about different industries, cultures, people, and places. You will become a more well-rounded character for doing so, with a higher ability to empathize with others and understand the complex mechanics of how the world works.

8. Be social. I made the mistake of hiding in my room on a computer for much of my childhood, and it wasn't until high school that I really began to understand the value of social interaction and maintaining strong solid friendships. It's as important to spend time away from the computer as it is too keep practicing.

All the best. Good luck!!

9. Bookmark this page, so you can come back and see the wonderful advice that you got when you were 12. It will make more sense as you get older.

I'm 27, and learned this over over years, but still I needed to hear this again! And I'm sure I will need this advice again. Bookmarked :)

Would you please provide an example on how one proceed with T shaped learning and expertise?

That's a good question. I left that intentionally broad so as not to prescribe a certain type of learning system or self-education.

There's a lot of ways to learn anything, but I would say that the first step is pursuing a variety of subjects that you find interesting, and even some you think are boring. A lot of seemingly boring things I've found turn out to be quite fascinating when it comes down to it. (e.g. database design)

Over time, you will pick things up, and put them back down. You will return to some, and stay away from others, but eventually you will build a repertoire of knowledge and experience in a variety of different fields. I would also bet that at least one subject will interest you so much that you keep coming back and in which you will ultimately develop expertise.

Experimentation is the name of the game!

Thanks for replying, Giles!

I see just dabbling in whatever interests and experimenting with them takes one forward.

What you wrote is profoundly impactful.

I work in cloud, and one of the things I look for when recruiting solution architects -- another role where broad is a pre-requisite and some narrow depth is expected -- is systems design skills. Besides whatever you study and learn how to do as a programmer. these days it's not often commercially applicable unless you know how to make it run well in the cloud, with all the caveats that apply.

Among the things that aren't common knowledge among basement programmers: network configuration (VPCs, load balancing, CDNs, security), security/IAM, non-relational datastores, ML model development & training, and the list goes on and on and on (microservices & containers, devops, serverless, advanced logging/monitoring & problem solving, high availability, HPC/grid, blockchain, and oh so much more.

The point I'm trying to make is that you can become a T shaped programmer by continuing to study and learn various programming skills, but you may want to alternatively consider becoming broad in ways that expand your value beyond just coding.

Thanks. That's valuable info indeed :)

A phrase I came up with that I like more than "T-shaped" is "molecular skillset". A chemist would probably laugh at me but I think of it as developing seemingly independent skills (molecules <--> skills, experience, knowledge...) that end up forming bonds with other skills, sometimes in surprising ways, that together prove really powerful and can bond up even further. "T" and other tree/tendril-ish shapes are then just special cases of different bond arrangements. You can acquire skills haphazardly (though as a serial dabbler I'd caution against actively pursuing the dabbler's path without a ton of introspection), you can also take the efficient road like a university sequence where most skills are visible and have unsurprising connections to the next skills. You might have to take some seemingly unrelated course too, but you can look at those optimistically as another potential source of surprising bonds. Duds are of course possible -- I have some useless things in my brain that I don't think will ever connect with anything useful, at best they might aid in social signaling one day. (A lot of education is just signaling to show you can talk about a lot of stuff...)

Trivial example: at some point around age 17 I installed Gentoo successfully and learned a lot about gnu/linux (and some bsd) in the process. Lots of little skills formed and connected, and they continue connecting with other things to this day. Got my initial hook on vim then. Saw how nicely colorized my terminal prompt was and that opened the door to many bash customizations. One of the skills I had to learn was to understand how hard drives worked (at least to the ability to successfully partition, format, and boot them -- I did accidentally wipe a windows xp partition the first time). Some years later (age 20?) I had accidentally rm'd a couple weeks of java work without having committed it yet. But prior knowledge of the system let me know the data was still there, and I was able to recover it by scanning the disk device partition directly for the bytes. (Ended up using "grope.tcl" referenced here http://wiki.yak.net/592 which was a lot more efficient than what I was making.)

That’s way too exciting !! Really fresh perspectives.

I once tried upgrading my Windows laptop while having very less space in C drive. Windows totally crashed. After that I used live bootable Ubuntu, but would still like to retrieve all of my windows data.

Do you know how can I do that?

https://www.cgsecurity.org/wiki/TestDisk might be able to help you, it helped me once when I tried to resize some partitions and ran into problems.

Many thanks. That’s awesome!!

+1 for not recommending build systems

9. Use webpack!

This is not just good advice for a successful programming career, but if you can pull it off, pretty much any field that one is passionate about.

As a scientist, if one is a scientist and has abided these ideas, they'd ramp up the food-chain pretty quickly.

This is awesome liam. im just over twice your age and am just now writing game of life simulators.... man, what was i doing at 12???

Trust me, there can be downsides to starting early. In my case, I virtually skipped having social interactions not on the internet until I was around 20 or so. Not everyone will have this kind of affliction, but you can't have it all - you gotta choose what's important to you. I hope this doesn't come off as being too negative, I mostly say this because I think you should try to appreciate what you did learn rather than lamenting about not being able to start with coding super early.

...that being said, this is way cleaner than any code I'd written at 12 I'm pretty sure.

Oh definitely I agree. Gotta find that middle path.

> I virtually skipped having social interactions

That was me, too. It was a trade-off for me. I was reading Hawking in primary school and there was no one to talk to, so I was feeling lonely.

Programming was my lifesaving straw, but I don't wish any kid the kind of isolation that I've felt in my youth. I had to learn many of the social stuff later in life so it evens out IMO.

With that said, it can also have the opposite effect. Success in online interactions at such an age gave me the confidence I needed to be successful in irl interactions.

I guess I'm the only one here doubting that the combination of both this project and the language used in the comment replies is coming from a 12 year old.

Take it as a compliment if it's legitimate.

That makes two of us. My first thought was "Why would someone lie about being 12?", but then i remembered this is the internet. I would ask for proof, but it looks like everyone has already made up their mind or doesn't care.

OP is actually a dog

Probably not the only one, it's very common for people to be extremely prejudiced against the young. As with other prejudices, of course, it's wrong and only scores accurate evaluations by chance. The only thing which prevents any person, of any flavor or age or background or whatever, from producing such work and comporting themselves in such a manner is not having learned it or not practicing it themselves. Anything blamed on 'age' is normally just an indictment of the society that hasn't provided appropriate education or incentives.

I was a precoscious kid myself, and took to online communities with vigor specifically because it permitted me to interact with adults as equals without them constantly dismissing me without basis apart from my age. I've not forgotten that, and if anything society now is even more egregiously bigoted towards the young than ever. Next time you read an article about "kids" or "teenagers" or "adolescents", try mentally substituting in "women" or "African Americans" for the group identity they're talking about. You might be surprised how identical the claims made, and the weak support given for them, are.

No, it's not the same as doing that for women or black people.

People lie about their gifted child to get them exposure all the time. People doing that with art comes to mind. Or even just a parent doing assignments for their kid.

You're not being creative enough if you can't imagine a world in which someone would lie about something on the internet for attention.

Hey man, great job! You got some great advice here from the HN crowd and I'd just like to add one more: blog.

Set up your own site and blog about your journey and the lessons you've picked up along the way. Learning is about working and reflection. Blogging is a good platform for that. The very act of writing the content itself is reinforcing your knowledge. Dive into the work, research, reflect, stay hungry and repeat. That's what it's all about. Apart from that, you'll be expanding your network as well. Having some amount of folks following your journey is not just a good way of inspiring oneself, but also gives you an opportunity to inspire others as well.

Anyways, keep it up! You're definitely going places.

IANAL and all that and it doesn't bother me, but the first thing I thought of was the age related ToS restrictions from the HBO Silicon Valley show.

From the HN "legal" page:

> If you are under 13 years of age, you are not authorized to register to use the Site.

https://www.ycombinator.com/legal/

I assume no 12 year olds actually try to follow this boilerplate on the internet. Do sites that require a DOB to sign up reject those under 12?

I just assumed no people actually try to follow this boilerplate on the internet, glad to see the this knowledge is being passed down through generations.

HN doesn't take PII (just a username/password), and doesn't take a birthdate either for verification, so COPPA probably would not apply.

If he's not authorized to register on the site then the ToS can't apply to him, can they? At which point he can register, but then...

The legal restrictions are not actually on Liam, or any user. They are on the site itself. The law in the US prevents the collection of any personal information from anyone under 13 years of age. Once you are 13 years old, you are a trackable, surveillable, profilable, saleable commodity. Prior to that, though, your personal info is protected from collection.

But... HN doesn't actually collect any personal info. So, that's likely either a reflex inclusion of such a clause or just an abundance of caution.

Good job! I mentor college CS majors whose code is not as clean as this. Seriously.

I would also recommend checking out The Coding Train. NYU ITP Daniel Shiffman's lectures on creative coding. It's time to get started with graphics ;)

https://www.youtube.com/channel/UCvjgXvBlbQiydffZU7m1_aw

And to get a headstart in Artificial Intelligence, from the same author comes the ml5.js library: friendly machine learning for the web. Best of luck!

https://ml5js.org/

I like how the array named 'map' reflects our original use of this word, rather than the special meaning which it has acquired to programmers.

In Javascript and other languages 'map' has become a kind of "functional operation" so we cant call things 'map' anymore (in public). We have to say 'space' or 'state' or 'grid' or something... but I do still think 'map' was good.

Congratulations on a very efficient and readable project. Bravo !

Programmers got it from mathematicians. I had the impression it was a fairly long standing use of the word.

Mathematicians got it directly from cartography. A “map” “maps” points from one set of coordinates to another, for example when you want to plot latitude and longitude on a flat paper.

I expect ideas of mapping functions are only a couple of hundred years old at most, and really quite obscure to most people who are just familiar with the concept of 'maps' and 'mapping' things 'out'.

In functional programming I understand map as just a clone and convert operation. I noticed C++ has a 'map' data structure which is most suitable for associating references. A C++ map is not suitable for storing positions in space.

I'm particularly fond of this because at 12 when I was learning to program I never thought it was important to keep my work safe, or even documented at all, and it all exists on a hard drive in landfill now.

If you pursue a career in programming, you'll look back at your early work and have a great personal self-documentation tool to remind you of how far you've come.

Everything about it is optimism-inducing on a very deep level. The post itself, the reaction, the answers, the discussion.

Well, you either congratulate, or you get downvoted for not being nice. Your choice here on HN.

It's amazing that at the same age Steven Wolfram himself had written his first 100 page book "concise directory of physics) and at just one year older wrote his first book on quantum physics.

I first thought this was a joke, but he did indeed.

"At the age of 12, he wrote a dictionary on physics. By 13 or 14, he had written three books on particle physics. They have not been published." https://en.wikipedia.org/wiki/Stephen_Wolfram#Early_life

From all the amazing things Mr. Wolfram has done at such a young age almost makes him the Chuck Norris of science.

Catching the disease of cellular automata isn’t a good thing. Fredkin was also affected. This young person has many years to do useful things.

Unfortunately it was cancer from the radiation of atomic bomb tests that killed John von Neumann far too early, not his experiments with cellular automata.

https://www.hnf.de/en/permanent-exhibition/exhibition-areas/...

https://en.wikipedia.org/wiki/Von_Neumann_cellular_automaton

https://en.wikipedia.org/wiki/Von_Neumann_universal_construc...

It's really not all that unusual. History is littered with amazing minds that accomplished tremendous things well before adulthood. The modern fiction that humans are useless until adulthood is one we have to construct by force, through active neglect of education, denigration of intellectual pursuits ('Go outside! Get a life! Do anything else but read or sit and think!'), and adamant insistence that it is normal to be ignorant and incapable until adulthood. Pascal recognized his triangle as a youth, Gauss was similarly known for early feats, etc. Those are just accidents of history, mostly, that they are remembered. The actual number of fertile and productive minds is doubtlessly orders of magnitude higher. It's only through diligent refusal to believe, or to tolerate the notion that maybe we should have expected a bit more of ourselves in our own youth, that we can keep up seeing intellectual achievements of the young as an unusual occurrence.

I feel bad for all the other 12 year olds whose parents just saw this :)

or all the 16 year olds

Or all the 42 year olds, ahem, who are currently learning JS/Node/React. :-/

Speaking of which, I'm also learning Spanish. Liam, I'm from the UK where we only speak one language. The one thing I really, really wish I'd done is to have learned another language. I'm doing it now, but it's so much easier when you're young.

It doesn't even matter what it is – pick one you like the sound of. French, German, Italian, whatever. When you bump in to someone later in life and you hear them speaking another language and you can interact with them, there's no greater satisfaction.

/off topic

It's almost like a programming language is a language...

It's great to be writing code at 12 if you want to write code, and clearly this kid has a talent for it, but that doesn't detract from what all the other 12 year olds are doing. So long as they're happy and safe to do what they want to do it's all good. Growing up is not a competition.

Also, while it may seem great to learn early, I've worked with people who didn't write a line of code until they were in their 50s and they were far better at it after a few years than I am despite having 20 years more experience. Some people have a gift. Others have to really work at it. Sadly I'm in the second group. :)

What about those of us in the 3rd group - started late and have to work hard at it!

Any parent who thinks they have a place in determining what pursuits their adolescent child is interested in is in for a wakeup call...

This is so great and I am really impressed, and even a little inspired! When I was 12 there was no internet and learning programming was really hard, but it was more fun for me than anything else. Your project has reminded me how I felt so many years ago when I was 12 and I would sneak out of bed to program QBASIC on my APPLE 2C. Thanks for the inspiration and keep up the great work! Keep working hard like this and you will become a master programmer of the first rank.

Congratulations!

Are you familiar with code golf? If not, I recommend it. In Code Golf, the objective is to complete a programming task in the shortest program possible, measured in bytes. While doing so, you really learn about the nuances of your language and its obscure features.

When I was about your age, I joined the code golf community at Stack Exchange, and since then my JS skills and general programming skills have improved greatly. I hope that you would find it to be useful too.

Congrats.

> dimensional cellular automata in witch all possible rules are encoded into one byte as described here

Small typo here - should be "which" instead of "witch"

Thanks for pointing that out. Fixed.

Hi Liam! I'm your occasional HN neighbour I live currently just below your post :) (Show HN: A smarthome dashboard concept using zircle-ui).

It's a pleasure to see the impressive reception that your post is having. Also I want to join all the voices of congratulations you have received for your work at your age! Keep doing this kind of things, It is very inspiring for all!

Nice work. I'm not a JS expert but I think I saw a bug.

Inside of animate() you are calling setTimeout() to automatically call animate repeatedly but this is also calling setTimeout() every time animate runs but setTimeout() should only be run once.

You should be able to take

  setTimeout(() => {
    animate();
  }, 100);

and paste it over animate() on the last line

What you mean is setInterval and does work well for synchronous execution. His approach is correct and would even work if the animate function is asynchronous as the next tick gets scheduled at the end of animate function execution.

Ah yes, you are right. Got those mixed up.

Recursive setTimeout is a great way to allow for easy timing/speed changes on the fly.

good point :)

That's really awesome. Here it is running in a Node repl in the browser if you want to try it quickly (the code verbatim but hardcoded the rule number to 18) https://repl.it/@amasad/wolfram-cellular-automata

You're off to an amazing start!

I'd say just find stuff that genuinely interests you and try to build it. I got my start focusing on games, but whatever gets you excited. You're already way ahead of the curve, so have fun with it, experiment, and you'll do great things :)

Nice work! I hacked together something similar for the browser after watching a FunFunFunction video about automata, https://brianzelip.github.io/automata

(not ideal on mobile)

Off topic for the main achievement :)

But: if I open the bitbucket link on mobile, the left part of the text is clipped and it doesn't allow scrolling left.

With "request desktop site", it works perfect.

Why does mobile mode even exist if the desktop site works better??

I tend to read my own code in bed on the phone after I commit.

The new Bitbucket interface really dropped the ball on this. It is almost unusable.

I hope they’ll fix it...

Send a request ...

This has suddenly became the most wholesome thread I've ever read on HN.

This is why HN is great.

This is really amazing, reminds me of my childhood days..

I showed this to my girlfriend's sister whom has shown interest in programming, today I learned that she decided to drop her art classes in favour of some programming courses!

Well done!

That’s so cool! When I was 12, the best I built was a game of Who wants to be a millionaire in Qbasic and this is whole another level from that! You are inspirational!

Great work :-) I had fun programming Basic and Z80 assembler when I was 11 as well. I always remember it as being a great time of discovery and passion.

Just stepping by to say, congrats for shipping, congrats for posting ! Im 32 and will be posting my public beta next week !

Congratulations Liam. This is great.

not sure why the age matters?

Congratulations Liam!!!

You're doing phenomenal. You'll be role model for up-coming generation.

Great!!

Congrats! I have shown this to my 13 and 12 year old sons to inspire them.

Well done little one. Keep it up and hats off to your parents.

Impressive. Keep it up!

Keep it up, this is exactly how great programmers are made!

This tells you that talent can be discovered at any age.

congrats! when I was 12 I was war dialling BBSes and pirating shareware games. hopefully your future turns out bright.

Great! Now redo this in Python and GoLang :)

Wow, you really had great potentials in CS!

You are amazing, keep learning more. :)

Thank you Lia. You inspired me today.

Well done mate, very impressive.

Very nice.

Holy $hit! Nice work man!

Holy $hit. Nice work man!

Very nice work!

its brilliant, congratulations!

Great work!

Cool!

Good work!

very inspiring!!!

Excuse my portion of scepticism, it's not the fact that a "12 year old" built this, sure anyone can learn how to code and we can't know how much help this person got along the way. But it's rather the person's language when writing plain text, here in the comments for example. It doesn't feel adolescent but more grownup to me ... But maybe it's just me realising my own shortcommings as a 12 year old, as I only was playing with my friends and the odd computer and video game.

Do kids really do their own science fair projects? We have different rules of ownership for age-appropriate levels of mentorship.

Also, I and likely you come from a different generation (I'm aged late 30s) where my parents knew nothing about coding and I genuinely had to learn everything on my own reading books at a library when my parents would drop me off.

I only had access to a very expensive family computer where my parents would freak out if I changed something on the fancy appliance they poorly understood and essentially every language other than qbasic needed a pricey proprietary compiler costing hundreds of dollars.

These days with the accessibility of the internet, explosion of open source, craigslist laptops perfect for learning going for under $75 and the commonality of programming as a stable career, it wouldn't surprise me to find children living in say, the silicon valley with programmer parents who successfully get their child excited about programming early.

It's a different set of circumstances and a 12 year old able to write graphics demos doesn't sound that implausible.

I think that to post here, is a good sign. Show HN wherever you are able to do, alone or after a 100M financing is one of the reason forums like this one exists. The other is to just share your code and meet digital people.

> It doesn't feel adolescent but more grownup to me

Showing skepticism is also fine, our industry is not just nice code review and beautiful moments where people want to encourage you.

I was extremely lucky and got Internet access in 1990 (pre-web) when I was myself 12 years old. I was entirely conscious of the fact that if the adults figured out that I was 12, I was dead meat. They wouldn't listen to a thing I had to say, and wouldn't feel the need to defend their claims against any argument from me. So, I spoke well. It's not all that hard to do. As such, I can't recall a single time I was ever called out. When I did on occasion choose to reveal my age, I often encountered disbelief, but by then it was usually after I'd already won them over so they could overcome their bigotry. And that's really what it is, I've realized in the years since. It's just simple 'they're not like me, any difference must be negative, it must be rooted in what they are and insurmountable, I will always be above them' bigotry. Despite the clear fact that if a 12 year old is taught to install a kitchen sink, something I'm sure anyone could see as possible, then clearly they'd be superior to most adults who don't know how to install kitchen sinks. So why is anything else different?

Technically, you shouldn't be using this site ;)

> If you are under 13 years of age, you are not authorized to register to use the Site.

But I applaud you for getting into programming this early.

When I was 12 I ignored this on pretty much any website I signed up for. Words do little to stop a determined 12 year old.

On the internet, nobody knows your a dog.

Nice work dude!

11 year olds follow the rules!

Great work. I will give you a different piece of advice. Make sure to work out 360 out of 365 days in a year, without weights.

Ok I'll bite....where is this very specific advice coming from?

From a guy that was doing cool things with a computer at age 12 but instead of starting to workout at age 12, started at age 18. Any age is good to start working out but sooner is better.

What bizarre advice for a programming thread. And what are those five days for?

spring break

Sick days. Health is the most important thing.

wow you code better than me

Good work.

For a next step, I would suggest giving a try at breaking it up into multiple files using import/export statements (it doesn't need it but it's good practice), and using a bundler like Parcel to build the finished file.

    npm i --save-dev parcel-bundler

...then move your source JS files to a folder named src, and add to package.json...

    "scripts": {
      "prepare": "parcel --target node src/index.js"
      "start": " node dist/index.js"
    }

In your individual files, you do like...

    // file-1.js

    import somethingName, { whateverName } from './file-2'

    // file-2.js

    export function whateverName { ... }

    const somethingName = '...';
    export default somethingName;

...and then run...

    npm run prepare
    npm run start

(The prepare script is also a "magic word" that will auto-run anytime someone does 'npm install', so usually an end user won't have to manually run it.)

A bundler isn't strictly necessary if you plan to only run something in Node, but it lets you make a package that's usable in both Node and web projects, and it makes it simple to use Babel, which lets you Javascript features from the future that aren't actually in Node proper yet.

That last thing a 12 year old needs is a complicated build toolchain.

Everybody needs a build toolchain eventually, and learning it early saves the eventual "this webpage has become a 5000-line JS file and it's incredibly unwieldy to maintain" hassle with a more complicated project.

Hell, I'm 50 and its the last thing I need.

It's a program that fits in one screen and has 3 functions. No need for these trendy overcomplications.

The fact that it's dead simple is the point of the recommendation. It gives an opportunity to learn the tool with the bare minimum of moving parts involved.

I'd prefer that we unlearn many of these things instead.

Unfortunately, my stock of browser-maintainer mind control rays has run out, so until somebody scrounges one up we'll still need ways to make the web half of the JS equation at least vaguely maintainable.

How about we let the kid learn how to program (which will outlive your favorite bundler and file hierarchy anyway) first? It's a 100 line script, one of the mistakes I made learning how to program was trying to be like the adults and focusing on minutiae like code organization rather than good code.

EDIT: story time. I once tried to make a 2D game engine in my late teens, but I wanted to be like the adults and have a proper class hierarchy. Suffice to say, I never finished it. Come to find a few years later, now class oriented OOP is passé and I was wrong to waste that time on it.

Learning to code and get things working is more cool (and in the end, more profitable) than making your code fit the norms of the day, which won't last anyway.

My intent is less "code structure", and more "most of the pieces you need to make a Node project also work in the browser". Imports/exports and Parcel would get liamilan most of the way towards being able to do...

   if (window) {
     document.write(...)
   } else {
     console.log(..)
   }

...and have it just work, without having to write everything twice.

Of the things to learn when one is starting out, I would not put code modularisation and toolchains at the top of my list (especially for something so small).

Additionally, this runs in Node, so you don't need to compile the source files with `parcel`.

im 12 and what is this

me too thanks

Reply on news.ycombinator.com