Ask HN: How to integrate coding into primary school?
Hi everyone, do you have any ideas about how to integrate coding into primary school? What's the best way to start coding with young children? Thanks for your help!! Why there's a need for children to learn coding? This is getting really crazy. Seems like everyone needs to code and everyone is pushing towards that. I don't see this happening for plumbing, to give you a short example, or any other professional area. EDIT: I was a coach at a Django Girls event. I support these kind of events, but still, I don't see the reason to really push children to learn coding in their 5 or 6 years old. At that age, they are supposed to play with other children in the backyard, getting dirty. We are losing that. Probably, my generation (1990) was the last having that joy. It's a good question. Because programming isn't plumbing. Programming is not just a skill, it's an expressive medium required for effective communication in the 21st century. Back in the day, you had to be literate. You had to write in cursive. This wasn't something you went to school for and did as a living, it was just a baseline from which you were able to do anything else. Programming is like that. Why folks think of it as a stand-alone skill is beyond me. Yes, its basis is math. But that's like saying that the basis of writing is painting. The purpose of writing is to enable other kinds of work with lots of people, some of which may be far away. The purpose of programming is to enable helping lots of people do repetitive and boring things from all sorts of problem areas, some of which you will never meet. >Programming is like that. No it's not. Making YouTube videos is like that. Knowing how to do social is like that. Being able to look stuff up on Google is like that. Graphic design is like that. (I'm often amazed by the quality of the graphic design and motion graphics on amateur YouTube channels.) Knowing how to tell when politicians are lying should be like that, but isn't. Programming is engineering. Not everyone is an engineer. Not everyone can be an engineer, because a lot of people's minds don't work in an engineery kind of way. I think it's fine to teach kids Scratch and maybe some Python. But I don't think it has anything to do with effective communication - it's just useful practice for basic logical thinking and problem solving. The kids who are good at it (maybe 10-15%) and enjoy it (maybe 25-50%) can specialise and do the harder stuff. In any case I don't expect the Web of 2035 to look much like the Web of today, so there should be no expectation that learning any specific language or system is going to be useful later - any more than it made sense to expect everyone would be using BASIC and CP/M in 2000 when we started seeing computers in schools around 1980. It works the other way around
>Graphic design is like that.
Graphic design is art. Not everyone is an artist. Not everyone can be an artist, because a lot of people's minds (or "eye) don't work in an artistic kind of way. But getting the basics of Graphical Design can still help both appreciate a good design and "understand" why "that stuff" feels ugly, and being able to discuss with a graphic designer and talk the "same language" Same thing with "Programming".
No one claims that 100% of the population should understand Rust borrow checker or functional purity But everyone should understand (and be able to copy-paste some code and "fix" it until it works) the concept of a sequence of instruction, a loop to generate repetitive stuff, the concept of variable (and therefore template and mail merge), a if (and therefore be able to do basic stuff in Excel).
So they can talk to engineer and get what they way.
And maybe everyone should be able to do FizzBuzz or display the number from 50 to 1 with a loop going from 0 to 49. At least in pseudo-code. But my point is that a lot of people who aren't trained graphic designers, know nothing about the history or practice of graphic design, and have probably never studied art are able to produce cool designs that are close to the standard of professional work. And also that being to do this is a huge personal and business benefit. Being able to code gives you - what? If you're not doing something useful with code - probably professionally - it's not a communicative, practical skill. Office, much as I hate it, kind of is. There may be some intangible benefits. But so far as I know there has been no research to suggest that learning to code improves personal, social, academic, or professional outcomes at school. Meanwhile there's a lot of research to suggest that learning a musical instrument or a second language has obvious measurable benefits. Obviously I'm not against coding. But I'm definitely against any mythology of coding that suggests it's a key literacy skill - because based on real evidence, there a lot of other skills with a better claim to that. let me clarify myself.
What I see as a "key litteracy skill" is what some calls "Computational Sense" ("A familiarity with the capabilities of computer applications and the ability to easily grasp the difficulty in implementing a computer-based solution. Typically acquired by learning a programming language.") aka https://xkcd.com/1425/ So the "learning to code" is just a mean to an end. Also I do think that learning to automate the boring stuff (https://automatetheboringstuff.com/ ) improves at least personal (e.g. do your own taxes in Excel, understand AND and OR so you filter email better), social (e.g. ifttt), academic ( e.g. R ), AND professional (e.g. everything) outcomes. >learning [...] a second language has obvious measurable benefits and don't you think that the ability to analyze a process and transform it into a sequence of instructions can have measurable benefits too? You wrote "it's just useful practice for basic logical thinking and problem solving."
I agree. It is only that I would remove the "just". I like what Papert wrote: " debugging is the essence of intellectual activity". >And also that being to do this is a huge personal and business benefit. All the "non technical" startup founders looking for/lacking a technical co-founder,
you don't think that they would benefit from learning a bit of coding so they can a) estimate what needs to be done (Computational Sense), b) hire the proper people c) specify what is needed ? Obviously, I am against coding as "learn js in 5 days" but not being able to FazzBazz(2,7).
I am for learning what a computer can do. Cursive writing is and was less a criterion of literary and more a social signifier. It is an encoding that makes things less intelligible and the luxury of choosing a less intelligible form is a mark of social status, e.g. a doctor's prescription is stereotypically poorly penned but the social status disparity meant nurses, pharmacists and patients just had to deal with it. Programming since the demise of Cobol has had a similar hermeneutic strand. Languages designed for intelligibility by ordinary persons are limited in expressibility and denigrated by the community of programmers, e.g. HTML isn't deemed a real programming language and its users aren't deemed programmers. Even though HTML is one of the more directly applicable languages to ordinary school day tasks it is eschewed for real languages and the objective becomes making baubles rather than layering in a communication technique on ordinary tasks. HTML can be integrated into the curriculum. It's forgiving...system failure is in the eyes of the beholder not the god of Booleans. And most importantly, HTML is a rational extension of many Indo European languages as punctuation and annotation. Definitely don't detract from kids getting a chance for physical play, preferably outdoors. That should be an axiom. With that said, I don't think it should necessarily be pushed as a career skill, nor necessarily even as an academic discipline, but why not just for fun? Most "educational" use of computers at the grade school level is a waste of the kids time. I'd rather kids sit in front of Scratch, than some glorified flash cards. Maybe the younger kids are, the less academic it needs to be. For older kids, my view is that computer programming is becoming analogous to a liberal art, i.e., something that every educated person should have been exposed to in order to be an informed member of society, even if they don't pursue it as a profession. My parents took me to music lessons throughout my childhood, for which I'm grateful. They also let me spend a lot of time playing outside. Edit: Kids are under a lot of pressure to choose "careers" before experiencing the subject matter, e.g., in engineering and computer science. Perhaps it would be a service to let the kids who don't like programming, find it out before they hit college. Why learn mathematics or science? If kids are good at abstract problem solving, denying them access to computer programming is like denying them sports if they like to kick balls around. Coding will be more and more the killer app that people have in their jobs. A person can crush many ordinary office jobs just with excel and word macro skills. I teach people simple excel forumulas like vlookup and their life changes forever. We should also teach plumbing, electrical and financial skills, and cut back on the interpretative dance, music and drama for kids who hate that crap. > If kids are good at abstract problem solving, denying them access to computer programming is like denying them sports if they like to kick balls around. Hear hear! We should be identifying students' aptitudes and interests and ensuring they're getting the best training possible towards up-skilling them these inclinations as early as possible. Not everyone follows the same path, and the disaster is that we make others feel bad for not following the mainstream path laid out in front of them. I'm one of those who tread from the path and nearly fell of the cliff, but managed to find a way with software in my very late teens, just in time. I can't imagine and almost don't want to think about those who never found a place in society as a failure of early education methods... I would make the same argument as you but in the opposite direction: everyone has to deal with plumbing at some point in their lives. Schools ought to teach basic proficiency in plumbing. It's useful to know the basics even if it only helps you recognise the skill that you need a professional for. Kids do /play/ with water in schools to some extent - whilst not plumbing it might be considered proto-plumbing in the way that the basic Hour of Code (http://studio.code.org/s/hourofcode/reset) is coding. A. It facilitates teaching some useful skills in approaching problems. B. Schools don't need to buy equipment. C. Anything that means running the MS office suite less is a win in my book. I help out at our local CoderDojo. I don't know how I feel about saying everyone should learn to code, but the main thing I get out of it is that the kids really love it. They use Scratch, they mostly make games, it's all around awesome. This shouldn't be a surprise to me or most people here, I guess, since I also love programming. Why wouldn't I want to share something I love with kids? Because programming is the most direct way to communicate with a computer, and computers are highly important in our lives? I suppose you could say the same for plumbing, but programming is in a league closer with reading/writing/math than plumbing, in that it applies to creating such a wide range of art and science. And it's not something bound to technology, the concepts of programming were mathematically sound well before the earliest home personal computer was built. Agree and disagree. The focus should not be on "code" but exposing them early on to problem solving: https://www.youtube.com/watch?v=mvK0UzFNw1Q If coding is the way to go to solve a problem then be it. If plumbing is the solution then that is what they should learn. The key is to empower them to solve problems early on. The good news is that that can even happen in their backyard getting dirty. Back in 1991, when I was 7, the Logo programming language did great for me. You have to write instructions for a pointer to move on the screen and draw things ("go ahead 20 steps; turn right; go ahead 10 steps...). We had a Logo class twice a week and we all kids couldn't wait for it to come. I bet some flavour of Logo is still around, and "turtle graphics" (the general concept of procedural drawing for kids) is a great way to introduce programming. https://en.wikipedia.org/wiki/Logo_%28programming_language%2... https://en.wikipedia.org/wiki/Turtle_graphics Scratch from MIT is another canonical suggestion when it comes to this matter. Some teachers are using Minecraft to let kids build imaginary things; I guess the underlying principle is "let kids plan ahead", which is the core ability for algorithmic thinking, thus computer programming (as in this board game http://www.robotturtles.com/ , ispired by Logo; it's for pre-school tho) I'd also check out "CS unplugged" http://csunplugged.org/ , a collection of open-air activities you can make that sneakily teach algorithms. My kids started with 'beebots' in their first year. This is simple logic - go forward 1 space, left 1 space, etc. My oldest child is still in primary and they have a lego robotics club- they build simple robots and do simple programming. He does well at maths and is a lego wiz so has been allowed to join the club 3 years early. I've been thinking about where to start and they love minecraft, so it makes sense to start there I think. I'll check out some of your other links as well. There are minecraft mods where you programme bots to do tasks for you - I like the sound of it but haven't tried it with my kids yet; it was on HN a week or two ago. Few links: http://www.computercraft.info/download/ http://www.learntomod.com/ (paid) I was 7 in 1993, and did some Logo programming. It was the final puzzle in the educational adventure/puzzle game "Crystal Rain Forest"[1] for the Acorn Archimedes. The puzzles started out with very simple programming (so simple it doesn't look like programming). Just setting a single variable, or two. It progresses to programming a one-dimentional robot (forward/backwards), then finally using Logo with loops. I was 10 in 2001 and also did lots of Logo programming. Same thing with institutionalized classes twice a week. It wasn't what got me into computers, but definitely what got me into coding. I'm thankful every day. https://scratch.mit.edu/
It is made for teaching code to young children We've been using scratch in a local code club for children and it is a great language for that purpose. You can teach the general flow, logic, and what goes into making a computer program without the "scary code" part. At what age do kids tend to start with Scratch? I am coteaching an afterschool club - 9-11 yrs. 12 boys and 3 girls. I am not a professional teacher so the materials on the MIT site are very helpful. In addition I have a NoStarch book which I recommend. My last class: https://scratch.mit.edu/projects/61701790/ Very rough but meant for demonstration. Was trying to show 2 things - how a loop works and one way to use simple math to change xy position. My spiel is in the notes. It is fun to teach this. We try to keep the initial demonstration down to 10 mins and let them try for the next 50 mins. It is productively chaotic as they ar ehelping each other and calling out for help from us (the teachers). The good thing is that it is very hands on. The kids have chromebooks in the classroom and the scratch.mit.edu site is pretty good as an "IDE". The kids save progress on the chromebook and then upload it to their google drive. From the site :
"What is the age range for Scratch?
While Scratch is primarily designed for 8 to 16 year olds, it is also used by people of all ages, including younger children with their parents." We saw children from 6 to 12 come try coding with scratch at our school during a coding for children event Code Club really is an excellent organisation, with some first class resources. Came here to say this. Don't reinvent the wheel. Not sure if you've seen this, but code.org created a teaching manual just for it. They also regularly run workshops throughout the United States. There's also this in the UK for children aged 9-11, you can volunteer to go into primary schools to run a club (teacher assisted): I do this. The curriculum is all available on github[0]. I think the first semester (Scratch) is really good. Personally I don't like the python and html/css stuff later. The material is quite old fashioned, using terminals and rather dry web pages. It's hard to get kids interested in this after they have spent a few months knocking out colourful and highly interactive games. I co-teach a local after school club using Scratch. Kids are 9-11 and they are very much into it. The school has in classroom Chromebooks and we use the online scratch.mit.edu .
That works nicely - we have an hour and try to teach some basic point in the first few minutes and then the kids are off. It is a noisy, fun process as they kids help each other and ask us how to do things. My coteacher is a CS professor at a local engineering university. Its a blast. Scratch allows them to make little games and animations. They honestly seem to pick it up themselves. Another resource that I like is NoStarch press which is an imprint (right word?) of O'Reilly. They have great instruction books focused to kids and to adults teaching kids. I have a Python title as well as a Scratch and a javascript one. Coincidentally listened to the most recent episode of "Talk Python to Me" http://www.talkpythontome.com/episodes/show/8/teaching-pytho... Professor Curran from Sydney was the guest and he has a site called Grok Learning which takes kids through a Pythin curriculum. He argues that by middle school kids should be taught a text based language. He says in the classroom he just uses the Python interpreter in a terminal. He aludes to the fact that it is a productive exercize in that you make mistakes and show the kids a process of making mistakes and fixing them. I liked Curran's focus on training teachers and his recent contributions to the Hour of Code are interesting. The Eliza bot example is one where you can start with a 'bunch of if statements' and refine it over time as you learn new constructs. I hope I got his points right - TBH I listen to podcasts as I wash dishes and get kids ready for bed so missed some. A good listen though if you are intersted in this topic. When I was a child I learned BASIC and LOGO when I was 5-7 years old. I found the skills I acquired in the process helped me with math and english, to the point where I was doing both at a college level by the time I was in 6th grade. I'm working on a project to "upgrade" classic BASIC versions and make them more interesting to today's children. We have a proof of concept at http://discorunner.com and there's also an introduction to BASIC tutorial on that site targeted toward primary school-age children. I really like the insight from Shriram Krishnamurthi at Brown who tried a similar project. "Teachers are hard to find" - Prof Curran makes a similar point in a recent podcast from "Talk Python to Me" It is training teachers not only coding skills but the process of doing something in front of a class where you are going to make mistakes and fix them in front of a live (and opinionated) audience. What integration of the curriculum usually entails is bringing Language Arts and mathematics into other subjects. By analogy, if you want to integrate coding into primary school then you have to pick encodings that are integrable. Logo, Scratch, Python, C, Cobol, whatever aren't because you can't really use them in Language Arts without a real stretch of the term integration and a lot of pretending that Logo or Cobol isn't a force fit in a language arts class...it's hard to write a poem in Perl that compiles. The way to integrate coding is to pick encodings that extend the ordinary tasks. The obvious extension of Language Arts is HTML because presentation is part of what primary schools teach and HTML is highly accessible and doesn't require any special tools. The less obvious extension of mathematics is J, which is simply an alternative to languages designed for the limitations of 1970's era hardware in the form of hand calculators. Forty years ago, the hand calculator focused on accuracy over hand calculations rather than extending the possibilities of numerical manipulation. But summing a list of numbers should be a basic operation...and in mathematics it is - Σ. Finland's official curriculum will include programming starting next year. This Reddit thread has good information on that. HTH! https://www.reddit.com/r/programming/comments/2ffw1d/program... I would hesitate to make primary schools kids learn 'coding' because personally, I find the underlying problem solving more inspiring than learning syntax. FIRST Lego League [1] offers a compelling alternative though. I competed in FLL with a team of other primary school kids and greatly enjoyed designing a robot (with motors and sensors) and then programming it to solve tasks using a drag and drop language. Each member of our team could focus on the part they enjoyed most rather than everyone sitting in front of the computer. "FLL challenges kids to think like scientists and engineers" and has a glamorous 'World Final' event that is honestly the coolest thing in the world when you're 11. Get a few Raspberry Pi's and start playing around with the projects on their web page. Maybe searching about "The Young Coders" may help: https://us.pycon.org/2014/events/letslearnpython/ http://therealkatie.net/blog/2015/feb/17/young-coders-why-tw... http://pycon.blogspot.com.tr/2013/03/how-kids-stole-show-you... I spent many hours being instructed cursive and "penmanship". So be it - its the essential foundation to all the notes, essays, etc you do in school. Likewise, if we want kids to have a foundation in computing, I'd advocate a computer penmanship course, probably around basic linux commands, basic networking, etc. Primary school seems too early to appreciate coding, so I'd focus more on rote learning and drilling. Boring i know, but I think it would pay off more than learning to drag n' drop to the LEGO GUI. I'm someone who never got that foundation, and it has been a burden on my ability to work at a professional level. I somewhat agree. I worry that a 'fashion' is developing in that it is 'cool to code' therefore all must learn it at all ages. You make a good point that it is not necessarily the best thing to do given how young kids skills differ and you list some really essential basics that I absolutely agree should form part of the curriculum. I'd also add that actually disassembling and rebuilding a computer or other types of electronics should be on the essential list. So many young kids (mine included) are excellent operators of their little black boxes but have no clue what's inside or how it physically works. BUT - all kids learn to write and to read and to count. These are basic skills needed for life. Given the reality of the world, proficiency with computers is becoming equally important. It seems prudent to introduce all those things you mentioned but I think it perhaps does kids a disservice to suggest they are to young to appreciate coding. One of my most notable memories from my early years of primary school is programming a little robotic turtle thing using a BBC Micro and had it completing mazes and drawing maps on large sheets of paper for us. I was probably 6 or 7 years old and it was AWESOME!!! We certainly understood what we were doing and how useful it could be. So, I guess I'm for a balanced approach. Google's CS-First program, http://www.cs-first.com/overview , is great for this. CS First is a free program that increases student access and exposure to computer science (CS) education through after-school, in-school, and summer programs. All clubs are run by teachers and/or community volunteers. I'm a volunteer and I've been quite successful starting them in various schools thus far. When I was in primary school (around 1989), I coded in BASIC (first, by typing some pieces of code from textbooks to see how they work, then altering them, then trying to write something on my own — e.g. tic-tac-toe, or date calculations). The best modern analogue of BASIC (though much more attractive) is probably Processing. Lua also works. Then kids should start to learn Javascript to understand that the real world is not that logical. I had the same experience. Typing in BASIC from books/magazines in the mid-1980s and modifying the code after I got it working. My 7-year-old daughter uses Scratch. My 6-year-old son watches me play code combat and tells me what to do. I explain what I'm doing as I type. I never see anyone talking about Processing. It's probably the quickest way to go from never-programmed-before to having visual results (well, maybe aside from JavaScript). http://openprocessing.org has a massive compendium of shit people have made in it over the years. There is a big project, Computing at School, to introduce computer science as a school subject in Britain. They have a sample curriculum thing which includes learning goals for all years, you may want to look at that. http://www.computingatschool.org.uk/data/uploads/CASPrimaryC... RoboRally is a turn based board game, each player controls a robot by selecting a number of command cards which will be executed by the robot like forward, turn left or right, fire laser :) and so on... Coding / Engineering is about problem solving and breaking things down into component steps. Critical Thinking is required. Don't teach a language teach the skill. Thanks!!!