Slashdot Mirror
Stephen Wolfram: No Need To Teach With 'Toy Programming Languages' Like Scratch (wolfram.com)
theodp writes: From Stephen Wolfram's blog post announcing the Wolfram Programming Lab: "It's a very important — and in fact transformative — moment for programming education. In the past one could use a 'toy programming language' like Scratch, or one could use a professional low-level programming language like C++ or Java. Scratch is easy to use, but is very limited. C++ or Java can ultimately do much more (though they don't have built-in knowledge), but you need to put in significant time—and get deep into the engineering details—to make programs that get beyond a toy level of functionality. With the Wolfram Language, though, it's a completely different story. Because now even beginners can write programs that do really interesting things. And the programs don't have to just be 'computer science exercises': they can be programs that immediately connect to the real world, and to what students study across the whole curriculum. Wolfram Programming Lab gives people a broad way to learn modern programming — and to acquire an incredibly valuable career-building practical skill. But it also helps develop the kind of computational thinking that's increasingly central to today's world." So, when it comes to programming education, are schools hitchIng their cart to the wrong horse?
Guy with profit motive thinks his pricey programming environment is better than one that is free.
Settling into a comfy chair & pouring myself a beer...
Confucius say, "Find worm in apple - bad. Find half a worm - worse."
Obviously no language can be serious without a high price tag and draconian licensing restrictions.
Not so long ago, a story came up on Slashdot about learning the Wolfram language (left as an exercise to the reader to locate).
As I recall, very quickly there were concrete things as a result of relatively easy things ... graphing or plotting or somesuch. And it went quick from there in terms of being able to do things people could relate to, because it's the same kinds of things they kids would learn in school.
So, I guess contrast this to the old fashioned languages we used to learn on like Logo ... ok, I can draw a line. I have always wanted to draw a line. How awesome, a line. (Don't get me wrong, Logo was my first programming, and it was cool.)
Wolfram seems like he's got a much more "results oriented" language for doing fairly practical things, and that it will be easier to see the immediacy of the pieces as it relates to something you grasp.
I wouldn't discount him out of hand when he says this. His stuff is far less abstract, and far more concrete in terms of the kinds of things you do.
Lost at C:>. Found at C.
Teaching "how to program" to the general population isn't about teaching a practical skill.
Just like Math, the point is to get students to understand Logic and Reasoning skills.
Similar to how P.E. class isn't meant to teach children how to play dodgeball, it's about making sure they understand the importance of being active, and know various ways they might be able to enjoy that.
If a "toy language" is more approachable, go for it.
-- 'The' Lord and Master Bitman On High, Master Of All
My daughter (7 years old) spend 10 minutes choosing proper shade of pink for a cat in her first Scratch game. I don't think that Wolfram can even start to compete in same category of fun. She was a lot more interested in possibilities of making things meow or bark rather than trivially connecting her results to per square-furlong gross national product of 10 most polluted cities in the world.
If you are targeting 12-15 year olds, sure, pick whatever. They are forced to learn French, they can be forced to learn any other strange programming language. But for 5-8 year olds, let us play with Scratch. And having to explain that father does bit different things at work than picking between pink colors for cats... he picks between different shades of blue for odd lines in table css... yes, it is kind of a cat for grownups, just square and painted in blue stripes.
I'm sorry, but in the world of instant gratification and kids already used to highly graphical and immersive games from a very young age, writing a program that 'outputs hello world' or having to learn to use some framework or library to do anything 'cool' is not going to engage 5,6,7 year old kids.
For these age groups, these kinds of graphical block programming languages are very much helpful in that they actually engage the child. I have not played with Scratch in a while, but the exercises on code.org are very engaging, very awesome and they even show you the JavaScript code that would be produced as a result. I think they even have some pure JavaScript exercises now too.
It's easy to say that we all learned on real languages and it was good enough for us, but much as I hate to say it, we grew up in a different time with different expectations on what a computer program should/can do.
Asking your carpenter to fit you for braces.
Asking a wood chipper for a blowjob.
Asking a physicist to write you a programming language.
Asking a squirrel to guard your strategic nut reserve.
Guess which one applies to this story?
I think there's a flawed premise, that there must be outreach to those not particularly inclined to do programming of their own volition.
I think that certain popular offshoring destinations demonstrate the result of such a strategy. There are good developers in those geographies, but the signal to noise ratio makes it tricky for a business person to tell the difference up front. I know in my experience, the 'cheap' flavor of offshore developers have been 5% immediately proficient (those will be gone in a month to get a better paying job, whether they move or not, people who underestimated themselves or had to take a filler job between good jobs), 15% will get to that point over half a year or so (and then get a better paying job, effectively those fresh out of not-much-better-than-high-school education and this is their first real world job). The other 80% of the cheap labor that US companies love so much either just aren't wired for the work or just don't care enough. They approach their job with all of the enthusiasm of a retail store stocker or grocery bagger.
XML is like violence. If it doesn't solve the problem, use more.
If I was designing a setup to teach kids programming, it'd be an IDE centered around creating - for example - a 2d game. Have easy, integrated ways to edit art assets and associate them with scripts and inputs. Have easily accessible commands to make sounds, move stuff around, and navigate between "levels". Kids like doing this; when I was a Cub Scout leader, the boys really enjoyed drawing pictures and bringing them to life and figuring out what would happen - but there's no way they would have been able to chain things together (in Haxe, since at the time a Flash game was the most accessible target for the kids to be able to play the final game at home) without me helping.
There's lots of languages that would be suitable for this, and I don't think that's necessarily the important part. The problem with just picking up, say, Java and writing a game is that there's a lot of ducks you have to have lined up before you can get an interesting result. There's a lot of unintuitive steps. When I learned Commodore 64 basic as a kid, I pushed through those challenges because that was the only way I could play a new game. Most kids now won't have that problem of "there's nothing to do on the computer".
Let's not stir that bag of worms...
C++ and Java are both great beginner's languages.
I've met folks in projects, who have been programming C++ and Java for 10 years . . . and they still program like they are beginners.
Hmmm . . . maybe a Fountain of Youth . . . ?
Schroedinger's Brexit: The UK is both in and out of the EU at the same time!
I wish people would stop quoting that. First, it's junk and second it's incredibly out of date. The quote is from 1975.
BASIC in 1975 was not very good.
BASIC by 1982 was much much better. BBC basic (first released in 1982) had simple structured programming with named procedures and functions, local variables and etc. By 1987 BBC BASIC had acquired proper blocks, and was quickly followed by QuickBASIC in 1988. Those are the only ones I know of.
The glib quote about BASIC has been out of date for approximately 30 years.
SJW n. One who posts facts.
Only if you have made enough mistakes in a language like Basic you will realize that there's a reason to code in better languages. Otherwise you will just get bad coders in a different language and nothing would have been achieved.
The language doesn't really matter, it's how you apply it that matters.
If builders built buildings the way programmers wrote programs, then the first woodpecker would destroy civilization.
I taught myself how to program using the BASIC books located in the Radio Shack stores and typed them into the (new) TRS-80's they had out. (Yes I am THAT old) I then moved on to using Assembly (Z80). At the moment I can code in 23 languages, and I think in C so there can be a progression.
Although I completely agree that one needs an introductory language to bridge the gap between language arts and programming, the last time I checked Dice there were no openings for Wolfram programmers. I do however remember all the hype around the various instances of BASIC and I can attest to a large number of VB apps that were written (very poorly) by non-programmers. Coding past an interpreter syntax does not qualify you as a programmer.
I see this entire discussion, including the various calls for CS education in the public schools as yet another instance of what killed my profession: the incorporation of unskilled labor. I am CCIE #12981 and there was a time when having that certification meant I could pull down a well paying job nearly anywhere. Now it almost doesn't matter because so few organizations need highly qualified networking resources. They have farmed out networking to a 3rd party, or they have a few slightly skilled resources that keep the lights on. I see the same thing happening to software development, and we as a culture will continue to suffer under the risks of running poorly written applications, because corporations don't see the need to hire highly skilled developers. Shoving all students into the pot via mandatory CS education, or promoting BASIC languages like Wolfram will only make that worse.
If you want a programmer you don't start with a language. You start with math and specifically with logic. The language used is a mere vehicle for the expression of concepts and as such learning its syntax is secondary. Rather teach principles, such as "Always check your inputs, and your return values" which is true in any language.
Exactly. Same goes for all the boiler plate code just to get things running. We have to start with a bunch of complicated stuff like "int main(int argc, char **argv)" and "#include ". Java makes it even worth by having to declare a class simply to write hello world. Languages like basic are much simpler for beginners because there is no boiler plate code. Hello world is a single line of code, and very easy for people to understand.
Anthropic principle: We see the universe the way it is because if it were different we would not be here to see it.
No you don't.
C++ and Java are both great beginner's languages.
Yeah, that's why there is LabView.
Some people are easily intimidated by text - not just procedural thinking, but actual text - not many of those people will participate in a text based discussion board like /. but they exist, watching YouTube videos, and working in your company. One way of thinking says: so let people who can handle text handle all the programming. Another way says: give them a mouse and a drag and drop language so we don't have to have endless meetings with these people trying to get them to explain something they want in terms that can actually be delivered.
If you give a person something they can work with directly, they can learn the limitations for themselves. Otherwise, they'll often assume your "can't be done" explanations are a form of laziness, or ignorance, rather than a communication of the physical realities of the universe we live in.
Yes... and I think it's important that they understand why. A beginning programming course might do well to teach the basics of what's going on in the computer itself, how code - no matter which language - is changed to machine instructions, and talk about the instruction pointer, simple memory allocation (like what happens when you declare a variable).
My son takes these crap programming courses in high school, and while it included Java (and scratch, and even python), and while he understood some of what the code he was writing did, he didn't understand that the complicated stuff was being handled by some "magic" libraries the teacher gave them, or what was provided in the IDE they were given. At the end of the day, using the programming environment he was given, he could compile a Java program to byte code and run it... but it didn't generate jar files and, even if it did, he could send it to me and it wouldn't work on my computer because I didn't have all the "magic" libraries that actually did all the work.
I actually wouldn't have a problem with that if the intent was to give them an understanding of programming concepts like structures or classes and why they are useful, but no, even though he gets great grades, he couldn't explain why he was creating classes in Java. He new he could use System.out.println(), but didn't really know what it was - he just knew that was the statement to print something.
Stupid sexy Flanders.
Since when Java can be considered a low-level programming language? Even C++ can be questioned as such!
Sent as ripples into the electromagnetic field. No single photon has been harmed in the process.
Ughh!!! I totally apologize to everyone for including the wrong link for Ogden's Basic English. A thousand pardons!
Here is where I meant to direct you: https://en.wikipedia.org/wiki/...
(And now too many people know where I let myself be distracted this morning!)
"The mind works quicker than you think!"
Actually, I disagree it should be about algorithms and data structures to begin with. In fact, it's a terrible idea. Don't shoot for "productively", don't look for "creating suitable models of real-world situations". That's way too much way too fast.
You don't teach a kid to build stuff by letting them build a real functioning bridge. You give them some legos or building blocks and let them run wild with it. Let them realize things fall over, or collapse, or don't fit.
What we need is a way to accomplish small, discrete, achievable tasks ... as quickly as possible, in a way they can fiddle with permutations and see the outcomes, and with as little abstract concepts as possible. At least, not ones which seem like abstract concepts.
You're not trying to make people who are professional programmers ... you're trying to establish "if I do X, Y happens ... if I want something which is kinda like Y, I need to do something which is mostly like X but different in this way".
One of the exercises I saw done in junior high/high school to teach people the root concept was to get them to "program the robot". You have them walk someone through the steps of doing an easy task for a human ... ideally the "robot" understands you can ONLY do EXACTLY what they tell you, EXACTLY as stated, and that you CAN'T interpret for them.
"Go get the book" results in nothing. Step with your left foot, step with your right foot, extend your hand, open your hand, close your hand. I had a class mate who just couldn't wrap her head around this until a teacher and I did this after class ... once she'd done it, she suddenly kinda went "oh, so I need to break this down into a bunch of small steps it already knows how to do". She was never going to be a professional coder, but suddenly she understood the underlying fact ... computers are idiots and can't do anything on their own.
The idea you need to formulate a set of steps, plan it out and describe it correctly to do something is what you're learning when you first get introduced to the concept of "programming". And it's incredibly eye-opening how much people struggle with even that as a concept.
Start throwing around words like algorithms and data structures, and people will switch off LONG before you've taught them a damned thing.
You need to lay the foundation to help people being to grasp what it means to start filling in those blanks, and extrapolating to a more generalized solution. You can do an awful lot of that adevelopers.slashdot.org
What you describe is not suitable for teaching kids, or establishing the concept. It's for people who are going to be pursuing something much more rigorous and formalized.
Lost at C:>. Found at C.
I learned to program in college before computers were found in the home, starting with Fortran. And, I could do it, but it didn't really come together for me till I learned assembly language. In class, the teacher started with a very simple model of a computer that had only an accumulator and a small instruction set. We didn't learn about index registers until we had had to write self-modifying code to go through a list. We learned about indirection and pointers and so on.
And it wasn't hard! OK, I already had experience, but really, a kid could have learned it easily enough. One could probably turn it into a kind of game without much trouble. And, after that, you just know.
In theory, theory and practice are the same; in practice they're different. (Yogi Berra & A. Einstein)
Programming is not a building block.
Were you ever told to study or take notes in school? Do you know how to do either of those? Think before you answer.
Did you know organization aids in memorization? Did your teachers tell you rhythm and rhyme increase the ease with which you can learn something, or only leverage that fact, most likely thinking they were adding entertainment to keep a class full of distracted kids attentive?
Surely someone tried to feed you acrostics. Even engineers know this one.
What about mathematics? Are you still counting on your fingers and carrying the two? If you memorize your multiplication tables (by brute force) and practice using a Japanese 4/1 abacus, you can immediately compute arithmetic operations in your head. Memorize a simple system of numerical storage (Dominick's, Mnemonic Major, number shape, PAO) and use a digital computation algorithm and you can keep three registers straight while you compute infinite digits in any square root in your head faster than you can write or voice the numbers.
People think too much about goals and not about foundations. They also think children too stupid to understand anything complex, instead of thinking about how people think. You would think folks would say, "Hey, we can describe memory to children in great technical detail, because a child will stare at you blankly, think for about four seconds, and immediately recognize the mechanism you've described!" Instead they say, "Associative? You want to tell children memory is visual and associative? They're not going to understand that! It's too complex!" It's ludicrous; it's like claiming you can't tell a child teeth grind up food and wet it with saliva so it can safely transport down the esophagus to the stomach. They bite a chicken nugget, chew, swallow, and feel it move, and immediately understand what you're babbling about.
As a result, we don't teach children to learn. We force them to learn by whatever means necessary, but give them no tool to drive information into their minds. We don't teach them study methods, note-taking methods, or deliberate practice; we don't teach them any concepts of executive function or mnemonics; and we even avoid showing them highly-structured, systematic approaches to basic mathematics, under the assumption that children cannot handle structure and require a sort of free-play type of classroom learning.
Children need to start with a basic study of the mind. First a brief overview of memory in function, including a high-level overview of the neurology involved and an introduction to mnemonic devices, but excluding mnemonic systems. Then an explanation of leveraging human memory through systems of study and note-taking, like SQ3R and the Affinity Diagram device. These provide the easy foundation to ingesting new information.
Once you've transferred these, you can teach and apply deliberate practice and executive function. Deliberate practice is a method of technical, goal-oriented practice producing constant and immediate results: you recognize your weaknesses and focus on those, while trying to judge if you're improving. Executive function includes a broad array of loosely-related behaviors, notably in eliminating distraction, managing time, and orga
Support my political activism on Patreon.
I started in BASIC. My software made me $4 million last year, and is often cited by other developers as being exceptionally well done.
I looked at teaching my kids scratch. I bought this book called super adventures in scratch. Sort of a comic book that knitted together diverse simple programs into a made up adventure story. No teaching of concepts just moving icons. I did not think much of it but I gave it to my kids.
Man was I wrong. that book was absolutely perfect for a 7 to 9 year old. they gobbled it up, and competed to finish chapters. played the games, and them tried to modify them. all on their own.
I could not have imagined a better introduction. And it reminded me a lot of my own self propelled learning by copying BASIC programs out of KILOBAUD magazine (dating myself).
The challenge then was that computers were slow so you had to figure out how to make programs go fast. The problem today is rather one of managing complexity. And this is where scratch beats wolfram as a language. Scratch has the ingredients we now consider essential most notably event dispatch, listeners and everything that makes objects work. The objects scratch mainly uses are literally iconified (usually a cat or something).
So yes, no one is writing a word processopr or computing sattelite trjaectories in scratch. but it cuts past the crap of languages (remebering syntax) but teaches you the abstract concepts just as a matter of course. There really isn't any good linear program in scratch and even calling subroutines is rare. You many dispatch messages to objects.
Wolfram is right if you already know about programming and are fixated on doing some calculations. but in regards to learning scratch is sugar coated health food that kids love.
Some drink at the fountain of knowledge. Others just gargle.
You should appreciate my ideals about teaching kids about memory, and then about studying and practicing. Memory is immediately recognizable and can produce immediate benefit: every mnemonist is a sort of charlatan who performs a routine, giving his audience a bunch of shit to remember and watching them utterly fail, then telling them to use one tiny little trick--usually the mnemonic linking system or the method of loci--and providing them a new list, which they promptly remember near-perfectly (typically about 1/3 of the audience gets it perfect). They never tell you how physically exhausting it is to train yourself to use these systems effectively... at first, anyway; it gets ridiculously easy with practice.
Imagine a child's reaction to a technical explanation which he can understand, and which immediately makes him *smart*. Then imagine explaining that the brain adapts: new things are hard and tiring, and then become easy and effortless after you force yourself through them for a while. This "force" can be as little as a good 10, 15, or 30 minutes each day practicing. Conclude this by pointing at his amazing, genius memory, and associating it with systems of study and systems of practice which leverage the human memory efficiently, allowing him to learn much more easily. What regard do you think you'll get for the long-term goal?
You'll get a classroom full of kids who want their minds filled, and who suddenly love math. Well, they'll probably still hate math; they'll just enjoy the feeling of dominating it with their elite mind powers. A half-hour lesson and they'll be able to imagine it; they'll crave it; they'll want you to teach them everything.
Support my political activism on Patreon.
With all due respect to Dijkstra, that's a load of crap. I'm from the generation of programmers that learned basic on their Commodores, Tis, and Ataris... The difference may be that I was motivated to learn on my own, but there's a whole generation of older, very well established and learned programmers out there from my age group that started with BASIC. And make fun of BASIC all you want, but I've seen Visual Basic, and while it's not my choice, it's a workable object oriented language that I've seen some pretty complicated and impressive programs written in.
Stupid sexy Flanders.
"Who ever heard of a language you had to pay to use?"
S/S-Plus, SPSS, Matlab, IDL, PV-WAVE, Mathematica, AutoLisp (AutoCAD's language) and a range of other smaller languages developed in the late 70s and 80s and targeted at very specific technical markets.
To some extent, Palantir, Tableau, Spotfire, and even Excel also all are programming environments that are close enough to languages to count here (they all have solutions close to Scratch in spirit).
I'm just scratching the surface - every industry has these languages and environments. Sure, Python (NumPy/Matplotlib), R, and many of the Java-based environments and C++ libraries can be used to replace these, but most customers don't necessarily have the resources or time to move to these solutions. And, those solutions are never complete replacements. Believe it or not (yeah, I know this is /.), many people see value in using commercial solutions that meet their needs better than the free-as-in-beer/Free-as-in-ideology alternatives.
-Chris
ps: Don't forget Oracle and their flavors of Java and PL/SQL that are tightly integrated with the database
Ever tried to describe to someone below the age of 10 why you need to declare variables?
I was a grader in college and can attest to frustrations at having to explain this to people over 10.
It must have been something you assimilated. . . .
That's called Williams Tube storage. Several early computers used it. One of my Computer Engineering profs started on those, then worked on a computer that used mercury delay line memories, a big improvement in storage density. You took a long pipe, formed it into a serpentine that fit the memory cabinet, put an ultrasonic transmitter on one end, and a receiver on the other, forming an acoustic delay line. You regenerated bits, or flipped them, as they came around. Of course, you had to wait for the bit you wanted to fly by.... Later he got involved in some new-fangled stuff called "core", and many of his graduate students went on to be big deals in magnetic core memory design. The last of his graduate students worked on magnetic bubble memory for their thesis work.
I'm not asking to teach how to program assembly before another language, just a very high level analogy of memory (I've seen it as a mailbox or cubby hole analogy), and how the instruction pointer moves from one to another, and I'm not going to do the whole thing here. It would be less than one 45 minute class to describe what I'm talking about; it's actually fairly simple (the gist of it), and would do worlds of good for students to understand what they're doing and what happens when you say "A=1, B=2, C=A+B."
Stupid sexy Flanders.
Thanks. Only the non-anonymous coward understands what I was saying. You can take 30 to 45 minutes - less than a full class period, to give a brief overview of what's happening behind the scenes. If you explain in simplified terms what happens when you program "A=1; B=2, C=A+B," you're off to a better start then handing someone an already implemented class that reads in an image file and resizes it, and then tell the students to modify the code to make a mirror of the image.... by using the built in "mirror" function that's already given in the super high level image library that doesn't come with the language. But hey, kids these days need pictures.
Stupid sexy Flanders.
I think this might apply to both of you so I will place it here.
I hold my PhD in Applied Mathematics and I could do maths by rote, in my head, really quickly and easily but I never understood it until I had a teacher explain why it was that The formula for finding the area of a triangle actually made sense. (Duh, you're squaring it. It's half of that.) At that point it just "clicked" in my head and from that point on, and with more instruction from that same teacher, I grasped the concepts and not just the rote - I understood and could visualize the maths involved.
I imagine that point is different and some people may never get that. I can look at a complicated algorithm, work it out, and visualize it. I might even be able to solve it in my head - even if it's kind of complicated.
All because a teacher told me that you're really just squaring and solving for half the value to find the area of a triangle. Sure, I knew the formula but it just had never clicked. When it clicked - I got it.
There are other things that sometimes just finally click and I understand them. Like, I had the hardest time (and I'm a real mathematician) understanding why a one-time-pad couldn't be brute forced and then it clicked - it can be anything. You don't know, it's everything that has that many letters. It's just something that clicked and I got but it took two people pretty much beating into my head before it finally clicked.
That's how my head works. I don't think it's unique but once it clicks, it's a "complete" thing. So, it's nice when it does click but it doesn't always click so I've done many things (not mathematically related) by rote. They've just not yet clicked. I don't have a better to put it. But, if you can find a teacher who can find the special way to make it click for their students - that's a good instructor in my humble opinion.
I don't know if that makes much sense but that's the best way I can think of to describe it. Hell, it might have even been you who was helping me with the OTP thing. I just couldn't grasp it - and then it just clicked.
"So long and thanks for all the fish."