Live data from Hacker News

Teaching how to code is broken

neil.computer

211–220 of 258 posts

Re: Teaching how to code is broken

#211
post #170

It’s a little confusing and sad to me that people have to try to pressure themselves to go from zero to professional engineer in a short period of time. That sounds stressful and, frankly, almost impossible. Like a lot of people here, I taught myself how to code as a child in a terribly inefficient manner. I spent probably around 100 hours making games on my TI-83 - you had write a formula to figure out how to color…

> It’s a little confusing and sad to me that people have to try to pressure themselves to go from zero to professional engineer in a short period of time. That sounds stressful and, frankly, almost impossible. ... That's what engineering school is for. At least, that's how aerospace engineers, EE and so on do it. > I’m grateful that my childhood obsession turned out to be absurdly lucrative. I find that, when there's…

Yes, undergrad is a good place to learn to code. I’m still glad I learned when I did, and I think it made college a lot easier, but no doubt four years is plenty of time, and allows ample opportunities to code for pleasure as well.

I was referring to the people who, say, try to switch careers to tech and go to a bootcamp to learn to code. That just sounds really hard to me, a horribly compressed timeline, and I feel like you would always be worried about gaining professional competency vs just relaxing and enjoying the ride.

I knew a guy who actually went back for an undergrad and masters in CS in his fifties. He was a classmate and briefly a coworker. He was an excellent student and won some department award, and very organized, but somehow not very good at coding things. He would get lost in a sea of sticky pads and notebooks and take two weeks to do very small amounts of work. He was let go in the end. We forget how hard this stuff can be. (Ironically to me it’s everything else that seems hard, from music to mechanics to sports, coding is just so… logical.)

Re: Teaching how to code is broken

#212

Earlier quoted context omitted.

I have never met a programmer I respected skill-wise that wouldn't randomly spend a weekend locked in a room solving an esoteric problem they felt they had to solve (for no reason). Over the Christmas holidays everyone at my company writes PoCs or hobby projects, being glad to have some free time away from paid programming in order to enjoy recreational programming. What I'm trying to get at is software engineering i…

This “you gotta program all day every day or else you are unfit for it”-attitude is incredibly toxic and harmful. No other profession has this. No dentist gets told he gotta setup a practice in his garden shed to practice pulling teeth, or else hell be a horrible dentist. No, doing job for 40 hours a week is enough to be one.

I agree on this, but I believe the comment was aimed more at the "Tinker-Attitude" that many STEM-People have. And I believe many other professions have this to an extend: I'd imagine that many journalists and authors enjoy reading in their spare time. Every mechanic I know has a cellar full of half-finished machines he fiddles with in the evening.

I don't have the energy to code on private projects for eight hours after eight hours of coding at work (besides occasionally enjoying not sitting in front of a monitor...). But every once in a while it bites me to blow 10 hours of tinkering on some strange useless problem, like simulating Logic-Gates with basic arithmetic.

Re: Teaching how to code is broken

#213

Earlier quoted context omitted.

I think that was neither nice nor fair from you. This is experience. If you had a cushy life, good for you. But don't tell others their experiences aren't real.

Ok, I'll bite. Citation? Any evidence of any kind to offer? See, that post just read like negative chatter. "Folks are always doing wrong stuff." Meaningless to offer without anything concrete.

This is a pseudonymous account and I am not willing to change that.

I will try without giving anything away, but you will need to help me here:

- Do you really think some people here haven't grown up with feminist teachers who took it way too far?

- How many times have you heard anything about "Girls who code" vs "Boys who code"?

- If you are actually interested and ready to see something ugly, just go back and try to read some "mens day" discussion on Twitter or I think even on HN too.

I've tried to be active in those days in a positive way. Last year I gave up. Any mention of today being mens day on company chat is immediately laughed off with something along the lines of "and so is 363 other days".

Meanwhile men die younger, have higher incarceration rates, more suicides, and gets less education.

And also the few times there is discussion it is often redirected to "how men are suffering from toxic masculinity" which is probably true but only a small problem.

Re: Teaching how to code is broken

#214
I've been coding since I was six years old. (I'm in my forties). For the first bit of that it was BASIC on a TRS-80 Model 100 laptop, with 16k and a tape drive. Then, really, Perl and Hypercard. Answering the question of but what for??? was a huge part of my friends' and my dilemma in building simple games and projects. In my experience, there is a chasm between getting good at doing something technically and knowing what the hell you're supposed to do with that knowledge. Screensavers; that was the best idea we had and that's what we competed against each other on. We'd sit around and talk about how we could make something like SimCity plus SimEarth plus realistic physics and make a universe simulator. Or a driving game that could generate cities to drive through. But I had the same experience with 3D modeling! I could spend a hundred hours to make a dinosaur in a pirated copy of Infini-D. Okay, now what?

Perhaps this is off track, but I guess to us at the time the joy was in discovering the technical aspects, and tinkering and learning how to use them. I read the BASIC manual that came with the TRS, and I remember being 7 years old and skipping the part about "Arrays" repeatedly because it seemed too complicated to light up dots on the screen. Until one day I just was like, okay, what's this "array" thing, I feel like I'm missing something. And suddenly that chapter blew my mind. Likewise, I was deep into my third or fourth PHP online store / shopping cart in the late 90s, when a kid I hung out with was like, what? You're not using a database? And I was like no, I just have text files for each product. There's a whole back end for the business to edit the text files and images. Then, because I couldn't not look at it, I had to look at mysql. Holy shit, you mean I don't have to create text files for every product? Or block one person from editing them when someone else is? The rest being history.

Okay, broader scope: You can't generate use cases until you need them, and also, you can't learn tools until you have the use case for them, or else you're going to forget what they were for. Tools and the experience of working with the tools go together. You might be able to simulate that experience by building a card game, but like all skills, they will be lost quickly unless they're exercised regularly.

Re: Teaching how to code is broken

#215

Earlier quoted context omitted.

I have never met a programmer I respected skill-wise that wouldn't randomly spend a weekend locked in a room solving an esoteric problem they felt they had to solve (for no reason). Over the Christmas holidays everyone at my company writes PoCs or hobby projects, being glad to have some free time away from paid programming in order to enjoy recreational programming. What I'm trying to get at is software engineering i…

This “you gotta program all day every day or else you are unfit for it”-attitude is incredibly toxic and harmful. No other profession has this. No dentist gets told he gotta setup a practice in his garden shed to practice pulling teeth, or else hell be a horrible dentist. No, doing job for 40 hours a week is enough to be one.

They're not stating a moral judgment.

Re: Teaching how to code is broken

#216
I'm not sure I agree with this argument. Learning stuff by diving in and doing stuff immediately without studying the basics first may be fun and interesting, but it might lead to fundamental gaps and misunderstandings too.

I learned to write code from the bottom up. It was sometimes dull and hard work, but I think in the end I got very good at it.

On the other hand, I learned to play golf by just playing games. It was fun, I got ok quite fast... and then I never got any better. My nephew started playing by having proper lessons and could beat me after a few weeks.

Re: Teaching how to code is broken

#217

Earlier quoted context omitted.

> 1. Girls in STEM: society needs to quit telling people "math is hard, tee hee" and quit shopping in the Pink Aisle at the toy store and reinforcing that culture. Buy your kids Lego and Raspberry Pi circuit kits and see what happens. Girls in STEM =/= Girls becoming professional programmers. There are plenty of girls in STEM fields such as medicine, biology, geology, physics, Math, civil engineering... The fact that…

Sitting at a computer for 10 hours a day isn't a gendered thing. There is nothing about computer science or programming that is gender specific. It, ideally, should sit around 50-50, ±5%. So yes, it is a problem that it so heavily skews male. As to why, it's multi-faceted cultural issue, with how our society treats boys and girls starting from birth. In other words, a pipeline issue.

> Sitting at a computer for 10 hours a day isn't a gendered thing.

Except the 10 hour a day part :) I believe the tendency to overwork is mostly a male thing.

Re: Teaching how to code is broken

#218

> I've found that modeling or simulating something real like a card game is very effective. I think the last paragraph of this article should have really been blown up to be the entire article. I want to know more about this experience the author had. Because one question I have is: should we even be teaching object-oriented programming to beginners? The rationale for the modified class schedule is that it answers th…

If you ask me easy teaching of programming died with the 8 bit machines that only did basic. You had instant feedback, the system was extremely simple and it was possible to hold it all in your head.

... and as you did more complex stuff you realized on your own that basic isn't quite enough and went to look on your own for more complex environments to manage the complexity you had created.

Re: Teaching how to code is broken

#219
As someone learning right now this reflects perfectly my frustration with the teaching systems I’ve found.

The first part, let’s say the basic logic of programing is fine and relatively fast to learn. I have no problem with lessons starting there.

Where I find the problem is with the actual translation to real projects with the languaje, how to interface the code with the data, the server and the outside world.

The jump is huge! Of course you can overcome it with effort and stack overflow, but the feeling is like walking in a nice hill with clear views and path, and suddenly arriving to a vertical cliff with no clear routes up. You see others climbing easily, hanging out there, but you don’t even know where to set your first hand.

In my case, I’m starting with javascript in codecademy. The first lessons are more or less clear. Some concepts are harder to grasp or to express in code correctly, but with some exercises and effort you get them.

Then you start with real application stuff (promises, requests, Get, Post) , at first is easy enough, the concepts are not hard (the syntax is a bit harder, or at least the variety of syntax can be confussing). But suddenly you get thrown into AJAX, JSON, Frameworks, boilerplate. All at once, from 0. Copy this code, change this variable, lots of instructions but little learning or at leat a highlevel view of where you are.

I could choose another course, but I rather keep with the good parts in this one and look for the lacking lessons outside, than start hopping schools. Also I have some great friends that are experienced programers that can help me when I get too stuck.

Is interesting how different teachers explain differently. Some lessons are perfect for begginers, clear explanations, useful exercises. But in some lessons you can clearly feel how they are created by people used to teaching experienced programmers. Several concepts or tools thrown at you without further explanation, handwaving lots of steps, repeat this piece of code several times and that’s it.

I know that I can go through the material, google it, see youtube tutorials and classes and advance. I’m doing it, most of you have done it before. Not am impossible task by any means.

Maybe learning to code is easier than ever, but still… I find that it is way more complicated or with a steeper courve than necessary at the level I am currently.

There are lots of great resources to learn, but is difficult to find them structured in a coherent logical way for the learner! Is a bit frustrating TBH.

Re: Teaching how to code is broken

#220

My approach is to employ folks with enthusiasm. If they're teachable, they'll figure out whatever pile of tools they get thrown into on the first task. Then they'll have some skills and the next task will seem less daunting. Of course some mentoring is useful. But just the stuff that's pertinent to the task at hand; concrete help that gets them moving forward. Because, somebody is paying for results. And because prod…

Enthusiasm is perhaps the easiest thing to fake in an interview, and something many of us fake automatically, as smiling and seeming engaged is something a lot of us learn to do when meeting new people. I doubt it correlates with productivity outside of something like a sales role. Most of the world's work seems to get done by people who don't appear particularly enthused.

> Most of the world's work seems to get done by people who don't appear particularly enthused.

Interesting! Can you share some sources to get deeper here? Until now, I thought apathy wasn't a sign of productivity.

Post reply on HN