Learn to Code, It's Harder Than You Think • Hacker News Board

Re: Learn to Code, It's Harder Than You Think

#111

Post by serge2k » Fri, Dec 04, 2015, 4:35 PM UTC

panglott wrote: post #101

paulcole wrote: post #27

There's a big gulf between knowing how to code and working as a professional software developer. I'm a writer for a marketing agency, but I write Google App Scripts, JavaScript, and PHP on a regular basis to automate tasks, try out ideas I have, etc. There's no way I could work as a professional software developer, but I know how to code enough to help make my professional/personal life easier and more fun. Isn't the…

Excatly. There's a huge difference between "everybody needs to learn to become a professional software developer" and "everyone needs a higher level of software literacy, including basic coding skills" Lots and lots of office jobs involve stuff like updating content on a Django app or shoestringing together Excel VBA and Access stuff. There's all kinds of tasks where scripts and low-level automation would be useful.…

I would argue that neither of those statements is true.

I think this idea that coding is some sort of silver bullet is ridiculous.

Re: Learn to Code, It's Harder Than You Think

#112

Post by moyix » Fri, Dec 04, 2015, 4:36 PM UTC

> Other research points to similar results. There seems to be a ‘double hump’ in the outcome of any programming course between those who can code and those who can’t.

> “In particular, most people can't learn to program: between 30% and 60% of every university computer science department's intake fail the first programming course.”

Argh, I knew when I opened the article that its evidence would end up pointing to the "camel has two humps" study.

That study has been retracted, and the author regrets the damage it may have caused to computer science education:

http://retractionwatch.com/2014/07/18/the-camel-doesnt-have-...

> Though it’s embarrassing, I feel it’s necessary to explain how and why I came to write “The camel has two humps” and its part-retraction in (Bornat et al., 2008). It’s in part a mental health story. In autumn 2005 I became clinically depressed. My physician put me on the then-standard treatment for depression, an SSRI. But she wasn’t aware that for some people an SSRI doesn’t gently treat depression, it puts them on the ceiling. I took the SSRI for three months, by which time I was grandiose, extremely self-righteous and very combative – myself turned up to one hundred and eleven. I did a number of very silly things whilst on the SSRI and some more in the immediate aftermath, amongst them writing “The camel has two humps”. I’m fairly sure that I believed, at the time, that there were people who couldn’t learn to program and that Dehnadi had proved it. Perhaps I wanted to believe it because it would explain why I’d so often failed to teach them. The paper doesn’t exactly make that claim, but it comes pretty close. It was an absurd claim because I didn’t have the extraordinary evidence needed to support it. I no longer believe it’s true.

Re: Learn to Code, It's Harder Than You Think

#113

Post by qntty » Fri, Dec 04, 2015, 4:37 PM UTC

conductr wrote: post #11

I liken coding to playing a musical instrument. Some people are musically inclined, they can learn easily and fast and they tend to the best. But most people require some level of training complimented by the dedication of what amounts to mostly self-practice. So in a way, you're self taught. The vast majority of people are willing to buy an instrument, just to see if they are musically inclined, but when they find o…

Quoted in SICP:

A computer is like a violin. You can imagine a novice trying first a phonograph and then a violin. The latter, he says, sounds terrible. That is the argument we have heard from our humanists and most of our computer scientists. Computer programs are good, they say, for particular purposes, but they aren't flexible. Neither is a violin, or a typewriter, until you learn how to use it.

Marvin Minsky, ``Why Programming Is a Good

Re: Learn to Code, It's Harder Than You Think

#114

Post by hockeybias » Fri, Dec 04, 2015, 4:38 PM UTC

sbov wrote: post #78

mattmanser wrote: post #35

Earlier quoted context omitted.

Err, yeah, you could work as a professional programmer if you wanted to. You can code. The whole point the article is trying to get across is that a lot of people are simply unable to learn how to code. You've learnt. You CAN become a professional. You just don't realize it. The step you've taken is the hard one. In fact, there's another big dirty secret in programming. A load of professional programmers aren't able…

There's a dirtier secret: A load of professional programmers are not able to make new programs. A load of professional programmers are not able to modify existing programs. And a load of professional programmers are not able to finish programs. Almost no-one is good at all 3. Most are extremely bad at least one of these. Lots of people who are "good" at making new programs are actually pretty bad.

That sounds quite harsh or I must be fortunate - I can do all three and work with devs that can do all three.

Re: Learn to Code, It's Harder Than You Think

#115

Post by gyardley » Fri, Dec 04, 2015, 4:39 PM UTC

joshcanhelp wrote: post #52

discardorama wrote: post #28

Coding requires a mindset that some people just don't have, or haven't developed yet. Ask any TA who has taught the basic "intro to CS" courses, and you'll hear lots of anecdotes which show that some people just don't get it. Here's an example: Q: write a program that reads 2 numbers from standard input, and outputs their sum. For example: First number: 3 Second number: 7 Sum is: 10 Here's a solution someone submitte…

There was a paper published about exactly this, summarized by Jeff Atwood [1]. Essentially, there appears, according to the authors, that there is a mindset that people either have or don't have that allows them to understand the very basics (variable assignment, iteration, etc) before they've had any training at all. Students were tested before 3 weeks of education and after and the results were essentially the same…

I really wish people wouldn't link to ten-year-old drafts without checking on current research. The article Atwood links to makes some totally bombastic claims - here's the retraction and clarification of them:

http://www.eis.mdx.ac.uk/staffpages/r_bornat/papers/camel_hu...

While a consistent mental model for things like assignment is a good sign, not initially having it doesn't mean you can't learn how to code. Lately, researchers have found that certain pedagogical techniques (especially pair programming) are pretty effective at overcoming these deficiencies.

I wish I could understand why so many programmers are eager to assume their skills are due to something 'innate'.

Re: Learn to Code, It's Harder Than You Think

#116

Post by vonmoltke » Fri, Dec 04, 2015, 4:41 PM UTC

Alex3917 wrote: post #47

paulcole wrote: post #27

There's a big gulf between knowing how to code and working as a professional software developer. I'm a writer for a marketing agency, but I write Google App Scripts, JavaScript, and PHP on a regular basis to automate tasks, try out ideas I have, etc. There's no way I could work as a professional software developer, but I know how to code enough to help make my professional/personal life easier and more fun. Isn't the…

> There's no way I could work as a professional software developer Actually given the way our economy is currently structured, it's easier to get a good job as a developer having never written a line of code than it is to get a good job as a marketer even with 5+ years of experience. It might take you a couple tries to pass the interviews, but that doesn't matter because getting interviews is relatively easy. (Albeit…

I think you are overstating how easy it is to get interviews for development positions. From what I have seen, someone without some demonstrated coding experience, professional or hobby, won't even get the time of day from companies looking to hire developers. They will never get to the take-home assignment phase.

Re: Learn to Code, It's Harder Than You Think

#117

Post by abathur » Fri, Dec 04, 2015, 4:41 PM UTC

There's a lot of back-and-forth over whether it is easy or not or whether one comparison or another works best.

I think it's fine to compare learning to write code with learning to write human language. The best of us will still make mistakes; plenty of us will always be incapable of writing well. Once upon a time the idea that everyone needed to know how to write was laughable. These days, it's arguable how well everyone needs to be able to write, but there's a general agreement that it's worth giving people the opportunity to reach at least a basic level of competency, even if their daily lives require little writing.

I had a miserable time learning to write (i.e., failed many assignments) until a few things fell into place, and I saw how writing could be an important part of thinking, learning, feeling, expressing, etc. I had a miserable time learning to code through coursework in highschool and college until a few things fell into place and I saw how code could be an important part of thinking, learning, feeling, expressing, etc.

How did/do we sell everyone on trying to learn to write human language? AFAIK, with some superlative rhetoric about how useful/necessary writing can be without much dithering over what portion of the population will ever be good enough to make a living at it.

So there are varying levels of irresponsibility when it comes to saying everyone should code. There's probably a big potential social upside to fudging things a little to suggest everyone should learn some programming throughout school; it's much more negligent to tell someone with a boring but stable job to quit and blow a few grand on a coding bootcamp without an inkling that they're prepared.

Re: Learn to Code, It's Harder Than You Think

#118

Post by serge2k » Fri, Dec 04, 2015, 4:43 PM UTC

Alex3917 wrote: post #75

bovermyer wrote: post #67

Earlier quoted context omitted.

Did you seriously just suggest that the majority of American adults are illiterate? I'd love to see the source you base that claim on.

> I'd love to see the source you base that claim on. Like the government's official data on adult literacy? http://nces.ed.gov/NAAL/PDF/2006470.PDF After 1994 they got rid of the 'advanced' literacy ability category, corresponding to roughly the level academic journal articles are written at, because not enough Americans qualified. So currently the top literacy level is 'proficient', which is roughly the minimum leve…

How is that paragraph evidence of illiteracy?

Re: Learn to Code, It's Harder Than You Think

#119

Post by lloyd-christmas » Fri, Dec 04, 2015, 4:43 PM UTC

bencollier49 wrote: post #102

jdiez17 wrote: post #43

This article reflects my experience at University very well. I'm a self taught programmer that ended up pursuing a CompSci degree. I'm in the last semester of my BSc, and in hindsight I should have studied something else, like Physics or Electrical Engineering. The standards are ridiculously low. Most of my coursemates (final year CS students) are literally unable to write software. They can write a single-file cpp f…

This matches my experience. I was trying to hire a junior programmer, and got someone in for interview who ostensibly had a first-class CS degree from a Russell Group university, yet could not explain to me what Functional Programming involved. He didn't even recognise the term. (Incidentally, the university refused to corroborate his degree, on the grounds that it would be a breach of the Data Protection Act, which…

From a non-CS degree perspective (on the hiring side), it seems that CS undergrad is meant to produce CS professors. Most (yes, MOST) have never touched anything higher than C++. While fundamentals are great, teaching them to the exclusion of immediately tangible skills does a disservice to people who are effectively paying $200k for a ticket into the working world. The reason we consider boot campers is because they typically have a work background demonstrating an ability to be useful as well as a recent educational background demonstrating an ability and willingness to pick up foreign tasks.

Re: Learn to Code, It's Harder Than You Think

#120

Post by jdiez17 » Fri, Dec 04, 2015, 4:44 PM UTC

learc83 wrote: post #110

jdiez17 wrote: post #43

This article reflects my experience at University very well. I'm a self taught programmer that ended up pursuing a CompSci degree. I'm in the last semester of my BSc, and in hindsight I should have studied something else, like Physics or Electrical Engineering. The standards are ridiculously low. Most of my coursemates (final year CS students) are literally unable to write software. They can write a single-file cpp f…

I went to back to University after working after working as a professional programmer for about 5 years. What I learned in Automata, Discrete Math, Algorithms, and the other more theoretical classes has been worth it's weight in gold. The things I learned in "Software Engineering" on the other hand, not so much.

>What I learned in Automata, Discrete Math, Algorithms, and the other more theoretical classes has been worth it's weight in gold.

That was my intention when I signed up for University too. However, I've only had two maths modules, and they haven't been particularly challenging.

One of the things I was looking forward to learn at University was hardcore cryptography, since I believe that's one of the few things for which getting formal, structured education is the best way to go. Sure, you can learn the fundamentals on your own fairly easily (even more so today - Stanford's online course in cryptography is very good), but you need to learn about attacks or you'll write code vulnerable to known problems like hash length extension attacks, ciphertext modification...

However, the only time cryptography was mentioned - at all - in my entire BSc was in first year, where we had a few lectures explaining the rudiments of RSA. Interesting, for sure, but pretty much useless without the remaining 80% of knowledge required to have a full picture of what's going on.