Learn to Code in 3 Months—or Less

I can understand the attitude - and your counterexample. I know a few people, who have law or humanities degrees, who can come up with some working Wordpress installation or some web form much more quicker than I could (I've been programming since 12). And I would love to see more people being able to automate their everyday tasks - and more and more people can. But the problem of programming I guess is that the limit is really your mind, and that's a bottomless pit. I'm still learning to develop an instinct for code, a feeling for balance between necessity and clarity. And I don't feel my journey will end sometimes soon.

Because filling the road with half-accomplished drivers is a pain in the derriere. The average driver is already annoyed and tense on the road, I can only imagine professional drivers get very frustrated by poor driving when they're outside the tracks. (On a side note, driving is much more vital than programming, so the analogy stops here. You should have the right to drive, even at a non-professional level. Can the same be said of programming?)

Just like less-than-average drivers will annoy the pros, the less-than-average programmer will limit the field. If you look at my original comment, I clearly mentioned: "How to make the real programmer on your team miserable". If the less-than-average programmer sits in his parents basement working on his own racoon-loving social network using the most mind numbing technologies, that's okay. But as soon as he interacts with others, he will inevitably cripple the advances of the field. Think of what happens each time a standard comittee reaches the conclusion: "It's a great idea, but too complex for people to grasp".

I know I sound very elitist, and I wish I wasn't. The truth is programming is hard. Very, very hard. Sure you can use your favorite framework to generate an html form and read DB fields, but this is not programming. Dijkstra has written extensively about the subject, calling programming "one of the most difficult branches of applied mathematics" or saying about BASIC (whose usage then is very analogous to the subject at hand now) "the teaching of BASIC should be rated as a criminal offence: it mutilates the mind beyond recovery". EWD was an extremist, he argued that CS students should focus on symbol manipulations, prohibiting them to manipulate computers for the first few years, focusing on an abstract formal language instead. He criticized universities for "[lacking] the courage to teach hard science" and their "infantilization" of the "radically new" idea that is computer science.

I won't go as far. If what he suggests is an extreme, a website promising you to learn code in 3 months is the opposite extreme. A 4 year CS degree, a couple of open source projects and the deliberative practice Norvig talked about should be a good moderate middle-ground.

In short, do I encourage people to start programming? Of course I do, it's the most fun I've ever had in my life. Do I agree that three months worth of learning and couple of RoR pre-fabricated functions is enough to say they learned how to "code"? Absolutely not.

To be fair, my diatribe is not geared at self-taught programmers. I have a lot of admiration for the people who learn programming that way, and I personally consider that most of the (theoretical, formal, CS) knowledge I got came from self-teaching, picking up a book and personal experimentation despite having earned a Masters in Engineering that taught me more about corporate bullshit like "Project Management" or "Work Ethics" than it did on programming - The sad state of engineering schools in France, a whole different subject.

What I'm merely pointing out is the ridiculously short amount of time the article is promising its readers. 3 months? Really?

I have been programming for a while now, borderline obsessing about it for the past 2 years and a half, and frankly, I feel like I still know so little. When yet another blogger tells her cool marketing friends that they too can learn to code with such few efforts, it makes me chuckle. And cry a little.

I'm also Australian! I'm led to believe Junior developers are at around $90,000. It's hard to say with the level of experience these people have, as it's very much entry level.

I don't know about "junior developers", but the average starting salary for a fresh BS from my public university (based on a voluntary exit survey, so there may be some upward bias) is around $80.5k.

This is particularly pertinent because the vast majority of the students here end up working in the Bay Area.

I feel the title of Software Engineering is misleading because it doesn't quite relate to other fields of engineering. The way Software Engineering is used is more like Software Construction, where the concern is more about programming using building blocks and tools of the trade, along with construction methods: like code organization, deployment, source control, etc.

Computer science might best be aimed at: Big Data Analysis, Compiler Writing like HipHop (Php compiler), or Distributed/Parallel Programming, Large Scale Deployment, etc.

I agree that writing an LALR recursive descent parser might not make someone an architect, but neither will a 300 hour/8 week course, mainly because I feel experience is the deciding factor.

Employers often want to create a repeatable scaleable process -- so yes this would be attractive to an employer.

Who says a CS degree is required for admission into development? I've worked at companies where many of the developers didn't have CS degrees, but there is a noticeable difference between those with a degree and those without. Those that don't have a CS degree, but somehow still make it, are often those that could have made it through a 4 year course. Sadly, there are also those with a CS degree, or even a graduate degree, that can't hack it either.

So in the end hiring is difficult, and there doesn't seem to be many indicators that some one will be a good developer and eventually an architect. But, when hiring I personally would give preference to a someone who holds a CS AND has the right attitude.

Also it's a matter of goals. If your goal is to stand up something cookie-cutter-ish, and then just duplicate that process with varying degrees of style, then sure anyone can probably learn to do that. But, if you need to integrate systems, write parsers to put data into a high-performance data warehouse, organize structures for a custom CMS, etc, would you really go to someone fresh out of a 300hr/8week course, or someone fresh from a CS program, if those were your only 2 choices?

You seem to disagree with a point I am not making. I'm saying that they are both valuable (maybe CS grads who want to work as Software Engineers should take such classes), but not equivalent.

There are values to an academically rigorous (CS) education that are unrealted to Software Engineering. Simply being aware of CS fundamentals is not equivalent.

There are many eng jobs outside of Startup land where a deep knowledge of CS is far more valuable than how to write and deploy a full stack web app.

Because the industry is so much more than throwing preformated auto-generated web forms at a database. There are things far more serious than your average "Facebook meets Fitocracy for lumberjacks" ideas...

If Carmack and others managed to do great without a formal degree good for them. But they're outstanding people and are more the exception than the rule. If you're as talented as these guys, fine let your skills speak for themselves. But assuming every person following this 3 month "program" will now call themselves a programmer is something completely different.

"the industry notion that a CS degree is needed to be a good engineer needs to be challenged."

true enough. but whether CS knowledge is required is a more interesting question. John Carmack (picking a name from your 'homegrown' list) is extremely knowledgable about algorithms, computer architecture, graphics etc.

of course, it also depends on the definition of 'good engineer'. If you define it as "someone who can build a simple CRUD app in Ruby on Rails", then probably very little CS knowledge is required.

I think mid level engineer needs to be contextualized. Mid level at top CS companies is a huge achievement and not something that '[anyone] can be made to resemble ... in a short space of time"