I was a computer geek from way back, had a few BBSes in the late 80s (yes, I'm a child of the 80s & 90s), learned QBasic & VisualBasic back in the day, and tinkered with Python for a few years off and on... mostly off. Other than that, nope.
The article shows us that nowadays anyone can easily start learning a programming language or even a framework, that the internet is an incredibly useful library for those who seek knowledge and that tabloid headlines like "How I became a programmer in 12 weeks" still get our attention.
I'm sorry but you can't become a _programmer_ in 12 weeks. You can become a pretty good code wrangler, you can put togeter a decent website or an application but you still have around 10 more years to go (as shown above/below http://norvig.com/21-days.html).
Suggesting that it takes ten years to become "a programmer" is absurd. An experienced programmer, perhaps, but as soon as you can write and reason about programs (inferring what they do without a compiler), I would call you a programmer. Whether you're good is another matter, though a talented programmer can outpace one a decade their elder.
This just goes back to the "people as resources" (which is code word for developers are widgets) garbage that constantly comes up within organizations, etc, it wouldn't surprise me at all if creativityhurts is a manager of some sort. People are unique, not drones, some have a naturally higher aptitude and interest in programming than others as well as other factors such as intelligence. IMO a programmer who loves what they do and spends time working on their craft outside of their 9-5 job can quickly outpace a more experienced programmer who has little passion and just does it "because its their job". I've seen this first-hand that some people with years and years of experience just don't have that enthusiasm for software development, as soon as you talk to them about a concept that they aren't already familiar with through sheer rote memorization they either lose interest completely or just can't seem to catch on. That said, obviously if you take 2 developers who have similar attributes as far as aptitude, intelligence and enthusiasm but 1 has 10 years more experience than the other, than yes, of course, the one with more experience is going to be better overall.
yeah seriously. I'm teaching myself how to program and for the last year, I ve been spending an average of 10 hours a day reading, practicing,studying Python and JavaScript. I m just getting into concepts like Test Driven Development and design patterns. I m a disciplined guy. But, saying this is like saying: " How I became an astronomer in 12 weeks". It's not just the theory, It's the experience that comes with time and things you realize by making mistakes that make you become something.
Let's not confuse correlation with causation; I once looked around for studies showing experimental interventions which actually increased work ethic (specifically, I was looking for studies on the Big Five construct 'Conscientiousness') and didn't find any.
So, I'm sure the work ethic that let him join and last 10 years in the USMC helped.
Lets not confuse an absence of evidence as an evidence of absence - just because there's no study yet that shows that 10 years in the USMC helps with work ethic, doesn't mean that 10 year in the USMC does NOT help with work ethic (unless you mean to say in your survey of the lit. you found something claiming the opposite of the op's claim).
That said, I'm inclined to agree with your last statement...
Well, in general terms, if the USMC (or the Army, or the Navy, or...) really did build up work ethic - a personality trait that correlates quite a bit with income - why do veterans seem to do so poorly, even veterans who never experience the confound of war? Because personality traits are, like IQ, very tough to alter. (This is why that psilocybin-increases-Openness study got so much play.)
You do poorly at kicking ass & making bank in civilian life. This is a tricky subject, of course (any statistics from 2001 and on are basically worthless because all the suicides and whatnot from Iraq & Afghanistan trash the data), and there are all sorts of correlation/causation problems (maybe the people who join would do poorly in the first place - these days, most of the graduating class of Harvard heads to Wall Street and not officer training).
But if you have any citations to the contrary - that veterans do better than one would predict from their pre-military life - I'd be glad to hear them. Conscientiousness is as important as IQ, and any intervention which genuinely improves either is really valuable and ought to be well known.
Don't confuse work ethic with motivation. I know a number of successful veterans who are kicking ass and making bank in civilian life. I also know a ton who are just 'retired' and enjoying their low-key life with friends and family.
You can't switch from studies and causation/correlation to just stating without citation that they suck at kicking ass.
He learned some Python, Django, HTML, CSS, Javascript and wrote programs. Can "programmer" not encompass both people getting their feet wet and 20 years experience mastering multiple languages?
He hasn't made any hard choices, he just picked what is most available and put the pieces together.
Frankly he'd have bought me if he made his hands dirty with some obscure 8bit assembler. There he'd have to really THINK his way out.
At least some sweat during rolling his own data structures and algos around them would make him something in a shape of a programmer, but not this.
Of course what he's done is cool, learning anything new is worth some pat on the back. But this is a newsfeed on hacking, and Lords, it ain't hacking. Let's call things what they are, ok?
There's also the problem that some people like him start calling themselves programmers on the job market and when we're trying to hire them, they occur to be only Django programmers with very limited abilities. Practice makes perfect, but they didn't do it yet. Have some interviewing experiences like that. Not funny.
Are you confusing Hacker News with Reddit? (for context, below is mattdeboard's deleted comment)
"Well, thankfully, I didn't write this for you. I wrote it for the people -- I suspect the vast, vast, vast majority of the HN readership -- who are not expert programmers, or even professional programmers, but they want to be."
What load of bull. What separates a 'programmer' from a 'Django programmer'? I can't think of any way to make that distinction that isn't logically self-defeating.
A programmer is language agnostic and has a deep understanding of both algorithms and computation. I find the term para-programmer to be an apt description of anyone else.
Why is this on the front page btw? Aren't we all hackers here? What makes this news?
Maybe a grad student has those things ( ;) ), but once you realise computers aren't the hard part about building software, you realise that there are a whole host of skills you need to be a successful programmer, that have nothing to do with algorithms or computation.
If all you have is a strong understanding of algorithms and computation... your just a computer geek. Doesn't make you a good 'programmer' (in the sense that you can produce something someone wants).
Huh? A good "programmer" is a good problem solver. You can't be a good problem solver if you don't have a conceptual understanding of algorithms or have any notion of what it means for a problem to be "hard".
Why do you need algorithms to be a good problem solver? Must of sucked to solve problems before the 60s given there were no good problem solvers :(
In fact, how did mathematics and physics even develop at all before then, without knowing what was hard, or how to solve problems.
Seriously, a lot of software doesn't require complex algorithms, sure there is a subset that does and that's cool. But saying someone isn't a programmer because they don't have a deep understanding of computation and algorithms is disregarding a large population of quality programmers.
> If all you have is a strong understanding of algorithms and computation... your just a computer geek. Doesn't make you a good 'programmer' (in the sense that you can produce something someone wants).
I think y'all are all using different terms (and I notice the parts-of-HN undercurrent of "technical skills don't matter, you need to be social and build products, startup pivot founder startup" but I digress).
To me, someone who can use algorithms et al is a "programmer", which is largely orthogonal to being able to produce something someone wants, call them a developer or something. Not every programmer has to be social client facing; getting detailed functional specs and creating good code to fulfill those specs absolutely does make one a programmer.
"It's not important to get things right/perfect. It's important to get things going"
Not realizing that fact held me back for years.
When I wanted to learn / do something, I would obsess over finding absolutely the best tutorial / process. Then I would obsess over the right learning environment. Always worrying about the incidentals, and ignoring the essentials.
HN has helped me a lot in this regard. Reading about big companies and how 'imperfectly' they started out, as well as the whole 'Lean Startup' concept has done much to change my mindset. What is important is starting.
I used to do that too - find the best practices and right ways of doing each new thing I came across in my projects. Although it wasn't efficient, I'm still glad I did, since it's a luxury that I don't as often have time for now as a professional. Having covered a lot of these things when it was still just a hobby has paid off, now that I can use it all again.
Agreed! I am only about 5 weeks in to learning Python with no previous experience and doing it very much part time as I work full time but there are some great tips in this article. Don't let the title throw you off here.
I highly recommend the O'Reilly School of Technology courses on Python if you are looking to learn Python. They are self-paced, reasonably priced, and provide some great tips on best practices. And the ability to have someone actual give you feedback and test you I find helps me keep moving forward learning to program.
Hmm, I started programming over a decade and a half ago, writing my own little game engines in dos, and a nice little x-mode GUI. I've worked as the software architect for a moderately popular (top 2000 alexa) social networking websites, I've worked at fortune 500s, I've been on the dev side, the test side, the pm side, and worked as an IC and as a lead/manager. I still don't really consider myself as being anywhere near being complete in the task of becoming a "programmer."
But kudos on taking your first few steps. Its not a bad vocation.
I envy that the author learnt programming and web development now as opposed to even a few years ago. Without devolving into "in my day I walked ten miles in the snow just to compile"; it's just so much more pleasant now. Stack Overflow, Heroku (and similar), GitHub, PDF Books. Obviously it takes time and effort to get better everyday but the learning curve is a bit flatter now.
I second this. Although my story is different. I would put my toe in the water and try to learn to program, but it all just seemed so absurd to me, so I pursued other areas of study.
Then earlier this year I started doing Hartl's Ruby on Rails tutorial and things just started clicking. Now I am teaching myself Ruby on Rails and I am looking for explore other more difficult languages (just for fun, I don't want a career as a developer).
For whatever reason, Ruby just got me more interested in programming. Now I am dipping my toes into SICP, which I would have never have guessed I would be interested in.
I think once you get your hands dirty with something relatively easy like Ruby on Rails, you start seeing the need for all the other computer science concepts and languages and that piques your curiosity.
However, I can say that teaching myself a little coding has really helped out in my day job, I notice I think differently about problems. My thought process is much more structured.
There's also a lot more to learn and possibly more confusing. In 'my day' it was just CGI with MySQL, Apache and perl, then comes Java, then along came PHP then Python, Ruby, now there is JQuery and Redis and Hadoop and.. and on HN now you see a hundred new frameworks each saying 'we're great, use us!'. While I welcome that there are easier ways to actually learn all these things and the amazing communities, I'm not sure it's really easier to actually just get started. Oh, and you've got Reddit and Facebook and HN sucking up your time and a fast internet connection to procrastinate on Netflix and Youtube.. I could go on.
Very true that there are so many choices now. Easy to suffer from analysis paralysis before you even get started.
I answered a questionnaire for Manning in exchange for a free book recently. I wanted to learn something new so that cut out a few topics. However, I still sat staring at their catalog for an eternity before deciding on one that was interesting/potentially useful (Lift in Action).
I fundamentally agree with Norvig's idea about how long it takes toward being a programmer. But this approach sort of makes sense - or just maybe I am reading too much into the software-commodity angle.
Either way - I cautiously approve this. This is similar to how modern warfare utilizes recruitment-training-deployment to weed out the unworthy from the able. Adapt or perish. Nothing wrong with that. The worst thing that can happen is that the developer in question may realize that he is just not that capable and venture into some other areas.
I realize I am reading HN but sometimes programming is treated as too much of a sacred cow.
I did something similar when I got started and yeah it got me a job but that was just the beginning, now, about 2 years later, I'm still learning and pushing myself to learn lower and lower level things about software development (as I started with high-level, dynamic languages) as well as more about design patterns, etc. Its definitely true that there's always more to learn so never be complacent. Its definitely possible to learn and apply the basics of CRUD development in a few months but IMO this is only the tip of the iceberg to having a fulfilling career in software development.
I applaud you for all the things you've achieved, it's pretty commendable ( always like to hear stories about people working hard and getting stuff done ). But the truth is, it wouldn't hurt to go and take the time to re-visit the topics he mentioned and tie up loose ends. It would help your career regardless of what was said here. 2 cents.
What's ridiculous about your comment and rorrr's trollish one is that he probably already has (re-)visited them. He writes that he watched 13 lectures of the MIT intro to computer science OCW class. That definitely touches on data structures and algorithms.
Have you made money as a result of code you've written? If so, you're a professional programmer. I wouldn't worry about whether you conform to a checklist an arrogant troll (I don't use these words lightly, but check his comment history) wrote on an internet forum. Hats off to you for doing it, especially under so much pressure.
Those are valid points, but I think they miss the purpose he was driving at. Maybe the title of the article could be different, but that's a minor issue.
Bottomon line: He wanted a job as a developer. He got that job. Success.
You're right though, people do have to learn these other concepts, but first they have to have some money to pay the bills, then they can just knock themselves out learning all about algorithms and data structures.
I know from professional experience that he can go from zero to competent very quickly. He will keep that job exactly as long as he wants to unless the company doesn't know how to manage their resources.
Very nice. When I first learned Django this year (after a year in Python and several years doing web development) I saw myself as an all-encompassing guru of awesomeness. A few months later I realized how big my knowledge gap is of the framework, and interviews helped me see how much I still have to work on. Easy to feel good on step#2; after that comes the true trial, and they go into cycles. Then you realize you did your 10 years and you still know very little LOL
Though I tend to agree with you, I think it goes both ways depending on how restrictive your framework is (and whether or not you work well with restrictions).
Yeah of course it's harder, because it's built on top of a platform you don't understand. It's easier to make good design decisions in Python if you come from a background rich with C experience, right? Because you understand what's going on under the hood. Without that experience -- like me -- you just fumble around like a big stupid klutz until you realize one day that you need to learn the lower level stuff.
So naturally when you're talking about dealing with an abstraction layer built on another abstraction layer it's going to be even hard to understand. That's why I say don't learn the framework. Learn the language, the concepts, the principles. It will make understanding the complexities easier to untangle.
I have been dabbling with Ruby and other languages for a year, but have never had this sort of discipline. For a couple years I have wanted a programming job but have always had serious doubts about whether I can really "hack it" as a professional. I don't know any programmers personally (well, one friend on the other side of the country in SF, heh), and I look at job openings and their requirements / desired experience all seem so intimidating. :-/
My mantra for the past 2 weeks has been that there's never been a better time to learn to code, simply because the resources are so abundant and available.
I'm self-learning web development right now. Between HTML Dog, W3 Schools and Google, I have more than enough information. I wanted to play around with WordPress-fired up a VirtualBox VM with Fedora, and an hour later had completed the 'famous 5-minute install.' When I begin to take on client work, most sites can be hosted on a cheap Linode or Bluehost.
Most of the information you need is free. Infrastructure (if you need it) costs peanuts. You have nothing to lose. Why not start now?
To offer a counter-point, I was running a consulting business at 17 (OK, a little later) went from hack to hack and didn't even know I was failing to learn and improve myself. Ran up debt, and went through what in retrospect looks a lot like bouts of depression. Now, I'm going on 30 in my first "real" job and I feel like I'm making up for lost time.
I don't do something unless the profits are clear. I once had $3,000.00 stolen from me, so I've learned early on to avoid anything too risky. And I'd rather make pennies doing something I love than dollars doing something I hate. It's absolutely different for everyone! But I'm confident you've probably learned a lot from your endeavors and have more respect for money. It's an interesting thing when you first start dealing with money...yikes...sometimes we have to learn the hard way.
HTML Dog is absolutely awesome for learning, but just make sure you follow that up with what's different in HTML5. It'll make your life a bit easier. net.tuts+ and webmonkey are great places to get up to date. I've been meaning to try out teamtreehouse.com too, but just haven't gotten around to it yet.
Right now I'm trying to learn Python using Zed Shaw's Learn Python the Hard Way and The New Boston tutorials. I think once I'm comfortable with the language I'm going to try to build something. It seems like the best way to learn how things really work.
> I was a computer geek from way back, had a few BBSes in the late 80s (yes, I'm a child of the 80s & 90s), learned QBasic & VisualBasic back in the day, and tinkered with Python for a few years off and on...
In every single one of the "how I learned to program in X days" stories, the author had some previous experience. I wouldn't count running a BBS and learning 3 programming languages as "no previous experience".
Learning any programming language, especially as a youth, puts you in a completely different class from someone who has no prior exposure to computer programming and its core concepts.
I'd argue that it still predisposes you very strongly. If nothing else, just in the abstract sense that you are familiar with the computer and don't see it as some scary or magical thing.
I "learned" Clipper and QBasic when I was 10. I can't write a single line in either today, but I'm 100% sure that experience made it a lot easier for me to become a developer. It's like riding a bike.
Knowing what a for loop is already puts you miles ahead of the pack. It implies knowledge of variables, counters, operators, etc. We take many concepts for granted that are not that obvious to most.
As I understand from the article, he didn't want to just get a site working but to learn how to program and picked, in my opinion, a good language to get started in.
Learning django was, I assume from reading the article, a way to build fun projects with the knowledge of python he had acquired.
Learn a language then how to make web pages with it is a better way to go in my view.
I understand his perspective. I dabbled in programming as a child and teen, and then was an advanced user, running Linux on my desktop for years. At the time I started actual programming, I knew what a 'for' loop was, but not how to write one in any language - even Bash. I didn't know what an array was, or say, the difference between C and C++, to say nothing of functional and object oriented concepts.
As far as I'm concerned, you can become a programmer in 12 hours. That's easily enough time to learn enough Python so you could start writing programs that satisfy your own needs, if they were simple enough.
Of course, there is no upper limit to how good a programmer you can become. We can only (maybe) set lower limits.
I don't understand it. Some of the people posting comments on this topic are acting extremely pompous. The word 'programmer' on a resume does not mean you're in a elite club or society. It means what the job title implies: a programmer is someone who writes computer programs. There are all different variations of skill level though, just like in any profession.
It's probably because you scare them. Here are people who have spent several years in school to gain a certificate that shows they know how to program. You may never be a competition to them in the job market but the knowledge that someone spent close to 12 weeks (not even a semester) and was able to land a basic job programming is probably rubbing their sense of entitlement the wrong way. Me personally, I have a M.S. in Computer Science but I work with a group of people, who, with sheer will and hard work, are in much better programming situation than I am. Not everybody gets to, or needs to, write a compiler or an operating system. Most of the applications are crud applications. Just because you are not of Linus Torvald's caliber doesn't mean that you are not a programmer.
I share your sentiment. It's disrupting their entire way of thinking regarding the programming field.
It would be amazing to get to the level of programming enlightenment to be able to create a compiler or kernel, though. That's a goal of mine with-in a decade.
I am certain most HN readers who are expressing negative sentiments here and who have spent several years in school feel very threatened by a Django coder.
Don't be too surprised or confused. There are a lot of very bright people here, but remember that being bright in one area doesn't necessarily mean they'd be bright in other areas, or nice, or humble. I guess someone who spent n years working at their craft and feels that he earned the title "programmer" might get offended when some young punk says he learned how to program in 12 days. I don't feel that way but I can understand why someone else might. I think if you had said something like "How I learned enough about programming to get a job doing it in 12 weeks", they might have been more encouraging (though it would have sucked as a title!). I think it was the title more than anything else that put them off.
And don't be too disappointed. I think the majority of the people here aren't nasty. They might have said some not so nice things in a moment of pique for the above reason, but the nice thing about HN is that most of us do try to be objective and helpful. Don't be surprised if in another thread, the very same people who dissed you here praise you, help you or openly agree with you. I usually don't even notice usernames when replying or voting on posts.
Lastly, congratulations on taking a very big step on the same path most of us here are on. The path is long and difficult at times, but infinitely rewarding. Welcome to the club.
My experience is similar to his. I have a business degree and worked in accounting, but decided I'd like to get back into programming after dabbling in it when I was younger. I wrote some Rails apps as side projects and put them on Github, which ended up helping quite a bit.
What made the difference in the end was good social skills. I met the CTO of the company I work for at a Ruby meetup, and had a good conversation with him. I ended up applying for a position he was trying to fill, and after an interview and a review of my projects on Github, I ended up with a development job without having any formal experience. I've been at it for over a year now.
So, this is definitely doable as long as you're committed to learning a lot, and are able to sell yourself well enough to convince someone to give you a chance.
The hardest part to that for me is the "get to know people" part. He already had a mentor before he started learning. How would one get a mentor in the first place? How do you get to know other people?
To be fair, the friend/mentor in question started out as a dude on a small private forum I used to frequent. The forum had an AIM chatroom and he made fun of me mercilessly me for years for making stupid decisions.
Go to meetups and lurk IRL IMO. Just start small conversations. Say hi, ask what someone's working on, what the program is they have open, if they read HN or Reddit or whatever. Let them know you admire their work and ask for help.
There are so many people in this world who pretend that something is preventing them from learning or doing something and you went out just did it.
You may (or may not) be the best engineer in the world, but I'd hire you in a hot second just because you have the tenacity to make it your mission to figure something out. Sometimes, 90% of writing code is just sticking with a problem until you get it.
I've given 100's of interviews to people who are either fresh out of school or have been programming for a while now. The vast majority of these people have no gumption to actually create a github account or do any sort of experiments outside of what tasks they are given at work. It is very sad to me that most of these people just want to collect a paycheck.
What I think is great is that he learned enough to get an 'in' in a relatively short time. Obviously getting from knowing some Python and being able to work with Django to being a master developer will take years, but it is so much easier to do that if you already have a fulltime programming job.
If you're trying to learn all the skills that a great developer needs in your free time, you'll realistically be able to spend no more than 4-5 hours a day at it over any sustained period. If your fulltime job is software development, even if it isn't hardcore by HN standards, then you will grow your skills so much faster.
189 comments
[ 3.5 ms ] story [ 144 ms ] threadhttp://norvig.com/21-days.html
This ought to be worth something too, though.
So, I'm sure the work ethic that let him join and last 10 years in the USMC helped.
That said, I'm inclined to agree with your last statement...
What are you saying that we do poorly at?
Your comments are more "confounding" to me than any war situation I was ever in.
But if you have any citations to the contrary - that veterans do better than one would predict from their pre-military life - I'd be glad to hear them. Conscientiousness is as important as IQ, and any intervention which genuinely improves either is really valuable and ought to be well known.
You can't switch from studies and causation/correlation to just stating without citation that they suck at kicking ass.
http://lesswrong.com/lw/ih/absence_of_evidence_is_evidence_o...
Frankly he'd have bought me if he made his hands dirty with some obscure 8bit assembler. There he'd have to really THINK his way out.
At least some sweat during rolling his own data structures and algos around them would make him something in a shape of a programmer, but not this.
Of course what he's done is cool, learning anything new is worth some pat on the back. But this is a newsfeed on hacking, and Lords, it ain't hacking. Let's call things what they are, ok?
There's also the problem that some people like him start calling themselves programmers on the job market and when we're trying to hire them, they occur to be only Django programmers with very limited abilities. Practice makes perfect, but they didn't do it yet. Have some interviewing experiences like that. Not funny.
"Well, thankfully, I didn't write this for you. I wrote it for the people -- I suspect the vast, vast, vast majority of the HN readership -- who are not expert programmers, or even professional programmers, but they want to be."
If he had started with some 8bit assembler I would simply think that he has a very poor approach towards learning.
Why is this on the front page btw? Aren't we all hackers here? What makes this news?
If all you have is a strong understanding of algorithms and computation... your just a computer geek. Doesn't make you a good 'programmer' (in the sense that you can produce something someone wants).
In fact, how did mathematics and physics even develop at all before then, without knowing what was hard, or how to solve problems.
Seriously, a lot of software doesn't require complex algorithms, sure there is a subset that does and that's cool. But saying someone isn't a programmer because they don't have a deep understanding of computation and algorithms is disregarding a large population of quality programmers.
I think y'all are all using different terms (and I notice the parts-of-HN undercurrent of "technical skills don't matter, you need to be social and build products, startup pivot founder startup" but I digress).
To me, someone who can use algorithms et al is a "programmer", which is largely orthogonal to being able to produce something someone wants, call them a developer or something. Not every programmer has to be social client facing; getting detailed functional specs and creating good code to fulfill those specs absolutely does make one a programmer.
"It's not important to get things right/perfect. It's important to get things going"
Congratulations and realize that the nth step in learning is realizing that you can always learn more.
Not realizing that fact held me back for years.
When I wanted to learn / do something, I would obsess over finding absolutely the best tutorial / process. Then I would obsess over the right learning environment. Always worrying about the incidentals, and ignoring the essentials.
HN has helped me a lot in this regard. Reading about big companies and how 'imperfectly' they started out, as well as the whole 'Lean Startup' concept has done much to change my mindset. What is important is starting.
Thanks for sharing.
A lot of good advice for people who would like to program rather than be programmed but aren't in a place to take a cs degree.
Call him a programmer or not, it seems he has learnt enough to program professionally and by most definitions, that makes him a programmer.
But kudos on taking your first few steps. Its not a bad vocation.
Then earlier this year I started doing Hartl's Ruby on Rails tutorial and things just started clicking. Now I am teaching myself Ruby on Rails and I am looking for explore other more difficult languages (just for fun, I don't want a career as a developer).
For whatever reason, Ruby just got me more interested in programming. Now I am dipping my toes into SICP, which I would have never have guessed I would be interested in.
I think once you get your hands dirty with something relatively easy like Ruby on Rails, you start seeing the need for all the other computer science concepts and languages and that piques your curiosity.
However, I can say that teaching myself a little coding has really helped out in my day job, I notice I think differently about problems. My thought process is much more structured.
I answered a questionnaire for Manning in exchange for a free book recently. I wanted to learn something new so that cut out a few topics. However, I still sat staring at their catalog for an eternity before deciding on one that was interesting/potentially useful (Lift in Action).
Either way - I cautiously approve this. This is similar to how modern warfare utilizes recruitment-training-deployment to weed out the unworthy from the able. Adapt or perish. Nothing wrong with that. The worst thing that can happen is that the developer in question may realize that he is just not that capable and venture into some other areas.
I realize I am reading HN but sometimes programming is treated as too much of a sacred cow.
Data structures are not there either.
Sorry, you're a django coder, not a programmer.
Bottomon line: He wanted a job as a developer. He got that job. Success.
You're right though, people do have to learn these other concepts, but first they have to have some money to pay the bills, then they can just knock themselves out learning all about algorithms and data structures.
It's not a protected title. However I doubt the guy in the article could write his own OS. Which I can.
Also keep in mind the phases of learning and understanding:
1. I know nothing...this is so hard.
2. I know everything...this is so easy.
3. I know nothing (again), but this time for real since I realized there is so much more to learn.
4. I really, really, really know nothing.
5. Hmm...maybe I might be getting the hang of this.
And in phase 5 other people start calling you a guru. However you'll have no problem finding a job in phase 2.
Learning frameworks is harder than learning programming languages, IMO, especially after you lean one language. You should learn both.
So naturally when you're talking about dealing with an abstraction layer built on another abstraction layer it's going to be even hard to understand. That's why I say don't learn the framework. Learn the language, the concepts, the principles. It will make understanding the complexities easier to untangle.
I have been dabbling with Ruby and other languages for a year, but have never had this sort of discipline. For a couple years I have wanted a programming job but have always had serious doubts about whether I can really "hack it" as a professional. I don't know any programmers personally (well, one friend on the other side of the country in SF, heh), and I look at job openings and their requirements / desired experience all seem so intimidating. :-/
I'm self-learning web development right now. Between HTML Dog, W3 Schools and Google, I have more than enough information. I wanted to play around with WordPress-fired up a VirtualBox VM with Fedora, and an hour later had completed the 'famous 5-minute install.' When I begin to take on client work, most sites can be hosted on a cheap Linode or Bluehost.
Most of the information you need is free. Infrastructure (if you need it) costs peanuts. You have nothing to lose. Why not start now?
There's no magic right way to do it.
I felt this way in 2001 when I started to teach myself HTML. There were plenty of free resources available online to learn.
The free education available now is exponentially more abundant. I would expect a similar leap during the next 10 years.
Right now I'm trying to learn Python using Zed Shaw's Learn Python the Hard Way and The New Boston tutorials. I think once I'm comfortable with the language I'm going to try to build something. It seems like the best way to learn how things really work.
In every single one of the "how I learned to program in X days" stories, the author had some previous experience. I wouldn't count running a BBS and learning 3 programming languages as "no previous experience".
Learning any programming language, especially as a youth, puts you in a completely different class from someone who has no prior exposure to computer programming and its core concepts.
Knowing what a for loop is already puts you miles ahead of the pack. It implies knowledge of variables, counters, operators, etc. We take many concepts for granted that are not that obvious to most.
Learning django was, I assume from reading the article, a way to build fun projects with the knowledge of python he had acquired.
Learn a language then how to make web pages with it is a better way to go in my view.
I
may be the times have changed and thats not true any more ..
Of course, there is no upper limit to how good a programmer you can become. We can only (maybe) set lower limits.
It would be amazing to get to the level of programming enlightenment to be able to create a compiler or kernel, though. That's a goal of mine with-in a decade.
And don't be too disappointed. I think the majority of the people here aren't nasty. They might have said some not so nice things in a moment of pique for the above reason, but the nice thing about HN is that most of us do try to be objective and helpful. Don't be surprised if in another thread, the very same people who dissed you here praise you, help you or openly agree with you. I usually don't even notice usernames when replying or voting on posts.
Lastly, congratulations on taking a very big step on the same path most of us here are on. The path is long and difficult at times, but infinitely rewarding. Welcome to the club.
What made the difference in the end was good social skills. I met the CTO of the company I work for at a Ruby meetup, and had a good conversation with him. I ended up applying for a position he was trying to fill, and after an interview and a review of my projects on Github, I ended up with a development job without having any formal experience. I've been at it for over a year now.
So, this is definitely doable as long as you're committed to learning a lot, and are able to sell yourself well enough to convince someone to give you a chance.
Go to meetups and lurk IRL IMO. Just start small conversations. Say hi, ask what someone's working on, what the program is they have open, if they read HN or Reddit or whatever. Let them know you admire their work and ask for help.
There are so many people in this world who pretend that something is preventing them from learning or doing something and you went out just did it.
You may (or may not) be the best engineer in the world, but I'd hire you in a hot second just because you have the tenacity to make it your mission to figure something out. Sometimes, 90% of writing code is just sticking with a problem until you get it.
I've given 100's of interviews to people who are either fresh out of school or have been programming for a while now. The vast majority of these people have no gumption to actually create a github account or do any sort of experiments outside of what tasks they are given at work. It is very sad to me that most of these people just want to collect a paycheck.
The learning never stops, keep it up.
If you're trying to learn all the skills that a great developer needs in your free time, you'll realistically be able to spend no more than 4-5 hours a day at it over any sustained period. If your fulltime job is software development, even if it isn't hardcore by HN standards, then you will grow your skills so much faster.