Hacker News Reader

I can relate to your sentiment. But note that your problems have very little to do with learning how to program, and everything to do with needing to navigate the high buy-in tax of learning all of this tooling before first building a simple site.

I HIGHLY, HIGHLY, HIGHLY recommend you give Steve Huffman's course on "Introduction to Web Development" at Udacity at shot. Unlike virtually every single resource I have found, Steve does a great job of explaining the basics of the infrastructure of how a website works while you are implementing it at the same time. Also, since he does it on Google App Engine, 100% of the "tooling" you will need will simply be installing the app engine runtime. I've found that it is a fantastic introduction to programming for the web. It will give you a solid understanding of the basic infrastructure, and you will then have a good mental model onto which you can slowly graft the many other tools that developers use when making a site.

I felt exactly like you did - and in my case, I had even programmed some in college but found it hopeless when trying to make a "real" product. Today I weave between backend and frontend technologies, preprocessors for my styling, grunt, etc. -- all of the stuff that I was put off by at first. And I'm confident it wouldn't have happened if I hadn't taken that course first.

We built www.lite-engine.com partly to fill this need; learn web development without admining servers or dealing with the backend language slash framework holy wars. With our paas you create web apps with html, css, JavaScript and sql. Great for learning these technologies an rapid app prototyping.

If you prefer to approach it from a full understanding of the fundamentals, like how jQuery instructions get translated to result in the browser, it's going to take a very long time. People go to university for four years and many don't understand fully.

On the other hand, if you can ignore your tendency to understand things and limit your scope to shallow understanding of the top 1 or 2 tiers of the stack, then you can learn enough to be useful within a short amount of time.

I'm wondering if Skype mentoring might help you. If you're serious about this, shoot me a message.

I think that by far the single most important thing is to start the "learning to code" process with a specific project in mind.

That sounds like you're using a tutorial aimed at developers from other languages. You can get started with much less and you may want to look around a bit more for things that ramp you up more gently, eg starting with serving static content to dynamic content from parameters to databases to the various bits of front-end tech.

At some point in software you will realize that there is more going on than you will ever know in detail and the best you can do is understand the salient points and some things are simply not worth learning.

...maybe start with an easier language set and then transfer the skills....even as a 10 year developer I found ruby weird/challenging to pick up and all the new web development languages/paradigms. It's taken me a good month! I couldnt really imagine starting out with ruby tbh.

If you learn java in depth - php will be easy, java script will make more sense. Having learnt one language in depth, picking up most other languages are easier, you'll be in a better position to understand how jquery works.

Omg, not java for sure. The important bits will be burried in the boilerplate code. How about starting with NO language at all. Learn what variables are, what assignment is, how do control structures work. And hey, you've just learned to programm. At least I learned this way. We had no access to computers so the lessons were more or less what I listed above. On paper. And when I did get access to the computer I already knew how to program. The only thing I needed was to find out which MSX-BASIC instructions are for the things I want. I. e. I had just to learn a syntax, not the concepts. Having mastered concepts is also very transferable, especially between the languages of the same family.

>Omg, not java for sure. The important bits will be burried in the boilerplate code.

So what? It doesn't make any difference for a beginner, I'd say it actually helps beginners since it makes programming a easier to approach.

>How about starting with NO language at all.

I hated this kind of advice when I was new, I followed it and my hatred has not subsided at all. Just pick a language and learn this stuff as you go, it will make zero difference and will save you a ton of time.

I think it's the individual differences between deductive and inductive learning styles at play here. Inductive learners, like myself, prefer to see examples and derive general rules. Deductive learning style is more about learning the generality first and moving on to examples.

Different strokes for different folks.

I too learned like you, I first learned Perl by trial and error example and the wonderfully awesome Learning Perl. I learned Java by frustratingly going through it, but once I got it I got it. Learning object oriented Perl and Python feel trivial now, thanks to that experience. Python, btw, feels like it needs to be practiced instead of learned, since I already have the concepts and generalities down.

The question of this thread is, how do we teach beginners the concepts in an interesting way. I realize how much of a genius Larry Wall and Randal Schwartz are when I see threads like this.

You can learn all those things you've just said, in Java, very little boiler plate. Java to me, is a mixture of all languages, or most languages have at least elements of java in them. Hence useful to start with, picking up others are easy.

I think the limiting factor for most people is motivation, not intelligence.

Reading books and doing tutorials will certainly help you, but you might get burned out quickly. Working on projects seems like a much better use of your time.

Eventually, you'll have to pick up some books and learn theory, but the main challenge is getting to that point where you'll find the theory useful.

> I think the limiting factor for most people is motivation, not intelligence

I think the limiting factor especially if one is learning on his or her own, is grit and persistence.

I started at zero and Hartl's tutorial was my first try. I got stuck too. Ruby &/or Rails has too much magic and doesn't teach you how things work. I switched to Python and Django. Python has tons of beginner learning material. Django was easy to pick up after going through Hartl's Rails Tutorial. Best Django material for me was "Tango with Django".

That is interesting because I did it the opposite way. I started with python and switched to ruby. Ruby IMO has more introductory material than python and I was terrified of the idea of reconciling python 2 and 3 which many books I found alluded to. To be sure, part of the problem with learning to program by ones self is that a lot of background material is lacking and the toolbox is incomplete. For instance, what is an API and how does one learn about it? What are threads? What does runtime really mean? Command line invocations and PATH? ASCII? and so on for all the entities one will require.

Learning programming and how IT works overall is like assembling a huge puzzle. At first you will struggle, but every now and then you're gonna fit in a new piece and eventually you will start seeing the big picture.

it sounds like php might be better for you to learn on. You can put small snippets of it directly into html and have it be immediately useful without learning a bunch of other stuff.

While it was nothing to do with websites, check out "Programming from the Ground Up". I think the title explains it all :-) It starts with assembly and ends with C.