On: Ask HN: Had an original idea for a start-up...but

[tsally]

I made this mistake for about a year. Ordered/downloaded tons of books, read published papers, etc. I learned a fair bit, but honestly you don't really learn until you start doing. Your return per hour spent coding is on average far higher than your return from books and papers. You shouldn't stop reading, it just has to be limited.

[jnovek → tsally]

My learning process usually works like this:

1) Learn the syntax/guts of a tool by going through a book, tutorial and trying trivial examples.

2) Work until I run into a problem. (How do I make merb-auth work with a flash uploader?)

3) Find examples/read a book/look at API docs until I figure it out. (Ohhh, you write a Rake middleware to handle it!)

4) Goto 2.

So reading materials are important, but I usually use them to solve real problems. There's no part of this process that doesn't involve writing code.

[MicahWedemeyer → jnovek]

Glad to see I'm not the only one that thinks like this. I sometimes feel like a fraud because I haven't read the latest-greatest book on X. Sure, I use X every day, but I don't understand every last teeny aspect about it.

[bkorte → jnovek]

Yeah, I don't think I've read a development book cover-to-cover, I always have something in mind that I want to build, so I stop the book half way and code.

I always skim the rest looking for an answer, but the point is that I will almost always get the best results just by diving in head first.