The value of your bootcamp came in its carefully curated curriculum. Instructors spend hours planning every aspect of the course: deliberating why webpack is easier to setup and understand than gulp, choosing the handful of node modules and ruby gems you’ll need, teaching a database that plays well with the data you’ll be storing.
As a result, you spend very little time on configuration and dedicate all of your precious time to creating. You can still namedrop webpack and babel and es-2015-presets, but a few months later when you decide to start a new project from scratch, you realize you have no idea where to begin (unless you choose your bootcamp’s MERN stack again).
The first sign comes with the StackOverflow developer survey – you notice the Node.js usage chart is relatively flat while Golang jumps 25 spots. Next you read myriad articles on Hacker News condemning Facebook’s React licensing and celebrating Vue. A large company using a NoSQL database where it shouldn’t causes MongoDB to lose favor among the developer community. WTF are containers?!
Inception of an idea
You hear a lot of buzzwords being thrown around your company and decide to learn some of the more important-sounding ones. After about a week of asking reddit for advice, you decide to build microservices with Docker that use tensorflow to trade cryptocurrencies. That sounds like a side project that would look incredible on a portfolio and give you solid experience with a bunch of hot technologies.
All right. Sounds good.
Let’s get started.
New episode of Rick and Morty first…
$ npm init
Pit of despair
Where do you even start? Docker? You follow a few tutorials but because everything is community-maintained nowadays, the steps aren’t always easy to follow and you sometimes get cryptic error messages. You find others on Github Issues who have seen the same error but the last update was four months (and one major release) ago.
Okay, what about Tensorflow? Surely a Google product must have immaculate documentation and an active support community? You do a little better this time – you manage to get through Tensorflow’s MNINST tutorial, but this only makes you more anxious. Maybe you need to take Andrew Ng’s machine learning course first? How many books do you need to read before you can do practical machine learning?
While your ego is already beat, you start thinking about cryptocurrencies. How much finance knowledge is required to understand arbitrage strategies? You’re definitely going to need faster computers to race against the banks already doing this.
Reliving the bootcamp
As rewarding as the bootcamp experience was, we need to forget about how fast we were able to learn while enrolled at the program – we were putting in 14 hour days completely immersed in the subject. We willingly gave up friends and family so we could consume all of the information over those three months. We were surrounded by 40+ peers who had the same questions and could be used as a support group and as mentors.
Now that we’ve graduated, we forgot about the miserable days spent staring at a recursion problem and wondering about base cases. We’ve forgotten about that one time we created a dumb rails migration and ruined our entire backend. All we remember is how quickly we learned and grew as developers.
Now, it’s just you and the internet. Except this time, you have the hours of 8pm – 9:30pm after work, assuming you haven’t made plans with friends or colleagues (and you aren’t too tired of typing yet). With your new schedule, you can expect a week of learning at this new pace to be equivalent to half a day during the course. It will feel slow.