Reinventing Wheels
It’s been about two weeks since I started writing the ruby on rails app for this blog. I recently took a look at some of the established RoR blogging engines and this feels more like a somewhat circular rock. Even though this app is lacking a lot of the regular features of a blog, I’m extremely happy with how it turned out.
Build to Learn
One of the things that I realized early in my programming career was that reading about how to do something was nice, but I never felt like I really understood it. However, when I started actually building something everything started falling in place. I think this is a pretty standard realization for a lot of developers and it’s always something I keep in mind when learning a new technology. Enter Ruby on Rails. Even though I read a ton of stuff, watched lots of videos, subscribed to podcasts and purchased expensive books, I didn’t realize how much I was missing until I set out to plan, build and deploy this application.
Referencing the Reference Books
By a huge margin the best ruby on rails book I found was The Rails Way (non affiliate link btw). There was just something about it’s straight forward nature, clever writing and almost-but-not-quite reference book feel that made it so easy to use. There was a great post on the Signal v Noise blog the other day about years of irrelevance where David spoke about how it took him some time to get up to speed with ruby’s syntax, idioms and just how things worked. I think I’ve finally reached that point with rails, the more and more I used the The Rails Way the less and less I needed it. Combine this awesome book with building this app and I think I made it through the learning curve faster than I would have following tutorials or watching videos.
What I Learned
Here’s just a short list of things I learned while building this app. I plan to expand this particular concept into a series of posts.
- Using RSpec for BDD Testing
- RESTful controllers
- Ruby syntax and idioms
- Customizing and extending Rails
- Deployment with Capistrano
- Git
Sadly I had spent some time already reading about all of these, but never understood how far off I actually was.
When NOT to Reinvent
While I’m hailing the virtues of building your own stuff from scratch it’s important to know when “rolling your own” is a bad idea. First, if you’re doing work for a client chances are you’re going to increase your profitability by finding a solution that’s been built and tested and proven to work, do you really want to build your own Authorize.net processor when you can use Active Merchant instead? Secondly, if you don’t have the time and patience to work on your project with the understanding that you’re trying to learn as you go, you run the risk to slapping together something that’s a hodgepodge of bad code and shortcuts, which will be more trouble down the road.
App Delivers; Would do Again
Like I said earlier, while I don’t have everything I’d like in this app, I’ve learned more doing this than I would have reading about how to do this so I’m happy. I know everything about this app and can add to and improve it as time goes on. Perhaps most importantly I have the satisfaction of having built something that’s now facing the world which is one of the more underrated accomplishments in software development.
