Fails on Rails
Todd and I met with a couple guys last week who are in the process of bootstrapping a web startup, specifically, a web startup using Ruby on Rails. First off, I need to say that I love talking with people who are trying to bootstrap, especially when these people are in Chicago. We have such a huge amount of talent around here, but too many of us are unwilling to take the red pill and be measured. But I digress. During our discussion, I was trying to explain to the non-technical, though passionate and inquisitive, startup founder the advantages of using Rails. I found myself telling him something that surprised me:
The total cost of development for a successful Ruby on Rails system is probably not that different from a successful Java EE system.
To be clear, I have absolutely no data to back that up, other than my gut instincts after working on Java and Ruby projects for the last 7 years. The point I was trying to make to the startup founder was that while total development cost might be the same, the time to market with Rails was much faster. A startup founder who chooses Rails is putting a priority on getting her vision out into the real world as soon as possible. Once her idea is out there, the work is far from over, though. Ruby is a relatively slow language, and that means that as you have success, you are going to need to optimize or replace the bottlenecks in your system sooner than you would with a faster language. If you’ve built your system correctly, though, these improvements should be relatively straightforward thanks to Ruby’s amazing conciseness and flexibility.
As Todd and I were walking back to the Studio, I was thinking about what I said, and I realized there was a flip-side to my first assertion:
The total cost of development for a failed Ruby on Rails system is much less than a failed Java EE system.
There’s a good reason this line of thinking didn’t come to me while we were talking with the startup founder. The guy was infectiously enthusiastic about his business, and talking with him about its failure didn’t even occur to me. But I think that like Agile Software Development, Ruby on Rails gives visionaries the opportunity to fail fast, learn from their first attempt, and regroup for a second effort. Or in the less likely event that they succeed, Rails is no less capable than any other platform to be optimized, tuned, or chunked into separate sub-systems in order to scale.
