I believe that Ruby on Rails hallmarks the next era in software development. But before I explain the scope and significance of my claim, a little background and history of software development is necessary.
The C programming language marked a new generation for software development. It was a high-level language, compared to assembly, which made software development a viable investment on an unprecedented scale at the time. Unix was developed with C and it created new opportunities for computing; new markets and potential beneficiaries of technology were born with the advent of Unix. (While there were other languages at the time that also opened new markets for computing, such as Pascal, I am using C as the “poster child” for this era.)
But application development was very expensive with C. There were many bugs, sustainable development was hard and overall the costs made it difficult to realize a return on technology investments. (Unix, because of the scale of its adoption, made the return worth the pains and costs.) Many tried to develop various applications and found it daunting and not economically viable with C.
Java brought on the next revolution. As the next generation, higher level language, Java caused a new market to explode: enterprise development. Before Java, IT development was fairly limited to few companies with very large budgets. While there have been many incredibly expensive Java projects, the number of participants in computing erupted; even small organizations with a few employees invested in IT projects.
While Java development had a lower cost profile compared to C, it still was high. This culminated in the mid to late 1990′s with the Standish Group’s Chaos Report(first published in 1995). This landmark report noted that 5/6 software projects fail. It also described a variety of other disturbing statistics. In short, many organizations were investing in software and most of those investments failed miserably.
The rise in outsourcing in the late 1990′s and early 2000 was in response to the economic environment of software depicted by the Chaos Report. I think most of the industry was throwing up their hands — “If we can’t figure out how to build software, let’s make projects cost less and make a portfolio management play.”
But outsourcing proved not to be a panacea. In a somewhat parallel track from enterprise, and in the aftermath of the Web 1.0 bubble, a new interest in consumer oriented applications took hold early this decade. PHP marked another new era in software development.
The LAMP stack (Linux, Apache, MySql, PHP) offered a new paradigm for development. Thousands of small, focused web-based applications sprang up. Two developers in a few months were accomplishing what twenty developers couldn’t achieve in a year. Pragmatism drove ideas and it was being demonstrated that small teams could be incredibly productive in short time frames. The economics of development changed once again.
But PHP has problems. First, it is limited to website development. This makes it hard to adopt for general enterprise software, as integration can’t be effectively done with PHP. Second, it has provided a great option for first version websites, but at least in my opinion, PHP does not provide for a robust language/environment for sustainable software development; this is a must for enterprise.
The breadth and volume of topics to which software is being applied is growing. This creates tremendous pressure for the economics of software to be manageable. Furthermore, the line between enterprise software and consumer-oriented web 2.0 is getting more and more blurry. I think the market needs and demands a technology option that can be used by both markets and is economically viable.
This brings us to Ruby on Rails. Ruby is an elegant and mature programming language. Object-oriented geeks tend to love it and feel empowered to create software with it. It is also a disruptive innovation (as described by Clayton Christensen) that changes the economics of software; Rails brought forth a new vision: opinionated, convention based platforms can dramatically reduce the cost of development. I think Ruby/Rails can and will replace Java as the language and platform of choice for software development in the enterprise and will similarly establish itself as the premier option for Web 2.0, cementing the bridge between both markets.
There is much work to be done on Rails though. There are scalability issues and integration patterns are immature (on average, at least). But compare Ruby/Rails to where Java was in the 1990s and I think Ruby/Rails is much further ahead in its technology and adoption lifecycle.
And what about the alternatives like Python and Django? There is some debate over this question and that can be a series of posts by itself. For the moment, I will address this with the following observation: at Pivotal we have many developers who know Ruby and Java and the same developers are several times more productive in Ruby. In this comparison, my money is more on Ruby compared to Python than Rails compared to Django; I can’t see how we can be as productive with Python as we are with Ruby.
The momentum of Rails is not be underestimated. Geoffrey Moore, in “Crossing the Chasm”, describes how markets need, elect and will support a market leader. The need for manageable software development costs, both in terms of upfront and sustained development, create tremendous pressure, that combined with the current momentum of Rails, will drive forward the adoption and development of a rich Ruby/Rails ecosystem. As Rails becomes more and more established, its position is harder and harder to derail. (Sorry, I couldn’t resist the pun.) Think about Apple’s iPod: it isn’t enough to offer a better mp3 player as the market has invested in an iPod ecosystem of speakers, docks, cases, etc. Similarly, Rail’s position only continues to be secured.
In short, as is the case with many things, economics will dictate the needs of software and will establish the supremacy of Ruby and Rails.