« Your stack - my stack? | Main | Ricky Gervais podcasts »
December 08, 2005
Open Source can commoditize too
While there has been enough written about the commoditization of software and how that leads to Open Source gaining traction in those areas, let's focus for a moment on what may be the commoditzation of Open Source projects themselves.
I've been following the ongoing discussions on the Apache Cocoon mailing lists and as a long-time "fan" of the platform, I think a lot of what we are seeing there is actually what we should be calling "commoditization". Something natural to any software solution, be that a commercial product or an Open Source project.
Or, as Geoffrey Moore would say, too much of the Cocoon core has now become context in the project. More and more resources are now needed to maintain the context and there is not much left for the core. This also means that developers, those that once drove the platform forward, have basically lost interest and are turning slowly to the "shiny" platforms like Ruby On Rails.
Over five years ago, Cocoon entered the marketplace with a radical idea - to separate concerns in the building of web-sites. Use XML and XSLT to format data into various formats - as needed. Back at the beginning of 2000, Cocoon was a new and shiny way of doing Web publishing. And over time the platform became more and more popular. As the popularity grew, so did the "itch scratching" and Cocoon became "enhanced" with a portal framework, forms framework, mvc underpinnings and much more.
But the cost of maintaining the beast also grew over time and adding new shiny things became more and more difficult. Also, the project grew self-centered and for a long time dwelled on Cocoon's virtues and ignored the growing popularity of frameworks like RoR.
The "innovator's dilemma" set in and in a way Cocoon tried to flee upmarket by adding more and more enhanced functions. Functions that also became increasingly hard to understand and master. Newbies into the Cocoon world are today overwhelmed with the plethora of APIs and "ways of doing things".
At first I thought the new discussion on the fate of Cocoon was a bad thing and the end of the world was near. Of course I realize how wrong that initial thought was. In many ways the ongoing discussion is exactly what is needed.
Why? Well, for one thing the current situation shows that Cocoon has reached a level of maturity that makes it maybe boring for geeky developers - but that makes it ideal for a platform that can be built on. Cocoon itself could now become the commodity that other successful platforms are built on.
On the other hand the Cocoon community will eventually turn to building the "next new thing". And that is the way it should be. The community is the strength of an Open Source project - not the code. The community with its tight-knit network for unique personalities can move forward and build something new. Using the knowledge gained in moving Cocoon forward through over 5 eventful years, I am sure that whatever results from the current discussions will be exactly what is needed.
Just as Cocoon was so many years ago.
It remains important to remember that an Open Source project is just as prone to becoming a commodity as a commercial offering is. This is not a bad thing. It is a good thing.
Bonus Link - Stefano posts his thoughts on the current Cocoon development.
Posted by Matthew at December 8, 2005 09:57 PM