OStatic

rPath CTO and Founder Erik Troan on the Future of Open Source

The Open Source Business Conference (OSBC) is coming up, to be held March 24th and 25th at the Palace Hotel in San Francisco. The conference will include much discussion on the future of open source. In preparation, OStatic has been  running a series of guest posts on this theme, featuring thought leaders from top open source projects. We checked in with Dries Buytaert, founder of the Drupal content management system, and co-founder of Acquia. Martin Schneider, director of product marketing for SugarCRM, weighed in on the open cloud, and Novell VP Justin Steinman wrote about open source and mass customization. Brian Gentile, CEO of Jaspersoft, also discussed the consumerization of information. In this latest installment in the series, we have a guest post on the future of open source from Erik Troan, founder and CTO of rPath, which offers a platform for software vendors to create and maintain software appliances and virtual appliances in multiple virtual machine formats.

The Future of Open Source: Scaling Developer Productivity

By Erik Troan, founder and CTO, rPath

Open source began its life serving developers. Some of the first few open source projects were tools like emacs and gcc. Richard Stallman was the father of both, which were aimed, essentially, at himself and people like him; developers who needed a quicker way to get things done. It turned out to be a clever target. He was living in the world of UNIX, which had lots of developers. It also had developers who weren't used to paying for tools, because UNIX had traditionally shipped with a full set of development tools. UNIX vendors didn't make any money off of those tools, so they were happy to see the Free Software Foundation come along and offer alternatives. In fact, many of them were happy to fund it to do so!

Simplifying things greatly, Linux and Apache came along and shifted the focus of open source development towards production applications. Of course, Linux is now used by companies of all shapes and sizes for mission critical applications. The Apache web server is used by, well, basically everyone to make the web work. Those projects, and others which complete open source operating systems, have transformed how companies procure, deploy, and manage production infrastructure.

Now that Linux and its related projects have become so standard, an interesting change is occurring: More and more open source projects are once again targeting developers. Not developer tools primarily (though Eclipse is a notable exception), but development frameworks and components. Projects like Spring, Ruby on Rails, and Django are fundamentally changing how developers build applications of all types. Software development is increasingly about finding the right open source projects to build around and writing just enough code to tie those projects together.

That "glue" code is where the intellectual property lies. Anything that is written but isn't intellectual property is probably either a waste of time (because there is an open source project which does the same thing, only better) or a good candidate to be open sourced (letting enough eyes make the bugs shallow [shamelessly borrowed from Eric Raymond]). This kind of reasoning as led to the large number of open source Java components that have gathered under the Apache Foundation. There is simply no reason for companies to shoulder the full development costs for code that contains no intellectual property. XML parsers just aren't that interesting.

It's interesting watching how this plays out in corporate America. Instead of monolithic applications that just need compiling, modern applications now depend on (typically) hundreds of various open source projects. The value of the code that the companies have written has never been higher (because it's all IP!) and the speed at which applications are being developed has increased dramatically. Projects like Spring, Ruby on Rails, and Django are all examples of this phenomenon, and I expect this trend will continue for a long time. By allowing increasingly rapid development and a strong focus on the business-critical portion of applications, open source will have its greatest impact yet.

What are your thoughts on open source's future? Please take a moment to fill out the Future of Open Source survey here and share your perspective.

The results will be announced at the Open Source Business Conference on March 24th–25th at the Palace Hotel in San Francisco, CA.

• Apache
• Eclipse
• rPath
• OSBC
• Erik Troan

Share Your Comments