The Changing Face of Open Source

By Andrew Binstock

January 1, 2006 —

The term “open source” may evoke a romantic image of programmers working late into the night on projects that have caught their interest. In the footsteps of Richard Stallman and Linus Torvalds, they selflessly toil away, contributing their work freely to the greater good, both in binary and source code form, all the while forgoing any remuneration. This enduring and endearing image, alas, is steadily losing any connection to the real world of open-source products as they exist today. And understanding the differences is key to finding your way around the open-source landscape, as we’ll see shortly.

First, the reality. Open-source programming today is almost entirely a commercial affair. On most projects, the majority of the code is written by full-time developers whose employers are interested in providing support to a specific project.

Earlier this year, I attempted to find out exactly how much these paid developers contribute when compared with independent contributors. I spoke with Marc Fleury, who heads up JBoss, the now-company that once built a full J2EE suite using the old-style community model with Fleury leading the effort. He told me that today, no more than 1 percent of the code comes from independent developers. Mike Milinkovich, who leads the Eclipse Foundation, speculated that the source-code contributions of independents was even less than that. (But, he pointed out, independents do make terrific beta testers.)

This shift from amateurs (in the word’s original sense of doing something for the love of it) to paid professionals has several causes.

The first derives from the complexity of programming today. Most developers must learn daunting codebases to find the routine to which they can add value. And even when they find it, making sure they are actually adding value is another complex process entirely. Only developers who really need the software feature have the motivation to consecrate so much time and effort to making small changes. (Large feature changes are, in many cases, insuperably difficult.)

Apparently easier tasks, such as providing documentation, also require more effort than volunteer communities are willing to provide. As a result, you get strange paradoxes, such as the Apache Portable Runtime, which is the portable C library that forms the basis of the Apache Web server and Subversion (as well as some other products). The library has never been documented, and so despite its many valuable features, its use is confined primarily to the developers who wrote portions of it or who have an employer willing to pay them to learn it.

The second—and important—reason for the commercialization of open-source products is that project managers far prefer to have developers over whom they can exert control. This enables them to plan specific feature sets and release dates, rather than to hope that someone will contribute the needed code.

This commercialization has distinct benefits: You get OpenOffice as opposed to the incomplete AbiWord. You get enterprise-grade threading support in the Linux kernel, rather than the various mixed approaches used previously…and so on.

However, it has limitations as well. Notably it suggests the important question of how—perhaps even whether—any large, grass-roots, open-source project could thrive today.

Research tends toward discouraging results on this point. In the past few years, most new code in the open-source community has come from existing codebases that commercial vendors have opened up (such as IBM Cloudscape, Sun Solaris and the like).

A few projects, such as Firefox and Spring, are flourishing by themselves, but they are small in scale (certainly smaller than the L-A-M portions of the LAMP stack) and few in number.

This commercialization of open source explains why a vendor’s decision to release source code for a product in order to reverse its commercial slide generally does not have that effect. All it does is open-source the project.

The commercialization also means, I believe, that the open-source aspect will no longer be much of an advantage for a product. Of greater importance will be the price (and here Stallman’s distinctions between free code and open-source come into play), the feature set and the support from the vendor or, in the rarer cases, from the community. And when you think about it, aren’t these the same criteria for closed-source, commercial software?

Andrew Binstock is the principal analyst at Pacific Data Works.

Share this link: http://www.sdtimes.com/link/29050

Git changes how developers manage source code versions At this year's Git Together, developers praise the platform's convenience and its ability to make source control easier. Changes for the future will focus on making it more Web 2.0-compatible by offering more on-demand repositories on GitHub.

News Briefs: June 15, 2008 Ingres creates a group of programs to help break in developers to the open-source concept, Xenocode launches its flagship virtualization program and Indigo Rose updates its MSI Factory program.

OSBC focus turns to best practices for open-source adoption Businesses are no longer asking why they should adopt, but how do they use open source legally and securely?

Add comment

Name*
Email*
Country

Comment
Preview

News on Monday
more>>
SharePoint Tech Report
more>>

Download Current Issue
ISSUE 3/15/2010 PDF

Need Back Issues?
DOWNLOAD HERE

Receive the print Edition?

Google Code turns 5
Google Code Turns 5, and adds a Paxos Algorithm to make the system more stable and reliable.
03/17/2010 11:16 AM EST

Test your Visual Studio 2010 know-how
Microsoft is offering free beta certification exams for Visual Studio 2010.
03/17/2010 11:08 AM EST

Microsoft lifts the hood on IE9
Microsoft is previewing IE9.
03/16/2010 01:10 PM EST