Proprietary vs open source

Linux and other open source software do not represent a breakthrough technology. Linux is a modernised derivative of Unix, born in the Bell Telephone Laboratories in 1969. Various contributors to the Unix family tree have developed important features, but it's not easy to declare any of them a Nobel Prize-worthy achievement. At least the creators of Unix, Dennis Ritchie and Ken Thompson, were awarded the A.M. Turing Award in 1983 for their original work.

Linux is a breakthrough of another kind, with major ramifications for the software industry beyond the much publicised threat to Microsoft. It's more akin to a historic moment like the fall of the Berlin Wall than a great scientific achievement. Linux--and the open source movement--is changing not just the economics of software, but also the way people think about software.

Today, the software world is divided into proprietary (such as Microsoft Windows) and open source philosophies (for example, the General Public License.

The proprietary camp believes the source code, whether original or derivative, should be kept behind a wall, preventing competitors, or even customers, from fully exploiting, hacking or stealing it. Companies trafficking in proprietary software believe their model for business and innovation has proven to deliver value to customers as well as to their shareholders in terms of recouping investment in research and development. Based on Microsoft's numbers, software margin is one of the top 10 wonders of the business world.

In a June memo to employees, Microsoft CEO Steve Ballmer said that IBM's endorsement of Linux added "credibility and an illusion of support and accountability, although the reality is there is no 'center of gravity,' or central body, investing in the health and growth of noncommercial software or innovating in critical areas like engineering, manageability, compatibility and security."

"There is always enthusiasm in our business for new concepts," he wrote. "So-called 'free software' is the latest new thing. We will rise to this challenge, and we will compete in a fair and responsible manner that puts our customers first. We will show that our approach offers better value, better security and better opportunity."

"While the noncommercial model may lead to many flavours of software, getting broad, consistent innovation requires coordination across many technology components. In the event of needed enhancements or fixes, the Linux development community, no matter how well-intentioned, simply cannot advance Linux the way we can--and must--innovate in Windows," Ballmer wrote.

I think Ballmer underestimates the strength and dedication of the open source movement, which includes well-heeled allies like IBM. The open source philosophy espouses that source code should be transparent--with no walls. The community provides the research and development, aided by groups like the Open Source Development Lab (OSDL), which serves as a "center of gravity for Linux" (in response to Ballmer's characterisation of Linux lacking a central body).

Businesses are built on top of the open source code, rather than by keeping the code behind a wall and charging for access to unique intellectual property and file formats. Companies must succeed on packaging, support and integration services. That's why companies with major services businesses and the hardware to run Linux, like IBM, HP and Unisys, are profiting more from open source than those--like Sun--that are trying to hang onto the past.

The operating system is not a palace of profit, unless you are Microsoft.

The battle is often painted as a confrontation of Microsoft versus Linux, David and Goliath, socialism versus capitalism. Bill Gates and team are lined up on one side with a US$50 billion war chest, an attitude of relentless pursuit, and customers who are starting to question their allegiance. Linus Torvalds and the ragtag army of open source developers and companies--including Red Hat, SuSE and MySQL, flanked by IBM on the other stalwarts--stand on the opposing side.

From a corporate IT perspective, however, both sides have legitimate claims and products to offer. It's not an either-or situation. It's about the price and service for goods rendered.

Today, the world is dominated by more proprietary software, but it's clear the open source movement will gain increasing "share of code" in the coming years. The issue is more about choice, as Red Hat CEO Matthew Szulik likes to say. CIOs are beginning to view operating systems, middleware or even applications as a currency that is changing in perceived value, and they want to take advantage of lower costs, increased standardisation, and an improved negotiation position.

At the bottom of the software stack, Linux has been gaining traction as an enterprise operating system. The OSDL is sponsoring Carrier Grade Linux and Data Center Linux working groups to accelerate development in those industry verticals. Going beyond the current sweet spot for Linux-- file, print, Web and appliance servers--is a significant challenge. Meta Group projects that more than 40 percent of new Intel-based servers shipped by 2006/07 will be Linux-based. The economics of the open source ecosystem will make Linux a major factor in the data center. However, Meta Group analyst Corey Ferengul said that while Linux is a less expensive operating system, it still requires the same level of support and management tools as any other high availability platform.

Ferengul predicts that Linux distributors will focus more on delivering manageability over the next several years, with a mixture of tools from new and established vendors and shareware filling the gap.

The Linux distributors, as well as Novell, are also trying to tackle the integration problem, which accounts for a huge portion of IT spending. Red Hat and SuSE, for example, are expanding their stacks by including more sets of middleware components in some of their distributions. However, this is a far steeper curve than gaining entry into data centers with a lower-cost, high-performance server operating system. On the other hand, for those distributors to build a lasting business proposition, they need to move up the stack.

There are open source alternatives like JBoss to fee-based J2EE-based application servers such as IBM's WebSphere, BEA's WebLogic, and Oracle's 9iAS. Sun doesn't appear inclined to enforce the intellectual property rights to Java that would castrate open source efforts. Ximian, recently acquired by Microsoft rival Novell, is developing software that will allow companies to run Microsoft .Net applications on Linux.

I spoke with Stuart Cohen, CEO of OSDL, who had a practical view of the proprietary and open source software contest. "From a software stack standpoint, customers will decide what make the most sense for them. I don't expect to see a full Linux and open source stack that will carry the day. Some proprietary software is really good and will continue to be the software of choice. For example, strong IBM, CA, and Oracle products will continue to have traction, as long as they do great job of responding to customer requirements. But, there will be price and service pressure put on those companies by users."

I asked Janet Perna, general manager for data management solutions in IBM's software group, whether she thinks an open source database will be competitive with IBM's DB2. "I obviously think about this quite a bit. At the higher end, when you are dealing high availability, large data volume, and different types of information, I don't think it will happen. There are a lot more people who know about and can work on operating systems than on databases. There is a skill in database technology that is not as prevalent in the world today as operating system expertise." That's sounds like wishful thinking. Database technology may be more arcane, such as in query optimisation, but the way Linux is scaling up proves that the open source movement can succeed in developing high-end system software.

Diane Greene, CEO of VMware, a leading provider of virtual machine software on Intel architecture systems, told me that as technology becomes commoditised over time, it could be contributed to the open source community. Typically, a company will have developed new software that provides differentiation and competitive advantage, she said.

Clearly, open source isn't going to eclipse proprietary software anytime soon. Granted, the people I interviewed either have a business based on proprietary technology or, in the case of OSDL's Cohen, partners like IBM, HP, CA, Fujitsu and Cisco who straddle and profit from both camps. But the motive to make money by developing unique and proprietary code is a deeply rooted part of our culture. It takes just one programmer to create an innovation that can be turned into monetary value, and it's up to that programmer to decide whether to profit directly by selling a product or donate it the community and possibly provide a fee-based service, as with Linux distributions. Similarly, not many corporations are willing to share their specially brewed code created in-house with their competitors.

The enterprise will be a hybrid world that continues to integrate both proprietary and open source code for a long time to come. With the emergence standards like Web services and software platforms like Java (which has similar characteristics to open source), in addition to open source, developers and IT organisations have better building blocks and an environment that fosters lower prices and continued innovation. In the end, you have more options and you can choose whatever makes the most sense to help your business.