The state of open source: Issues and opportunities

Roundtable: 11 leaders from the open source and vendor communities discuss the current open source climate and outline the challenges and opportunities ahead

Question No. 1: What do you see as the most pressing challenges and opportunities for open source given the current tech climate?


Bruce Perens

Creator of the Open Source Definition
Co-founder of the Open Source Initiative

Perens: The biggest problem facing open source, and software in general, is software patenting. Copyright is sufficient to protect the proprietary software folks, but they are threatened by software patents, too, as you can see from all of the court cases reported about them. As the situation exists today, it's very easy to get a software patent for something that isn't really an invention at all. What happens to the companies that have to defend themselves from an unjustly granted software patent? It can easily cost them $7 million in legal fees to win -- that figure is from the American Intellectual Property Law Association's Economic Survey. That's bigger than the entire funding of most startups and many ongoing concerns. So winning isn't economically viable for them, and their only real choice is to settle for whatever the plaintiff wants, whether there is any justice to that or not.

It is not possible today for a nontrivial program to be noninfringing on software patents granted in the U.S. There are just too many granted patents, on too many general principles that everyone uses. If enforcement of all of those patents was complete, there would not be a software industry in the U.S. at all.

The problem is even worse for open source developers, who aren't necessarily getting any income from their software and thus have no funds to pay lawyers and patent royalties. Tech experts and economists both understand there's a need for reform, but we haven't been able to make it happen politically yet. Part of the problem is that pharmaceutical companies are on the other side of the argument, not because they like software patents but because the law doesn't distinguish software from drugs or any other kind of technology. Of course, the pharmaceutical companies have lots of money to fight for what they need. We need to decouple software patenting from pharmaceutical patenting to win this fight.

This has started to be a real problem for open source developers. The big guys aren't the only ones being sued. There's a developer who makes, of all things, open source model railroad software (JMRI: Java Model Railroad Interface) who is a defendant in a patent case.

Obviously, open source is a new and very effective means of doing innovation. As a nation or world, we can't afford to throw out this new and powerful means of innovation because it conflicts with a bad law. We've got to fix the law.


Eric S. Raymond
Programmer, author, and
open source software advocate

Raymond: Opportunity? The utter failure of Vista to gain traction even among Microsoft's most loyal users, and Apple's decision to morph into a cellphone and consumer-electronics company that has taken "Computers" out of its name. These two developments have left a huge Linux-shaped hole in the center of the OS market.

UMPCs [ultramobile PCs] like the Asus Eee PC, running Linux, are flooding into that hole from below -- consumers are actually buying them, by the truckload. We've already seen one VP at Sony publicly worrying about what he calls a "race to the bottom" because these sub-$200 machines could knock the crap out of their bread-and-butter market for expensive home PCs in the very near term. And, of course, at the high end, Linux continues to clobber Windows in comparative numbers of Internet-facing servers.

Our challenge, basically, is to gain enough market share to break Microsoft's monopoly before it can recover -- if it can. If UMPC sales keep showing geometric growth, we'll take the consumer market by storm, and Linux might very well go over 50 percent share this Christmas. Don't laugh -- that Sony VP wouldn't be fretting in public if this weren't a real possibility.

I predicted seven years ago that what would eventually break Microsoft's monopoly is PC OEMs trying to claw back margin as hardware costs drop so low that a Windows license is the biggest single item in their cost to produce. UMPCs have reached that level, and I think the rest of the PC market is going to follow them down.


Dave Rosenberg
CEO and co-founder

Rosenberg: In the last five years we’ve seen open source go from geek tools to mainstream applications. The challenge for open source, as with any emerging technology force, is to continue to be innovative while delivering high-quality products. 

I see the current tech climate as ripe with opportunity for open source. With the murky U.S. economy, companies are much less interested in spending huge amounts of their budgets on up-front license fees to proprietary vendors. IT shops are more interested than ever in controlling their fate -- and controlling their destiny.

The shaky economy means that traditional software companies with expensive sales models will become further relegated to the dustbin, as open source distribution puts software closer to customers


Javier Soltero

Soltero: The biggest challenge for open source companies remains finding a scalable way of generating revenue while maintaining the openness and community focus that makes their business possible. Today's economic climate will put even more pressure on business models that are not sustainable because not enough value is being delivered to customers to motivate them to pay. The opportunity, of course, lies with the fact that open source companies have a much more cost-effective way to engage with their customers and prospects. This gives companies assurance that, in the face of slower growth, they are investing time engaging with prospects who are already using their products and are hopefully members of their communities. The economic model of traditional enterprise software companies, which do more "top-down" selling, is much more capital-intensive and inefficient than the "sell to users" model that is the foundation of commercial open source.


Matt Asay
Vice president of business development

Asay: On the one hand, in a recession we typically see a "flight to value." Alfresco includes veterans from Oracle, Documentum, Business Objects, and others. We've weathered these downturns before. Oracle, for example, has emerged stronger from each downturn because, at the time, it offered significant value for the money. The tables have turned now, and I suspect we'll see companies like MySQL cutting into the proprietary incumbents. Well-run open source projects -- community-sponsored and corporate-sponsored -- deliver superior technology at a lower cost. Hence, open source should actually gain ground on proprietary software in a recession.

That said, as IT budgets dry up, there will be much less inclination to bet on new projects. At least, not those that require significant capital investment. What we may end up seeing is a lot of dabbling in open source during the recession, preparing to ramp up payments into open source once the economy resumes growth.


Sam Ramji
Senior director of platform technology

Ramji: In terms of challenges, I think you have to start with the fact that most software today -- whether measured by usage or by lines of code -- is not open source, and is sold or written by commercial organizations using a proprietary model. For established companies, the shift to open source is not just about understanding what the strategy should be -- it’s about “programmatizing” open source in an organization when the primary revenue model is around traditional.

It’s also important to distinguish between commercial and community software. The maturity models for assessing community software are not well-established yet, which results in some confusion about what projects are ready to adopt at what level of importance or mission-criticality. This is an unsolved issue and represents an opportunity for the next wave of software companies or consulting organizations.

Finally, the word “open source” has become used to describe development models, licensing models, community models, distribution models, sales and marketing models, philosophical and ethical models, and is now being applied to politics. Clearly there are powerful core concepts of transparency and sharing at the heart of this. It’s starting to blur the original ideas articulated by Eric Raymond, Danese Cooper, et al, which are about the source code itself and the developers who share it. The risk is that the term itself loses meaning over time, which is unfortunate as it’s a powerful idea.


Andy Astor

Astor: The chief challenge open source faces is the need to tease apart the development and community model from the distribution and business model. Open source is used as a label for two very different things. First, as a way to develop software, and second, as a way to distribute and make money on software. Because the label is used interchangeably, the two get confused fairly often. As open source development and open source capitalists, like me, become more prevalent, the great challenge will be to recognize that there are two very different currents in the open source movement, each with its own particular requirements to succeed.


Zack Urlocker
Vice president of products

Urlocker: As with most systems software, it needs to continue to get easier to use. MySQL, Apache, PHP, Linux, JBoss, etc., are popular because they are powerful and easy to use. They are far less complex than some of the old proprietary software that was developed in the 1990s. But there's still a ways to go to ensure that all the software works well together with a single, simplified installation.

Open source has made the transition into IT and is being used for very complex systems development. But I think the infrastructure software is still more popular than open source applications. Still, we're seeing the start of that with companies like SugarCRM, JasperSoft, Pentaho, and others.

As we head into a recession with more IT budget crunch, I think we'll see the next wave of open source adoption. If it's good enough for telcos, banks, and the largest Web sites, maybe it's good enough for broader adoption.


Robert Sutor
Vice president of open source and standards

Sutor: I think the open source community has to focus on the final issues needed to really get broad adoption of Linux on the desktop. This means dealing with device drivers, breadth of applications, look and feel, and usability. To me, this means trying to make the Linux desktop as capable and as friendly as the Mac, rather than trying to emulate Windows, for example. In the same way, I want to see more open source applications be the recognized stars of their product categories and be the first to showcase innovations.


Mark Spencer
Founder and CTO

Spencer: I see the challenge to be finding and building the right combination of technologies to address the demand for the integration of different software applications and systems such as accounting, reporting, ERP, CRM, etc. Open source is uniquely positioned to allow easy adaptation to address these needs.

Open source programs manager Google

Chris DiBona

DiBona: I think that open source's biggest challenge is to ignore the current tech climate and continue to create software for its end-users. The tech climate, which I take to mean the mix of commercial and market influences on technology direction, is often shortsighted and selfish. Open source software developers should be influenced only by its developer base first, and its user base second.

[ Roundtable home | Topic No. 2: Evolving trends ]

Copyright © 2008 IDG Communications, Inc.

How to choose a low-code development platform