The state of open source: Missteps and lessons learned

Dave Rosenberg: CEO and co-founder Mulesource

There was no other way for open source to progress without trial and error related to both development and business. The biggest business misstep is probably the obsessive focus on licensing, which, while important, distracted people for way too long.

There are too many lessons learned to count, so I would say that open source projects and companies should have been much more aggressive at proving use cases in mission-critical environments. This has been going on for quite some time but has only started to get noticed in the last year or so.

Javier Soltero: CEO Hyperic

I'd argue the biggest lesson has been and will remain the "desktop issue." People jumped on Linux as a potential desktop replacement very early on, and the debate continues to this day. Meanwhile, the operating system and the associated infrastructure tools such as Apache, PHP, etc., continue to enjoy explosive growth and disruption on the server side. I know there are many who believe that Windows and even Mac OS X are bad because they're not open source. That might be true, but one cannot argue that the user experience for both of these operating systems blows away anything you'd get out of Linux (at least today). The lesson here is that open source can deliver more immediate, tangible benefits in certain areas than in others, and the market is smart enough to figure that out. We'll get a good Linux desktop one day, I just don't see it happening any time soon, and I don't know that that's the right priority given where desktop computing is headed.

Mark Spencer: Founder and CTO Digium

We have learned a lot of lessons. The project started without an organization, and nonprofit organizations helped and taught us how to interact in better ways with the community. Those were the early days. The next steps are to continue to grow and be relevant in a commercial organization to take open source to the next level with corporate focus and identity and penetrate mainstream businesses and markets.

Andy Astor: CEO EnterpriseDB

It's the same lesson that proprietary vendors have learned over the years, and that is that great software is not enough. I looked on SourceForge earlier today, and there are over 170,000 projects, and most of them are dead. Successful software, whether it's open soruce or proprietary, needs a combination of great development and features -- and distribution, marketing, salesmanship, and so forth. Linux would not be a mainstream operating system without Red Hat, and I'm sure that some of my good friends with disagree with me on that. I don't think we'd get too much disagreement that JBoss would be nowhere without the JBoss corporation, or MySQL without MySQL, and Sugar without Sugar. But I think it's true for independent organizations, too. In spite of being a fantastic database, PostgreSQL has had only modest success because it hasn't had a company behind it. We're working to make EnterpriseDB that company. It's more than a matter of building great software. You need marketing, service, salesmanship, support, and documentation -- all the things that open source projects don't necessarily want to do. And that's what capitalists do with software.