i‎ > ‎

s

Enterprise Computing NYT NOW

Open Source and the Challenge of Making Money

By Quentin Hardy
July 23, 2014 2:28 pm July 23, 2014 2:28 pm PhotoA trend of businesses that start with open source models and then switch to more proprietary offerings raises the question of whether it makes sense to build free stuff at all.A trend of businesses that start with open source models and then switch to more proprietary offerings raises the question of whether it makes sense to build free stuff at all.Credit Screen shot, via Big Switch Networks

Remember how the open source software movement was supposed to be like Woodstock, with everybody sharing and everything free? An entire economy where you gave a little to get a lot, in a place of love and software?

At the risk of bringing down your summer, it’s time to admit that this idea didn’t work out.

Take Big Switch Networks, a company that hoped to be for computer networking what Linux operating system software is for computer servers. A few years ago, Big Switch proposed building networking controller software that was crowd-created and free, which could demolish proprietary networking boxes. It would also offer a commercial version, with a few tweaks, that could be the basis of a great, profitable empire.

On Tuesday, Big Switch announced what amounts to a more sophisticated and older-style business model. The company announced a proprietary software product that lets relatively unsophisticated engineers allocate networking resources across a data center. The open source controller is still a part of Big Switch’s proposition, but in a much smaller role.

“We were going to be open source and proprietary, allowing people to build their own applications, and we’d upsell them with what we made,” said Doug Murray, the chief executive of Big Switch.

Over time, however, “we found out that customers wanted products,” he said. “They didn’t want to get involved in writing things themselves.”

Many other supposedly open source companies now work the same way: Get a software project going that outsiders can work on, and with your own team sell proprietary stuff that makes the open source project actually useful.

“We talk about doing good and doing well,” said Dries Buytaert, the creator of Drupal, an open source content management system for websites, and chief technology officer of Acquia. The company gives away the Drupal code and sells things, like site search or ways to run the code in Amazon Web Services, that people pay for.

“All the open source companies have learned that this is what you have to do,” he said.

Which raises the question of whether it makes sense to build free stuff at all.

“Is open source a crappy business? Very true,” said Marten Mickos, the former chief executive of MySQL, a company that created an open source database management system and was sold to Sun Microsystems for $1 billion in 2008. “We didn’t realize this when we started out, but eventually we realized that you have to sell something that is just your own.”

In 2010 Sun was purchased by Oracle, a maker of proprietary databases, for $7.4 billion, in part because Sun owned free database software that posed a threat to Oracle.

“Open is not a business model, it is a production model,” said Mr. Mickos, who now runs Eucalyptus Systems, an open source software company for cloud computing that includes lots of proprietary software. The company has stumbled in the past two years, he said, but is now gaining traction because the quality is slowly improving.

Having many people in many places working on something, he said, is “a way to experiment with a lot of possibilities, make a lot of mistakes quickly, figure out what works faster.”

The open source method may be effective if enough people play along, but it does not make money in itself. Moreover, by definition it implies that open source projects have many more mistakes, bad code and failed efforts on their way to succeeding, compared with conventional projects.

Both Mr. Mickos and Mr. Buytaert compared open source to evolution, in which experimentation in the form of a million failed freaks finds a successful new adaptation for a species. “Would you call nature cruel and wasteful?” Mr. Mickos said. “O.K., then open source is wasteful.”

“Open source is Darwinian,” Mr. Buytaert said. “Eventually the best idea wins, but it is much more wasteful. A regular company couldn’t have experimented with creating 10 versions of an online photo album, then picked the best one.”

Not only that, but the profit margins in open source are lower than in proprietary software. There is, after all, a big chunk of the product that is free, and the point is to put the incumbent players in a corner with lower prices. Big Switch is hoping to undercut the likes of Cisco Systems by selling its software on commodity boxes from Dell and Quanta.

“Now the total switching business is $18 billion,” said Mike Volpi, a former Cisco executive who is now with Index Ventures, a fund backing Big Switch. “When all is said and done, it will be $10 billion to $15 billion, with a lot more activity, and a lot lower margins.”

Like nature, for all the nice feelings, the tech business remains red in tooth and claw.

#auto

Subpages (8): 1 2 a e g i k q
Comments