Scott McNealy, Obama, and Open Source


Image by opensource.com
submit to reddit
 
(5 votes)

Gene Quinn's recent post titled "What Happened to the Obama Open Source Initiative?" criticizes, in turns, open source software, Scott McNealy, the Obama administration, and "business newbies" who want to use the open source software model.

Early in the Administration, President Obama asked Scott McNealy, a co-founder of Sun Microsystems, to prepare a report on how the federal government could employ open source software, but Quinn notes that "as yet, some 26 months later there has been no mention of the report or across the board government adoption of open source software." The article then draws this strange conclusion:, "Perhaps the trouble associated with coming out with a report or even a government wide coherent approach is that open source software is not really free really."

Yeah, I didn’t understand, either. To summarize:

  • The President shouldn’t have commissioned a report on open source software.

  • Scott McNealy shouldn’t have been asked.

  • The follow-through on the report has been poor.

  • Therefore, open source isn’t really free.

Disappointingly, Quinn doesn't stitch these points together, robbing me of the opportunity to refute him. Instead, he uses the McNealy story as a christmas tree from which he may hang his favorite open source bugbears. I'm forced to content myself with refuting these instead.

He first makes the argument that open source is not a sustainable business model, and drags in a number of news articles to support this idea. Those four articles, if read in their entirety, present a very different picture. For example:

"…the business model for open source has been elusive, that there is doubt whether the business model can stand the test of time and that really only one company – Red Hat – is successfully making money." – New York Times

This article actually discusses how a number of companies, including MySQL, SpringSource and XenSource were acquired rather than going for an IPO. These companies are clearly desirable, and have created a great deal of value. Acquisition is a legitimate exit strategy, and I'm not sure why an open source company has to IPO to be legitimate.

The NYT also describes how open source has changed as it’s matured, being embraced by many companies as part of a larger strategy. Nowhere does it indict open source software as a business-killer. Instead, the article describes the billions of dollars in value that open source projects are creating for a broad array of companies.

"…the way to make a small fortune with open source software is to start with a large fortune, and then went on to point out that, ‘The open source revolution began at least two decades ago, but businesses and programmers are still struggling to understand the best way to share wonderful code and pay the mortgage.’" – InfoWorld

This InfoWorld piece illustrates how difficult it is to build large pure-play open source companies. This is true, but an open source project can still be useful to many, many companies and create a tremendous amount of value without requiring a pure-play open source business behind it.

In fact, the "open core" businesses wouldn’t exist without open source. The argument here seems to mutate from the assertion that open source is unsustainable into an even less defensible assertion that only pure-play open source businesses are evidence of open source’s business value. Quinn’s article then moves on to patents and open source:

"When the law offers a mechanism to create an advantage it is hardly unfair to exploit it. What am I talking about? Patents. If you have a software innovation you can obtain a patent on it and pursue a proprietary model...In the early stages of a business life cycle it makes all the sense in the world to copy and take from others. Open source must seem to business newbies akin to being a kid in a candy shop. But then after you create yourself it doesn’t seem quite so wonderful. Depending upon the open source regime you have copied from you may have no ability to prevent others from taking what you contribute that is new, which doesn’t sound so great after you have spent time and money creating that which is new. How will you recoup your investments? Exactly. You won’t, but you will be able to sell your time as a service."

This betrays not only a misunderstanding of why open source is written, but how innovation works. Sciences continue to innovate at a breakneck pace despite the widespread practice of "sharing", "peer review" and "collaboration". It’s been said many times over, but apparently bears repeating: open source isn’t about free [as in beer] software. The fact that the software has no licensing cost is a means to an end, which is the creation of an efficient marketplace of ideas and their rapid commoditization. That’s immensely valuable to a business, as we learned in the NYT piece.

But wait, weren’t we talking about Obama and McNealy? Ah, yes:

"..when you set out an initiative on the second day in office and you still don’t have anything to show for it exactly 26 months later I think it is legitimate to ask why. So why hasn’t there been more progress? Perhaps it isn’t as easy as it would seem to adopt something that the government thought would largely be free."

I’m sorely disappointed. I was hoping for a substantial argument against all the open source projects in the Federal government that started in the last 26 months. Instead, the article uses the notional McNealy report as an excuse to trot out these road-weary straw men.

Open source in the federal government is actually just fine. Companies like Red Hat (where I work), Acquia, JasperSoft, and EnterpriseDB are doing great in the government market. Consultancies like Accenture and Booz Allen Hamilton make use of open source under government contracts all the time. I'm meeting new small businesses every day that support open source in Federal agencies. Even Lockheed Martin, General Dynamics, and the other large government contractors are using more open source than ever.

In the last 26 months, we also saw the Technology Neutrality memo come from OMB. We saw the White House release Drupal code, the OMB release the IT Dashboard platform, and the Veteran’s Administration start to open source their VistA system. Lockheed Martin released EurekaStreams. We continue to see open source projects from NASA, who just held their first Open Source Conference last week. Even the State Department held a Tech@State conference to discuss how it could open source both internally and as part of its international development efforts.

These companies and agencies are not blind open source ideologues. They just find that open source is good for the mission and good for their business. This is, in fact, part of a larger trend in the marketplace. I think that’s a wonderful thing, and I don’t need a report from Scott McNealy to tell me so.

""
Creative Commons License

1 Comment

Brian Masinick's picture

From the very first moment that someone had the boldness to use free software - even insist on freely available software, there have been mockers and scorners. Richard Stallman wrote his initial GNU Manifesto way back in 1983 and started the Free Software Foundation in 1984 and created some real difference makers in GNU software, starting with GNU Emacs, but then moving, one by one, to the majority of UNIX utilities, replacing them with superior GNU equivalents, until nearly the entire system was complete. The one big fumble, where they never really got it right was the kernel. The GNU project bet wrong on using the Mach-based kernel and the development floundered, whereas a college student named Linus Torvalds, now a well known man, developed an interesting kernel that could run on a PC, could load the GNU C compiler (now the GNU Compiler Collection), and an entire new ecosystem was formed.

Yes, it was an academic exercise, but only at first. It did not take long to realize that it was not only usable, it was profitable to base a hardware infrastructure on a free software architecture. The just emerging World Wide Web of the nineties brought Linux a reason for being, and became a great place to "check it out". Soon both individuals and companies found out that it not only worked, it often worked better (and clearly far less expensively) than other options.

As the software stack began to grow, Web sites, routers, and other network based applications began to experiment with Linux. They all worked. As database technologies began to move to Linux, now it looked like at some time in the near future, an entire back end could be built on an "open source" platform using Linux and other related technologies.

Red Hat was one of the first companies to realize this, and they wisely built their entire business model around the idea of creating an entire application stack that uses "open source" technology. They are certainly one of the most successful companies with an "open source" label, but what I think many people miss is that Linux and other open source technologies are widely used in many companies for a variety of purposes. You need not be "an open source company" to benefit from what open source has to offer - nimble flexibility to create a complete solution that meets your needs, the ability to change one or more components in that infrastructure to meet evolving requirements, cost savings in license costs that can be invested in software development, infrastructure, and strategy, and a whole lot more.