I agree with Michael Wheeler’s recent remark.  In an Open Solutions Alliance newsletter, the CIO and Finance Officer of Redmayne-Bentley said “Our view is that there must be fantastic opportunities for long-term investors among various software companies with at least a foothold in the open source arena.”

We need to look at this in context – what does the current economic melt-down mean to us?  Open source is clearly a better position if you compare it directly with the traditional space, because it’s more flexible and more predictable in terms of cost.

Today’s climate requires new ideas and a strategic look at how to optimize basic parts of our business across the board.  Three rules will cover it:

• Optimize Costs
  Costs are tight right now.  Expense costs are lower in our arena and clearly we need to generate a very simple, fast ROI.  The open source model (try before you buy) is a big advantage in validating solutions.
• Optimize Efficiency
  This means adding value to the core business for partners.  We need to concentrate on producing solutions that customers really need and steer clear of the speculative.
• Optimize Productivity
  We must generate more value in terms of productivity.  This doesn’t mean doing the same things as the competition.  We can’t just provide the same tools for the same performance; we need to provide better performance.

If we can address all three points, we’re not just resisting the financial crisis – we won’t have a financial crisis.

This has been our core strategy since we started the company and we plan to continue along those lines.

Bertrand

“In 2002, David A. Wheeler published a well-regarded study that examined the Software Lines of Code (SLOC) present in a typical Linux distribution (Red Hat Linux 7.1). His findings? At that time it would cost over $1.2 billion to develop a Linux distribution by conventional proprietary means in the U.S.” (introduction to a 2002 study which estimates the total development cost of a Linux Distribution).

The Linux Foundation – a non-profit consortium dedicated to fostering the growth of Linux – recently published an update to this study written by Amanda McPherson, Brian Proffitt, and Ron Hale-Evans with the help of IDC. Using the same tools and methods to update these findings, the authors estimate that with today’s software development costs “it would take approximately $10.8 billion to build the Fedora 9 distribution in today’s dollars. It would take $1.4 billion to develop the Linux kernel alone.”

According to the press release published by the Linux Foundation, the Fedora 9 distribution contains 204.5 million lines of code in 5,547 application packages. The development effort estimate comes close to 60,000 Person-Years. Using the same calculation, the Linux kernel included in Fedora 9 counts 6.8 million lines of code and its value is estimated at $1.4 billion. The development effort estimate for the kernel alone exceeds 7,500 Person-Years.

From a client perspective, these investments are a bargain. As Matt Asay wrote, that’s “$10.8 billion that we don’t have to spend to get an exceptionally robust operating system. $10.8 billion that we depend upon every day when using Google, Amazon, and a dizzying array of websites, as well as many of the applications we use within our own companies.” Indeed, more and more enterprises are realizing that open source solutions exist to make them happier. A new IDC study states that the growth in adoption of stand-alone open-source software is accelerating and that the total market will be worth US$5.8 billion in 2011. The market reached $1.8 billion in 2006, and is predicted to grow 26 percent annually for the next four years.

There are several sides to this debate, however. On his blog this week, Dana Blankenhorn, ZDNet business journalist, asks “Are vendors vital to open source?”

Blankenhorn writes that “the most vital open source projects are run by non-profits – Linux, Firefox, Eclipse” but he overlooks Joe Brockmeier’s (Community Manager for openSUSE, a community Linux distribution sponsored by Novell) opinion: “To be clear, Linux isn’t run by a non-profit. Linux is an open project that accepts contributions from many vendors, including Red Hat, Novell, IBM, HP, and dozens of others. The Linux Foundation, which pays Linus’ salary, is a non-profit, but it isn’t really accurate to say that it ‘runs’ Linux. Firefox is run by a non-profit which works in conjunction with a corporation to help fund the operations of the project – and it receives much of that money from Google.”

We at Talend are, of course, convinced that commercial open source is a win-win model, far more efficient than proprietary solutions that lock clients into closed software, with the only goal to make money.

In her Linux Foundation blog, Amanda McPherson quotes Matt Asay who says, “open source eliminates the vendor lock-in that created the enormous margins of the proprietary software world. It has eliminated the terrible inefficiencies created by companies competing and trying to differentiate on platform components that should be commodities. Now they collaborate, as they do with Linux. So just as the margin has migrated, I would say so has the development burden. Just as we use pooled money in the form of taxes to create roads or airports, you want to share the development costs of your computing infrastructure, in this case the operating system.”

Talend invests in its own R&D to develop and enhance its solutions. Talend’s solutions are not developed by community contributors. Of course, we rely on our community for specifications, tests, and analysis of market demand and that is no small thing. This method – listening and taking into account client needs – has never been adopted as a real strategy by the proprietary vendors. However, that is one of the real values open source vendors bring to the market. For example, 30% of the connectors natively integrated in our products come from contributors. A business will finance the development of a connector in response to its own needs, and then transfer it to the community via the vendor’s new version. But the vast majority of our core code is produced by engineers who are on Talend’s payroll.

Evangelization, training, deployment services, and high-level integration services are others things that can really help customers.

In the real world, vendors like Talend dreamed of finding a model that would create a balance between client satisfaction and profit, R&D and sales. And I think we’ve been pretty successful so far: 500,000 copies of Talend Open Studio have been downloaded, and more and more customers are subscribing to Talend Integration Suite.

I will conclude by asking the question I’ve been mulling over since I read Blankerhorn’s post: does this kind of debate really benefit users? It is similar to the discussion surrounding the free nature of open source software which surfaced some years ago. User organizations were never fooled by this misleading concept. I rarely – if ever – met customers who sincerely thought that open source was a no-cost solution.

Bertrand