It has now been almost three years since I stepped away from working daily on open source opportunities for Microsoft. I've watched the team that took on Shared Source morph it into some extremely positive collaborative work that is exactly what OSS is all about. The paltry few projects that were in place in 2006 have blossomed into thousands of projects, and some of the core source release programs for flagship technologies are still operational today. All good.
Unfortunately, I can't say that I'm seeing the same understanding about collaborative development when I go out and meet with customers, governments, partners...whomever.
I haven't been blogging much due to a great deal of travel, and the huge backlog of other work catching up to me following the close of the Open XML process. One of my most interesting trips was down to South Africa to talk interop, document formats, and yes...open source software.
South Africa has taken a most unfortunate position of late - the government has sought to put a political mandate in place for the adoption of open source software. I am against all technology mandates, and this one is no different. Ultimately, it constrains decision-making away from technology, solution quality, ROI on existing investments, people issues...in short value-for-money - all in the name of a political position. Worse, it is pushing CIOs into decisions that they don't want to make - essentially taking working environments representing huge investments and moving to untested, more expensive solutions.
But, the most serious issue to me is that they are not looking at the real benefits that OSS can bring them. Politically, every conversation about the OSS mandate is really a Windows vs. Linux discussion. This is in no small part assisted by the local presence of Ubuntu. There is absolutely no comprehension that the Linux they will deploy on an enterprise scale will be completely locked down by commercial services agreements and version controls by the apps vendors (e.g. SAP). This is absolutely fine from a decision point of view for enterprise systems, but it is most certainly not any gain due to open source.
The real value of OSS to a government that is looking to:
- save money
- bring development skills in-country
- address local issues with home-grown, customized solutions
- gain operational as well as financial efficiencies across government agencies
- foster local services and ISV opportunities
- etc. etc.
is to apply OSS development and licensing methodologies at the app-dev and tools layers, rather than thinking of the core OS as an OSS opportunity for them. Deep dev of the core OS is not likely to happen in South Africa today on any large scale. Students at the university still grappling with coding skills are not going to dive into the inner-working of Linux. Any innovation on Linux that is broadly applicable will immediately be picked up by Red Hat or Novell and commercialized globally with little economic benefit coming back to SA.
Yet, if they looked instead at the existing investments in infrastructure, and thought about the use of OSS against the custom needs of their government activities. Or for intra-government projects. Or for building key bridges to enable eGovernment solutions. Or any of a number of projects - they could bring in academics, local ISVs, local services providers, engage existing government developers, etc.
All of that could happen regardless of whether the platform is Windows, Linux, AS/400, OS/X...whatever.
The developing world still views OSS as "free as in no money," and that is widely known to not be the case. Technology solutions are expensive no matter what the development and/or philosophical model are underlying them.
I heard this same point of view for 5 years all over Asia, parts of Europe, and Latin America. I saw governments try to incubate OSS businesses solely because "OSS" was in the title and mandate. Then, those businesses failed, and the mandated solutions turned out to be far more expensive than other commercial alternatives. Almost uniformly this came about through a misunderstand (in my humble opinion) of what OSS can do for organizations.
There is a reason that Harvard Business School found that more than 95% of all OSS venture funding went into fewer than 20 projects. Those heavily commercialized projects are just another way to deliver high-value, mass-consumed, supported technology. Finding the value of OSS beyond those projects for the average organization is all about applying collaborative development against real-world problem sets in small, efficient projects.
Ahhh...it feels good to blog again. 🙂