Computer Society of India (CSI) Annual Conference
Indian Institute of Technology (IIT), New Delhi, India
December 2003
Robert Adkins of Technetra presented on the three pillars of open source – Collaboration, Precedent and Practice. He talked about how the trail of Unix and commodization has helped Linux grow into the enterprise.
Being the richest guy around is certainly no substitute for respect in high places. Especially when many governments just don’t seem to listen. Japan, Korea, China, India, Germany, England, France, Sweden, Norway, Italy, Finland, Netherlands, Spain, Australia, South Africa, Brazil, and many others are all looking for ways to use Linux and Open Source Software (OSS) to break the stranglehold of proprietary products.
Wait just a moment, says Mr Bill.
Such moves, we are assured, are wrongheaded and based on a false economy. Why waste tax dollars and lose productivity by adopting clearly inferior, inefficient and unprofessional open source software. OSS simply doesn’t have the quality you can trust!
And, believe it or not, the more trustworthy alternative is even cheaper than “free” software.
How do we arrive at this counter-intuitive insight? Easy really.
Software is but a tiny fraction of the cost of large ICT (Information Communication Technology) projects, where non-software components such as personnel and infrastructure easily dominate. The professionalism and polish of commercial, proprietary software leads to greater efficiency and productivity in these non-software components and thus pays for itself many times over.
Thank you, Mr Bill.
However, based on experience, many governments are now saying commercial, proprietary software may not be better. Let’s look at Mr Bill’s claims of “cheaper”, “more proficient”, and “more trustworthy” in a little more detail.
The marketing message of commercial software usually glosses over the costs, especially to developing countries, inflicted by skyrocketing license fees and by installment payments using forced upgrade cycles. Like patented drugs, proprietary software is usually set to one high price across the world, regardless of the user’s ability to pay. The OSS threat has been shown to introduce a healthy dose of competition, which can crack the global monolithic price structure of proprietary software.
Furthermore, government ICT has often invested in Unix and various non-Microsoft technologies. OSS represents a natural migration path for upgrading many of these government shops because Linux and OSS fit nicely into the traditional Unix model. In addition, with the introduction of mainframe Linux, OSS can help modernize many legacy mainframe infrastructures. By reusing technical skills and infrastructure, “free” software can be dramatically cheaper for these players.
The design point of PC software has always been the GUI and therefore it tends to be efficient for a single user. By contrast, the scripting and server tools available in the OSS world work far more efficiently in multiple server environments. Studies show that typical OSS administrators can manage more servers than their counterparts using GUI-based “designer” software. However, aside from server proficiency, an increasing number of proponents feel that current OSS desktop tools more than meet the needs of most users because many of the features hyped in proprietary packages are never used. The relentless self-promotion of proprietary software may cause us to confuse “bloat” with “proficiency”.
Many governments now realize that a single brand of software, the software monoculture, is fragile and dangerously insecure. The virus and vulnerability record of proprietary software has not been good. In the end, we only have the proprietor’s assurances that code continually patched for security is now clean and secure. Unlike OSS, there is no independent verification and no peer review.
It’s easy to claim that the transparency of OSS is somehow inherently less secure but hard to demonstrate that secretiveness produces better results. In fact, you would expect just the opposite according to most security experts.
So, Mr Bill, the best way to earn the respect of today’s governments is to become a better team player and to support all the needs of your customers, including the price competition and multi-cultural flexibility introduced by OSS. Dismissing any government who disagrees with you as wrongheaded and engaged in false economies reflects an attitude that will never earn the respect you seek, and will continue to come back to haunt you in the form of anti-monopoly litigation and the erosion of an important customer base.
To whichever ear they can bend, the top technology vendors have been articulating their visions of the future of computing. The biggest ear benders - IBM and Microsoft - have been composing and refining their visions and trying to make them reality. IBM, a corporate champion of open source software and Linux, envisions the next generation of computing to be governed by “outsourcing” open source based technology services. Rival Microsoft sees empowerment of the individual using its own brand of closed source software as key to the future of computing.
IBM predicts that users of technology in industries ranging from auto makers to xylophone manufacturers will lose interest in implementing computing services in-house. Instead, they will buy computing services and support from technology vendors. Users can then use only what they need and pay for only what they use - on-demand. Further, IBM believes that open standards and open source software are the technology factories for shaping the new on-demand computing paradigm. And IBM sees Linux and its rapid spread across the enterprise as the main assembly line.
Not surprisingly, Microsoft doesn’t completely agree with IBM. As in all things Microsoft, the individual user is the design point where further empowerment of the individual is key to the next generation of computing. Outsourcing computing services minimizes the individual’s personal control over computing resources because the on-demand model, according to Microsoft, is just a throwback to an earlier time-sharing paradigm. Instead, individuals on personal computers will use standards-based tools with Microsoft enhancements for everything from simple tasks to intricate applications. The individual is the smallest indivisible unit in a “per box” sales strategy - so Microsoft’s vision works hand-in-glove with its marketing strategy to maintain domination in the PC world. To empower individuals everywhere, Microsoft would build powerful tools based on open standards, embracing yet enhancing them with proprietary value-added extensions. For Microsoft, open standards provide the floors and walls of the factory but proprietary, closed source software continues to be the assembly line.
On the surface, who can argue with either proposition: pay for only the computing you need and empower the user. The problem with IBM’s vision is that though many technophobic organizations would happily embrace on-demand services provided by a remote computing utility, many would not. Organizations who need to control their infrastructure for competitive advantage or for mission-critical reasons (e.g., national security) are not in the on-demand computing future as currently envisioned by IBM. This might be called “in-control” computing and requires the user empowerment envisioned by Microsoft.
However, the problem with Microsoft’s version of user empowerment is that users cannot be truly empowered by embrace and extend strategies. Just the opposite. The appearance of empowerment hides the trap door of cost and dependency.
Open source software comes to the rescue. Applications built for on-demand computing can be migrated to equivalent in-house resources as long as open source is used in both places. The reverse is true as well, where open source in-house applications can be moved to an ondemand environment. This universality of open source guarantees empowerment of the user because the users ultimately have access to the core of the technology and services they are buying and thereby retain the ultimate power in a free market: the choice to change suppliers.
The beauty of open source software is that it can do it all. It can cover the breadth of on-demand computing as well as “in-control” computing.
© 2000-2009 Technetra. All rights reserved. Contact | Terms of Use
