Linux proves that the decentralized model works, and works better. But having the best technology isn't all it takes...Linux needs advocacy in order to win the adoption war.
It occurs to me that there are two basic sides to the debate between open source and closed source software. There are those who believe that any and all software should be open source and that users of the software should be free to enhance and alter that software as they see fit. By the same token, there are those who will never understand the open source movement and will always discount or dismiss development done under the decentralized model frequently used in open source. I believe Linux alone has proven that the decentralized model works by providing a stable and robust solution that's being used to solve real-world problems. However, Linux advocates need to realize that simply having the best product does not guarantee success.
We're being consumed by the fact that we like to solve computer problems, and we can't understand why other people just don't see the benefits as plainly as we do. History shows far too many superior technologies that are rendered obsolete because of a more easily accessible alternative. Anyone who has owned a Betamax VCR can attest to this. I'd hate to see the same happen to Linux and OSS.
The grass-roots nature of Linux and OSS development thrives on community participation. The same community participation is now needed to work with a community advocacy organization. A community advocacy organization such as the Open Source Initiative is necessary to provide a rallying point for activism, so that Linux and OSS can get a seat at the table in vendor conferences and bodies, and to attempt to provide a repository for ongoing software improvement. I realize that these are lofty and possibly unattainable goals given the nature of Linux and OSS. Without reaching agreement to even move toward an organization such as this, we may find ourselves in a niche market. If we, as a community, are not ready to discuss a roadmap for the future of Linux and OSS then it's only a matter of time until we become irrelevant.
A major problem with Linux and OSS advocacy is overcoming the self-evident nature of the benefits of the software. Too frequently we take important features for granted when discussing Linux and OSS. These include software restarts, installing additional software, and the open nature of configuration and log files.
When a major service such as a Web server is consuming memory or otherwise acting flaky, Linux users simply stop or kill the Web server and restart it. When I've encountered what seems like the same issue in Windows, I find that only a reboot of the entire computer will fix the issue. Exceptions to this rule exist. I can sometimes restart IIS in Windows using the Services Control Panel applet or a resource kit application, but when IIS gets augered, a reboot is the only solution.
When installing software such as a Web server or adding components to it, Linux users just restart the process. When I've done the same in Windows I find that a reboot is too frequently required. The only reboots required for Linux are usually kernel updates, which may happen a few times a year. Usually though, Linux users aren't required to update the kernel unless they're fixing a specific problem. This uptime is vital to delivering robust solutions for business.
Troubleshooting problems in Linux and with OSS is much easier than on Windows. For the most part, configuration files are plain text and there's at least some level of documentation for the configuration options available. There are no hidden registry values. I'll grant that exceptions exist on both sides of this issue, especially as I think back to learning m4 to try to translate a sendmail configuration. Many times there are also logfiles available for services in Linux. Linux includes the tools to parse these logfiles easily.
Those three items - service restarting, rebooting due to installation, and troubleshooting based on configuration file settings - are things that Linux users have had available to them for years and simply take for granted as an inherent advantage of Linux over Microsoft Windows. However, users of Microsoft Windows may not know that these are available or may not have the experience necessary to utilize them efficiently. By not having nearly enough advocacy, Linux and OSS are losing market share for no technological reason.
The bad news is that those three advantages to Linux have largely been either solved or made to work better in Microsoft Windows Server 2003. Windows Server 2003 requires less-frequent rebooting than any previous version, and Microsoft is also moving toward better documentation of configuration files as well as XML-based configuration such as the IIS configuration files. While it's great that these features are gaining acceptance among other vendors, it does nothing to promote the use of Linux and OSS. Microsoft doesn't say "Improved reliability thanks to Linus Torvalds."
Are those three items the only advantages that I see with Linux and OSS? Hardly, and in fact the biggest single benefit of Linux and OSS is the open code. There is no lock-in to a single vendor path tying your entire organization to the fate of another. The problem with this aspect is that most users of Linux and OSS will never realize any benefit from having open code, they just want software that works. If I can't get a Web developer to read a paragraph of documentation about the Apache Web server, what makes me think that I can get that same person to ever look at a line of C code?
Linux and OSS can continue to thrive and be the innovative and robust solutions that they are today without additional advocacy. I believe the open source development model is sustainable long term. However, I'd like to see Linux and OSS gain market share by leaps and bounds, simply because I know that some of the most difficult problems can be solved with solutions based on Linux and OSS. These same problems cannot be solved or cannot be solved on any type of budget using proprietary software and technologies.
It is up to those of us directly involved in the Linux and OSS community to face the advocacy problem head-on. First we must define the problems that we face in getting Linux and OSS implemented, both in business and for the home user. Once we can define the problems, we can attack them with real solutions. If you truly believe that Linux and OSS can deliver a better solution, then it's time to stand up and say so. We can win the technological battles but lose the adoption war. Remember the Betamax!
Subscribe to Linux Email News Alerts
Subscribe via RSS
