The LSB stamp.

Linux's future lies in its community's hands

By: Kevin Lynch

Posted: 10/8/04

Picture a future devoid of all happiness and freedom. The poor are scrounging for pennies and pieces of bread to survive on, while the rich are lounging in front of lavish fireplaces, smoking on expensive humidors, grinning. What self respecting businessman wouldn't be proud of himself for surviving the few years of turmoil during early and mid oughts? After all, they were the ones who survived those troubled times by sticking with the corporations and fighting off the open source epidemic. Now those poor Linux-loving hippies are starving to death, wishing they had sided with the big Scrooge, Bill Gates.

No, this is not the future yet, but it is one of the many possible futures that the Ghost of Linux Future has predicted. Without any clear-cut plan, Linux and the open source community will surely collapse; at the very least, stagnate and fade away like all ridiculously successful fads. However, as long as open source activists and those in the Linux political scene play their cards right, open source will be here for a long time and this future will never occur. The dilemma lies in which direction Linux should go and which goals should be set.

One of the biggest difficulties Linux and open source as a whole faces is standardization. One of the main purposes and selling points of Linux are its openness to endless possibilities and applications. Linux was raised on the principles of adaptability and configurability. How many other operating systems are flexible enough to record television programs on a TiVo, provide internet access through home routers, drive Spirit and Opportunity across the barren Martian terrain, and still be a desktop operating system to boot?

Unfortunately, too much of anything can end up being detrimental in the long run. One of the biggest problems software developers face is adapting their software to work with all distributions. Though this doesn't seem like a difficult task, it can rapidly get out of control. Each distribution of Linux is slightly different, and it can become a nightmare if a developer is working on a large scale product for Linux. Simple things such as installation procedures and locating other programs can become a challenge. This is where the Free Standards Group comes in with their Linux Standard Base (LSB). The goal of the LSB is to create general standards for the different Linux distributions to follow, including standard libraries, utilities, the layout of the filesystem hierarchy, and even standard installation methods.

However, like everything in the open source world as well as the rest of the world, standards face criticism. Most of the controversy surrounding the LSB is over the chosen installation package method, the Red Hat's Package Manager format. Zealous Debian GNU/Linux developers and users believe that the Debian package system is far superior. Likewise, Gentoo Linux must redesign its entire package system to conform to the LSB standards. Fortunately, both organizations have vowed to change and remain LSB compliant to help further the progress of Linux.

While some of the standards of Linux are still being hammered out, others have already proven success. One of the markets which Linux has always been pitched to was the developing market. These markets, not large enough to afford the high costs of proprietary software, can be anything ranging from small companies to large countries. This is where the internationalization of Linux comes into play. Linux, and computing in general, has always been written in English, and anyone who wants it in a different language has to do it themselves. Fortunately, the Free Standards Group realized the inefficiencies in this system early on and proposed the OpenI18N Linux internationalization standards (the 18 refers to the number of letters replaced in "internationalization").

The internationalization effort allows the input of complex languages, such as Chinese, Japanese, Arabic and more with the help of Unicode, an international character set containing characters from all languages. Internationalization allows for a much simpler translation process. With the help of OpenI18N, now a subset of the LSB, the Linux userbase of developing countries such as Africa has been growing rapidly. The internationalization efforts can be considered a Linux success story since Red Hat Enterprise Linux became the first vendor to achieve LSB Internationalization Runtime Environment certification.

One of the more important forefronts for Linux is actually one that is the least noticeable or even important to the average Linux user, but is either at the top of the list or very near it for corporate users. Security is of utmost importance to corporations and those of the tinfoil hat variety. Security-Enhanced Linux (SELinux), primarily developed by the National Security Agency, is a recent project which extends the basic access controls currently seen in Linux and any Unix based operating system. With SELinux, there is no longer the simple system of trusted administrators and untrusted users. Instead, SELinux uses a type of user and process hierarchy, only allowing the specific user or process to access information, other processes, or devices in the same level. This allows users and processes the minimum amount of access required for operation and no more, helping to prevent rogue applications from destroying critical systems or users from looking at the wrong information.

So what exactly does this mean to the end user? How does this influence Windows users to switch over to Linux? Well, these few things do not directly help the end user, but they do help make a substantial difference in the long run. All of these features help cut back on development time, allowing programmers to work on other aspects of the program, such as performance and usability. This means that open source applications such as the OpenOffice.org office suite and the K Desktop Environment will be perform and behave much better. The end result of all of these standards for Linux directly affects the users. No longer will throngs of Linux users get frustrated at the inefficiencies of the programs and switch back to the "easy to use" Microsoft Office and Windows.

Although the standardization of Linux at first seems to infringe on the core belief of customizability of Linux, it does not. The taming of Linux in no way hampers the customizability. Linux users are still able to tweak settings and features as they see fit, and in many cases it could become easier. With a standard setup, there is no guess work that needs to be done when a user wants to change something. Likewise, things are easier to fix when everything is in a standard place. Without standards, without a direction, Linux will end up fragmenting into several different directions as the behemoth Unix operating system has in the past. If all of the standards mentioned do succeed without major criticism, then Linux will have secured its place as it progresses from a geek operating system to a full fledged mission critical operating system for years to come, avoiding the fate the Ghost of Linux Future foretold.

Kevin Lynch is a sophomore majoring in computer engineering.