LWN.net Weekly Edition for June 26, 2003

The age of the Linux desktop

Not that long ago the Giga Group - not one of the more friendly analysts-for-hire out there - warned businesses to stay away from desktop Linux until 2005. More to the point, desktop Linux has long been regarded as a distant dream, even by many strong Linux supporters. It is commonly assumed that Linux is still far from where it needs to be to move out of the server and onto the desks of "ordinary people."

The Giga Group may be right that it will take another year or two before Linux is a common and safe choice for desktop deployments. For everybody who does not seek permission from analysts, however, desktop Linux is becoming a viable option rather sooner.

Consider, for example, the deployment of 80,000 desktop Linux systems in Extremadura, Spain. Linux and the GNOME desktop were considered to be more than good enough for students across the region; Linux systems were also used to set up 33 centers for general use.

Or consider CorelRescue, a shareholder effort to block the acquisition of Corel by Vector CC Holdings. This group, which claims to have over four million shares committed to voting against the acquisition, believes that Corel would be better off to continue as an independent company and reinvigorate its Linux desktop efforts - especially WordPerfect. Tux was even drafted as the group's logo.

For the clincher, consider this Business Week article about Apple, which happens to mention the following:

Apple's MacOS is generally considered to be the most advanced desktop operating system out there. That perception may not change, but the fact is that users are voting with their keyboards. Linux will displace MacOS as the second most popular desktop operating system within the year.

Once, not all that long ago, Linux was considered to be a toy system suitable only for hobbyists. Over time, Linux has proved its worth in many contexts, from personal video recorders to supercomputers - a Linux cluster is now the third fastest computer on the planet. Success on the desktop has taken longer, but it is now within reach. Nobody can say that a system which has surpassed Apple in the marketplace is not suitable for the desktop.

Comments (10 posted)

Legislative notes

As mentioned here last week, there has been a renewed push for the adoption of software patents in Europe. It now appears that the final scenes will be played out even more quickly than expected: according to the European Parliament Observatory, the full plenary vote on software patents could happen as soon as June 30. That does not leave a whole lot of time for concerned Europeans to contact their MEPs and get their feelings across. According to some sources (see, for example, this writeup by Xavi Drudis Ferran), it should not be assumed that the plenary session will simply rubber-stamp the software patent directive. Efforts to educate parliament members over the next few days could have a significant effect.

On the other side of the pond, representatives Zoe Lofgren and John Dolittle have announced their intent to introduce the Public Domain Enhancement Act into Congress. The PDEA was covered here in the June 5 Weekly Edition; it would require that copyrights be renewed after fifty years. Any material for which the copyrights are not explicitly renewed would pass into the public domain. This law would not reduce the copyright protection available to anybody; it would just ensure that works which are no longer being commercially exploited become part of the intellectual commons. The idea should not be particularly controversial, but the media industry is likely to lobby against it just the same. So it could be a long path between the introduction of the PDEA into Congress and its becoming law. That introduction is a necessary step in the right direction, however.

Comments (2 posted)

Making money with free software

[This article was contributed by Joe 'Zonker' Brockmeier]

The Linux Router Project is dead. So says Dave Cinege, the creator of the project. Though the project has been stagnant for some time, it still came as a surprise to see it officially pronounced dead, particularly given the bitterness of Cinege's eulogy for the project:

Apparently the cause of death was Cinege's inability to translate his work with LRP into a source of income.

While it's unfortunate that Cinege didn't benefit financially from his work on LRP, it's also an illustration that developers shouldn't depend on their contributions to free and open source software to land them a job or otherwise put money in their pockets. While a number of developers have, indeed, landed jobs as a result of their work with open source, it's hardly a guarantee of gainful employment. And it's true that companies may not even choose to publicly acknowledge the projects they've used to build their products. Vortech Consulting, for example, based Coyote Linux on LRP, but there's nary a mention of the Linux Routing Project on the Coyote Linux site.

The relationship between free software developers and companies is often uneasy. A recent bit of company bashing on the linux-kernel list led to this outburst:

While Cinege and many others see commercial companies as parasites using their work for profit without any kickback for the original contributors, others see open source as a parasite on proprietary software. There is a fair amount of mistrust and misunderstanding going in both directions. Many unknowns remain in the equation of how free software and money-making enterprises will work together; this situation is likely to persist for some time.

It's very clear right now, however, that if a developer hopes to earn a living off of their contributions to open source, he or she will need to come up with a workable plan beyond releasing software and hoping for job offers, contract work or grants to fund further development efforts. Even then, as with any entrepreneurial enterprise, it's no guarantee that they'll be able to pull it off. And, it's possible that someone else will come along and do a better job of capitalizing on your work. Part of releasing software under an open source license is giving up full control of the work.

Writing software is just one aspect of what makes a software company, open source or otherwise, successful. Brilliant software isn't enough to ensure a steady flow of clients. Developers who want to make a living off of their open source project will also need to wear the marketing hat, the sales hat, and so forth to turn a freely-available project into money. Some developers aren't interested or adept at doing those things, which is fine. In that case, they need to align themselves with partners or a company that will do that work for them if they hope to turn open source development into a money-maker. That, or resign themselves to the idea that someone else may do it without them.

Comments (5 posted)

Page editor: Jonathan Corbet

Inside this week's LWN.net Weekly Edition

• Security: Greylisting; new vulnerabilities in ethereal, osh, xterm, ypserv, others...
• Kernel: Multiple initialization functions; fixing cryptoloop; looking forward to 2.7.
• Distributions: Lindows.com - Friend or Foe?; Debootstrap/LVM with LNX-BBC 2.1; KnoppiXMAME
• Development: Contracts for Python, New versions of SAP DB, FreeMED, mnoGoSearch, Plone, WebSphere SDK, BEAST/BSE, CheeseTracker, Horgand, WaveSurfer, Mozilla, BZFlag, Sodipodi, SPTK, Netatalk, Wine, AbiWord, KOffice, GnomeMeeting, OpenMCL, PHP, Zsh, GDB.
• Press: Another pile of SCO articles, Linare offers cheap Linux PCs, Linux in the Airlines, desktop Linux history, Torvalds interview, Linux supercomputer is third fastest.
• Announcements: 4th Libre Software Meeting, Linuxtag in Karlsruhe, EuroPython report, International PHP Conf. CFP, GU4DEC slides, Ogg Vorbis Help.
• Letters: Finding SCO's code in Linux

Next page: Security>>