[LWN Logo]

Bringing you the latest news from the Linux World.
Dedicated to keeping Linux users up-to-date, with concise news for all interests

 Main page
 On the Desktop
 Linux in the news
 Linux History
All in one big page

Other LWN stuff:
 Daily Updates
 Linux Stocks Page
 Book reviews
 Penguin Gallery

 Use LWN headlines
 Advertise here
 Contact us

Recent features:
- RMS Interview
- 2001 Timeline
- O'Reilly Open Source Conference
- OLS 2001
- GaŽl Duval
- Kernel Summit
- Singapore Linux Conference
- djbdns

Here is the permanent site for this page.

See also: last week's LWN.

Leading items and editorials

A new, proprietary web? Back in mid-August, the World Wide Web Consortium (W3C) put out a draft policy on the incorporation of patented technology into web standards. The W3C, evidently, did not feel much need to publicize this draft, and it came dangerously near to ending its comment period with almost nobody realizing what was in it. That would have been an unfortunate development. Fortunately, Adam Warner was paying attention and issued a detailed call for action telling the community what was up.

So, what is up, exactly? The draft policy would allow the W3C to incorporate patented technology into web standards as long as the patent holder agreed to license the technology in a "reasonable and non-discriminatory" manner. This policy, called RAND, makes it possible for vendors of proprietary products to know that they can use the given technology.

The problem, of course, is that it is not generally possible for free software to use patented algorithms. If an implementation of an algorithm can not be distributed without the payment of royalties, the code obviously can not be put up for download on the net. And the GPL, of course, does not allow the inclusion of code with such restrictions.

Imagine a future web whose standards include patented technology. That is a web that can only be accessed with proprietary software - a very different web than the one we have now. It is, conceivably, a web without Apache, Mozilla, Konqueror, and many other tools we depend on. It's a web that would make certain vendors very happy, since it would eliminate the free software threat.

This is a real scenario, and one that could happen quickly. Consider the "Scalable Vector Graphics" (SVG) standard, adopted by the W3C on September 4. This standard, in fact, includes patented technology (of the worst, stupid software patent variety) from Apple, which will be available under RAND terms. In other words, the W3C is already behaving as if the new policy were in force, which casts some doubt on its public comment policy. SVG is not only proprietary, but the licensing terms are not publicly available. SVG, in other words, is inaccessible to free software. See Daniel Phillips's well-researched comments for more information on SVG.

There is a possible alternative scenario, of course to the proprietary web: the W3C, by endorsing proprietary standards, finds itself left behind by a web that does not want to go that way. The Open Group's attempt to take X11 proprietary has been put forward as an example of how this could happen. That time, however, we had the XFree86 project, which was already the real heart of X11 development. It is not clear that there is a body that is well positioned to supersede the W3C in this manner. It may not be so easy this time.

If necessary, however, that is what we will have to do.

In response to the last-minute outcry, the W3C has extended the comment period, but only to October 11. Now is the time to send in polite, well-reasoned arguments on why this policy should not be adopted. The people attempting to push through this policy know very well what they are doing, but they may still respond to a determined show of opposition. There may still be time to make a difference here. Perhaps we can avoid a corporate takeover of the web.

(See also: the patent policy comment list, which contains the comments posted so far. Included therein are comments from Alan Cox, Andrew Tridgell, Bruce Perens, Chuck Mead, Dan York, Eben Moglen (on behalf of the Free Software Foundation), Ian Clarke, Jeremy Allison, Joe "Zonker" Brockmeier, John Gilmore, Richard Stallman, Russell Nelson, Theo de Raadt, Tim O'Reilly, and many, many others).

Opportunities in migration services? Last week, the Gartner Group suggested that businesses should consider moving away from IIS toward other, more secure web servers. The latest Netcraft survey suggests that a number of businesses are doing exactly that - tens of thousands of IIS-based web sites have disappeared from the net recently. Corporate IT operations will stay with a given course through inertia for a long time, but, with a sufficient push, they will start looking at alternatives. Microsoft's more recent licensing and upgrade policies are giving more businesses reasons to look around as well.

It seems there should be an opportunity here. After being told for years that superior alternatives exist in the form of free software, some businesses are starting to look at them. But if all those businesses encounter is a pointer to a download site and a collection of HOWTOs, they may not look for very long. Now is the time for enterprising free software businesses to step forward with well-designed, targeted migration services.

Starnix has been quick to see the opportunity here: the Toronto-based company has announced a new service to help companies migrate away from IIS. The package includes a version of Apache hosted on a secured Linux system, consulting, and support services. Here is an offering that should be successful; it is a narrowly-targeted service which meets a pressing business need.

There should be other such opportunities in the free software world. Linux is increasingly ready to take on a wider role in corporate computing, and that will create opportunities for companies that can help. Those seeking to profit off such opportunities should proceed carefully, however; there are several prerequisites that must be met:

  • There must be a real motivation for companies to change. For many, Apache's advantages have been insufficient until the recent set of IIS security problems started causing real pain. The blue screen of death does not appear to be enough to make many desktop users look elsewhere. The pressure to adopt a new solution has to be strong before most people will think seriously about it.

  • Companies offering migration services must host substantial expertise on both the old and new systems. Linux expertise alone is insufficient; you must know all about the system you intend to replace as well. As Larry Augustin pointed out a few weeks ago, many people who are pushing a Linux desktop have failed to understand what Microsoft power users are really doing with their systems. Approaching the market with only half the necessary knowledge is a recipe for failure.

  • There must be convincing support options available. Nobody likes the fear that they could be stuck trying to solve problems on their own.

  • The free software alternative must be truly superior. For most users, the fact that a particular program is free is not enough; it must be demonstrably better at solving their problems. The free office suites, despite their amazing progress, have probably not yet reached that point for many users.

Free software support offerings probably need another look as well. These offerings look much like support services for proprietary software. The software in question becomes a black box, and the support provider takes charge of making it work. But one of the advantages of free software is that it is not a black box. "We'll take charge" services can deprive their customers of some of the advantages of free software. True free software support services should bring their customers into the free software community.

Helping customers migrate from proprietary products may not be the most exciting business to be in. And it certainly will not be an easy business. But it could well prove to be a workable business model for those working with free software, even in these difficult times.

Inside this LWN.net weekly edition:

  • Security: The top 20 Internet security vulnerabilities; more trouble with PHP.
  • Kernel: Maintaining 0.01; security module licensing; handling high network loads.
  • Distributions: Buy a Box Set or Download?; Mandrake Linux 8.1; SuSE Linux 7.3. MaxOS and BearOps Linux.
  • On the Desktop: The gamers way - playing games on Linux, Sun releases StarOffice 6.0 beta and Ximian adds channels.
  • Development: Dedicated Systems, measuring cluster performance, LPRng 3.7.8, Midgard 1.4.2, Python 2.2a4.
  • Commerce: SuSE closes funding round; C|Net sells LinuxDevices.com; Alta Terra Ventures Corp.
  • History: What will happen to the Linux VARs?; Linux Myths; Corel/Microsoft alliance.
  • Letters: Audio editors; patented web standards; non-free security modules.
...plus the usual array of reports, updates, and announcements.

This Week's LWN was brought to you by:

October 4, 2001


Next: Security

Eklektix, Inc. Linux powered! Copyright © 2001 Eklektix, Inc., all rights reserved
Linux ® is a registered trademark of Linus Torvalds