The X Situation (editorial)

Jim Knoble has written us an editorial on the recent situation involving the OpenGroup and their new cost structure related to X11 6.4. I'd say if Netscape was the biggest Open Source success of the year, this situation may very well be the biggest failure. Fortunately XF86 will remain free, and will very likely define the future of X. Anyway check the link below and read Jim's take on the situation.

The following is an editorial by Slashdot reader Jim Knoble Why the Hue and Cry? Thoughts on the Recent X11R6.4 `License' (or: What Can Public Radio in the United States Teach Us About Free Software?)

I've been spending far too much of my time recently following the ``Licensing charges for X11R6.4'' thread on comp.windows.x (at the time of this writing, over 227 articles in the thread, plus several spinoff threads).

In case the distinguished Slashdot readership isn't aware (uhh ... right), the Open Group, current stewards of the X11 copright, have announced a new licensing scheme for the recently released X11R6.4. Any one who makes money from selling X11 is required to pay a per-copy license fee. There is also a non-commercial use license which does not require the license fee.

According to the Open Group, ``We are unable to continue to support this technology without the participation of those companies who build business from re-selling this technology.''

That sounds reasonable, doesn't it? Those who make money from selling X11 ought to pass on some of the dough to the folks that make it available ... right?

Two questions pose themselves from the above:

1. Isn't ``selling'' X11 a rather broad term? What does it mean to ``build business from re-selling this technology''? Some of the Usenet discussion has centered around clarifying this point (especially because the license and other info on the Open Group's site are somewhat unclear). Does it mean that companies or organizations such as Red Hat, Caldera, SuSE, Debian, Walnut Creek, CheapBytes, etc., who were to distribute X11R6.4 as part of a Linux distribution which is sold on CD-ROM would have to pay the license fee? (According to a somewhat vocal member of the Open Group, they would). What about the XFree86 Project? (They're covered by the `non-commercial' license). What about folks who press CDs of archives which contain `non-commercial' X11R6.4 source code and sell them? (They'd have to pay). Et cetera.

This is not what i'm here to talk about. But it leads to the second question.

2. Is it necessarily obvious that the folks who make money from selling X11 ought to be the ones to support its development? This is what i'm here to talk about. I don't think it is.

Back for a moment to Usenet. Why is there so much uproar about this new `arrangement' for supporting development of X11? I think there are three reasons:

1. Change.
2. Surprise.
3. Uncertainty.

The announcement of a `commercial-use' license for X11R6.4 caught more than a few folks off guard. We've been so used to X being free software for so long that we didn't much consider the possibility that it might change. Many folks find change uncomfortable. Why fix what we didn't perceive as being broken? This change is one that potentially affects quite a few folks in the Linux community.

The new licensing scheme also casts some doubt on the future of X. Although the license fees are currently somewhat reasonable for a large volume of sales ($0.15US per unit for 50,000 units for non-Open-Group members), who is to say that the Open Group won't raise `prices' later? Or that X will remain `free' (think beer) for non-commercial use?

Here's the point: The Open Group's announcement of licensing fees for commercial redistribution of X11 is amounting to a public relations fiasco in the open source community ... especially on the heels of Netscape's official open-source-ification of Communicator 5.

Why is that? What did the Open Group do wrong? I think their biggest mistakes were the following:

1. They did not consult with the open source community.
2. They did not consider alternative methods of supporting the X Project Team.

1. They did not consult with the open source community. I am not a conspiracy theorist; i leave that to those who are better at it than i am. For whatever reason, the folks who make open source software obviously did not have much input into the decision to license commercial redistribution of X ... that goes against the very core of what open source software is. Had the Open Group consulted with the open source community, at the very least, the words ``reluctantly'' or ``We regret that'' would have appeared somewhere near the words ``There is a fee associated with...''. At best ... well, that brings me to the second mistake.

2. They did not consider alternative methods of supporting the X Project Team. Had the Open Group considered for a moment the importance of the relationship between the X Window System and the open source software community, they would have scrambled to look for alternatives to implementing per-unit redistribution license fees. Similar license fees are the reason almost no open source software is written using Motif or CDE. Putting license fees together with software that has historically been available without commercial redistribution fees is just asking for bad press.

What could the Open Group have done differently to avoid the slew of bad press they're getting now? I believe that they could have used a more collaborative approach rather than the authoritative one they chose. I also believe that the open source software community can learn from their mistakes. Let me explain.

First, a bit of background for the Slashdot readership outside the United States (if you know the significance of the words `National Public Radio', you can skip this paragraph). In the United States, besides the `regular' privatized commercial radio stations that broadcast thirty minutes of music and thirty minutes of advertisements per hour (or so it sometimes seems :), we have what are called `public' radio stations, whose broadcasts consist almost entirely of programming without advertisements. These radio stations used to be heavily funded by both the federal government and some state governments (this ought to sound familiar to residents of many countries in Europe and elsewhere, where such public broadcasting is the norm). In recent years, funding for public radio has gotten slimmer and slimmer, as state and federal governments have been required to `balance budgets' and `cut spending'. Many public radio stations now raise a large percentage of the money they need in order to operate by holding seasonal `on-air fundraisers', where regular programming is surrounded by the local radio staff asking listeners to donate money to the radio station.

Anyone who has heard their local public radio station announce that 70 percent or more of their yearly operating budget comes from listener donations knows how effective the seasonal fundraisers can be. The importance of this fundraising model is that it involves the listeners. Not only do the listeners get a chance to `feel good' about supporting public radio, but sending in their donation often gives them a chance to give feedback to the station, and the fundraiser suddenly becomes a two-way channel of communication. The listeners get a chance to collaborate in the work of the radio station.

I believe a similar fundraising model might have worked in the Open Group's favor. Instead of imposing an authoritarian license fee, they could have opened up the development of X as a collaborative effort: ``We're doing the main development ... can you help support us?'' Suddenly the whole view of the situation changes. And, X would not have distribution restrictions caused by silly license fees.

Then again, perhaps such a fundraising model would not have worked for the Open Group. Perhaps some other alternative (such as dissolving the X Project Team and donating X11R6.4 to the open source software community at large) would have worked instead, or would have worked even better. The point is that they acted without consulting the fastest-growing group of folks that would be affected, and without considering alternatives to the outdated `build once, license many' paradigm.

We are in the middle of a time when alternatives to the traditional methods of distributing and supporting software are not merely available, not just fashionable, but increasingly necessary. We need to find ways to maintain freely distributable software that benefit both ourselves and the community, otherwise we lose the support of the community, who are also our customers. Let's all learn from the Open Group's mistakes.