10 Apr 2004


In the course of several public appearances this past month, I’ve found myself being asked about “coexistence” surprisingly often. “Can free and proprietary software work together?” seems to be the question of the day.

In one respect, the very asking of the question is an extraordinary demonstration of the change in the global software industry, particularly when (as happened recently) I’m asked by the chief executive of a major multinational software firm. Just a few short years ago the proprietary software industry was almost entirely ignorant of our existence. Now they are wondering outright about the terms of coexistence.

Sometimes the coexistence question seems to arise in a relatively narrow context, as a question about the GPL. “How can I have customers run my proprietary application on Linux-based systems without the GPL requiring me to make my application open source?” Put this way, the question reveals deep misimpressions about how free software licenses, and the copyleft in particular, work. I explain that the licensing structure originated by the Free Software Foundation for the GNU Project—licensing the operating system kernel and userland tools under GPL and the C library and other essential API components under LGPL—was specifically designed to make it possible for programs released under any license terms to be distributed for use on free software systems, while ensuring that enhancements to the OS itself would remain permanently free. The result is a system that offers coexistence not by accident, but by design. The free software movement is committed to the freedom of free software, but our licenses have been designed to achieve that goal while preserving complete interoperability with proprietary software. Copyrighting interfaces and restricting people’s ability to program to them is an unacceptable practice, in the movement’s view, as unacceptable as controlling the content of scientific research through copyright. Our legal theory is consistent with our technical approach: we work with anybody who works with us.

But the issue of coexistence also arises in a larger context. “How can any large-scale enterprise make money out of this stuff?” wondered one senior manager I spoke to recently. Of course, organizations such as IBM and Hewlett-Packard have realized that they can make money from free software. Novell is essentially betting the farm on what we can do for the firm, even as Apple has reinvigorated its core business with the massive infusion of free software that powers Mac OS X. Hardware and services—everything from software support to network architecture—can make money coexisting with free software. If, like IBM or Hewlett-Packard, businesses in those lines also have successful proprietary software products, they can often succeed, at least for a while, in fostering their established products against free competitors. But the balance between free and proprietary software in the customers’ installations is not their primary concern; they can succeed no matter how high up what they call “the software value chain” the principle of freedom ultimately extends.

For the free software movement, on the other hand, firms’ success or failure in the marketplace is immaterial, just as scientific academies should be indifferent to the price of pharmaceutical shares. We want the knowledge expressed in software to be free knowledge, available to every user, modifiable by every user, available to be shared among people as seeds are shared among farmers, tools among carpenters, books among scholars. We regret the lost knowledge locked up in programs that users aren’t allowed to understand, and the frustration experienced by technically self-reliant people injured by software bugs they weren’t allowed to fix. Monopoly is a problem experienced in the market. What the movement cares about is the technical and social harm pervasive unfree software does in an extensively networked society. Everybody bemoaned the monopoly, ineffectively, until—pursuing our broader aim—we showed them something truly effective they could do about it.

Thus free software has become not only a movement for liberty of thought, not only a technological project of endless interest, complexity and power, but also an instrument of competition within a competitive industry. Our sole aim was, and remains, to institute and protect freedom. We have begun rather well. But in the next phase of its history, our movement will be part of a new and more complex context: strongly attacked and resolutely defended in the midst of an industry undergoing disruptive change. Coexistence may well become more difficult.

This column was first published in the UK in Linux User. It is also available in PostScript and PDF formats.

permalink | columns/lu | 2004.04.10-00:00.00

Comments are closed for this story.

Trackbacks are closed for this story.