Free and open source software (FOSS) began as an effort to create an option to proprietary software. For years, it was generally viewed in that way on both sides, with FOSS developers mistrusting proprietary software, and corporations equally dismissive of FOSS. However, over the years, the distinction has blurred. Today, FOSS has become more successful than its pioneers had ever imagined, yet that success may have come at too large a price. Although a minority still dream of a completely free alternative, increasingly the emphasis in FOSS seems to be on accepting coexistence with proprietary software. In fact, a few question if existing FOSS licenses have outlived their usefulness and if an alternative is needed.

While this perception is not widespread, consider these data points:

• Until recently, Debian organized its repositories on the assumption that users wanted a free system. In October 2022, in a general resolution, the project voted to include a non-free firmware repository in the standard installation. The repository contains the hardware drivers necessary for maximum hardware performance.
• Nobara is a distribution based on Fedora, but with proprietary drivers and gaming applications that Fedora does not carry. Although only a couple of years old, it is solidly in 13th place on DistroWatch's Page Hit Ranking.
• In early 2024, the SCM License was released by the Free Software Foundation (FSF). The license is a modified version of the GNU General Public License (GPL) with an exception clause. The exception is that “if you link the SCM library with other files to produce an executable, this does not by itself cause the resulting executable to be covered by the GNU General Public License. Your use of that executable is in no way restricted on account of linking the SCM library code into it.” In other words, the license allows a link to proprietary software that would otherwise be impossible.
• In June 2023, access to the source code of Red Hat Enterprise Linux (RHEL) was restricted to individual developers and 16 servers. Instead, access to the code would be via CentOS Stream, the development version of RHEL. The move is widely condemned as a violation of the spirit and purpose of open source and has sent commercial rivals scrambling to deal with the policy.

What these events show is that FOSS is no longer an end in itself. Instead, large portions of the FOSS community have become more pragmatic. Rather than attempting to create an alternative to proprietary software, the emphasis has switched to how FOSS can interact with proprietary software and business practices.

On the one hand, this shift in priorities might be seen as a belated dose of realism. The FSF may maintain a list of completely free distributions, but many FOSS supporters value performance as well as software freedom. If that means using proprietary drivers or applications until a FOSS alternative is perfected, many will do so. Many, too, have always preferred the permissive BSD licenses, which permits combining FOSS and proprietary software. From some perspectives, Debian’s newest repository or Nobara’s popularity is simply an admission of the true state of affairs. As for RHEL’s source code access, although it means more work for community-based distributions like AlmaLinux and Rocky Linux, the fact that they can still function might indicate that FOSS licenses are still able to defend against attempts to work around them.

On the other hand, the FOSS philosophy may be weakened because it no longer has a strong advocate. Sixteen years ago, the FSF reached a peak of authority in the discussions of 2006-2007 about the structure of GPLv3 – then immediately lost that authority by not reaching a consensus. That was followed by the cancellation of Richard Stallman in 2017, which, deserved or not, had the side effect of silencing free software’s most influential representative. Today the FSF that Stallman led continues to function, with Stallman returned to the board of directors, but its actions go unreported, and it seems to speak to a much smaller group of loyalists. The Linux Foundation, with its corporate emphasis, is not an adequate substitution. In these circumstances, there is reason to wonder whether FOSS has lost its way.

While the issue has yet to reach the mainstream, Bruce Perens, one of the coiners of the term “open source” in 1998, is already trying to describe what he calls the Post-Open Source era. Not only does Perens believe that FOSS licenses no longer fulfill their original purpose, but they no longer inform or benefit the average user. According to Perens,

“Open Source has completely failed to serve the common person. For the most part, if they use us at all they do so through a proprietary software company’s systems, like Apple iOS or Google Android, both of which use Open Source for infrastructure but the apps are mostly proprietary. The common person doesn’t know about Open Source, they don’t know about the freedoms we promote which are increasingly in their interest. Indeed, Open Source is used today to surveil and even oppress them.”

As a remedy, Perens proposes that licenses should be replaced by contracts. He envisions that companies pay for the benefits they receive from using FOSS. Compliance for each contract would be checked, renewed, and paid for yearly, and the payments would go towards funding FOSS development. Individuals and nonprofits would continue to use FOSS for free.

In March 2024, Perens posted a draft Post-Open license. The draft includes a description of the contract-related files to be shipped with FOSS software, a description of the status of derivative works, how revenue is collected, and conditions of termination. The draft has yet to be reviewed by a lawyer, but what is immediately noticeable is how it draws on both contract language and FOSS licenses to produce something different. For example, under derivative works, it includes shims, pieces of software that act as an intermediary between proprietary and FOSS software, like the one Debian developed for Secure Boot.

Whether Perens’s ideas are legal is still uncertain. As he points out, they also needs a formal mechanism for the collection of fees, as well as a popular movement to promote them. At best, they are a long-term solution. The problem is, modern FOSS is so large and diverse that a stance that was possible in the early days of FOSS seems far less likely today. The little attention that the draft has received in the five months since it was posted perhaps indicates how difficult encouraging participation may be.

And what about the corporate response? Will companies agree, however reluctantly, to pay for what they currently receive for free? Some might argue that they already pay for FOSS through contributions for development. And would Post-Open contracts cause some to walk away from FOSS, or are they already so invested that they cannot afford to?

Still, if nothing else, Perens’s efforts are a starting point for discussions. As the events described above suggest, FOSS has shifted direction and moved away from its original intention. Moreover, free licenses are straining to respond to loopholes, and a discussion needs to be had about whether they are adequate to modern pressures. If the Post-Open contract is not the only solution, at least it is a start towards addressing problems that are unlikely to be solved overnight.