The Free Software Foundation: 30 years in
The Free Software Foundation: 30 years in
Welcome back, folks, to a new Six Degrees column. As usual, please send your thoughts on this piece to the comment box and your suggestions for future columns to my inbox.
Now, I have to be honest with you all, this column went a little differently than I expected.
A few weeks ago when thinking what to write, I mused over the notion of a piece about the Free Software Foundation celebrating its 30 year anniversary and how relevant and important its work is in today's computing climate.
To add some meat I figured I would interview John Sullivan, executive director of the FSF. My plan was typical of many of my pieces: thread together an interesting narrative and quote pieces of the interview to give it color.
Well, that all went out the window when John sent me a tremendously detailed, thoughtful, and descriptive interview. I decided therefore to present it in full as the main event, and to add some commentary throughout. Thus, this is quite a long column, but I think it paints a fascinating picture of a fascinating organization. I recommend you grab a cup of something delicious and settle in for a solid read.
The sands of change
The Free Software Foundation was founded in 1985. To paint a picture of what computing was like back then, the Amiga 1000 was released, C++ was becoming a dominant language, Aldus PageMaker was announced, and networking was just starting to grow. Oh, and that year Careless Whisper by Wham! was a major hit.
Things have changed a lot in 30 years. Back in 1985 the FSF was primarily focused on building free pieces of software that were primarily useful to nerdy computer people. These days we have software, services, social networks, and more to consider.
I first wanted to get a sense of what John feels are most prominent risks to software freedom today.
"I think there's widespread agreement on the biggest risks for computer user freedom today, but maybe not on the names for them."
"The first is what we might as well just call 'tiny computers everywhere.' The free software movement has succeeded to the point where laptops, desktops, and servers can run fully free operating systems doing anything users of proprietary systems can do. There are still a few holes, but they'll be closed. The challenge that remains in this area is to cut through the billion dollar marketing budgets and legal regimes working against us to actually get the systems into users hands."
"However, we have a serious problem on the set of computers whose primary common trait is that they are very small. Even though a car is not especially small, the computers in it are, so I include that form factor in this category, along with phones, tablets, glasses, watches, and so on. While these computers often have a basis in free software—for example, using the kernel Linux along with other free software like Android or GNU—their primary uses are to run proprietary applications and be shims for services that replace local computing with computing done on a server over which the user has no control. Since these devices serve vital functions, with some being primary means of communication for huge populations, some sitting very close to our bodies and our actual vital functions, some bearing responsibility for our physical safety, it is imperative that they run fully free systems under their users' control. Right now, they don't."
John feels the risk here is not just the platforms and form factors, but the services integrates into them.
"The services many of these devices talk to are the second major threat we face. It does us little good booting into a free system if we do our actual work and entertainment on companies' servers running software we have no access to at all. The point of free software is that we can see, modify, and share code. The existence of those freedoms even for nontechnical users provides a shield that prevents companies from controlling us. None of these freedoms exist for users of Facebook or Salesforce or Google Docs. Even more worrisome, we see a trend where people are accepting proprietary restrictions imposed on their local machines in order to have access to certain services. Browsers—including Firefox—are now automatically installing a DRM plugin in order to appease Netflix and other video giants. We need to work harder at developing free software decentralized replacements for media distribution that can actually empower users, artists, and user-artists, and for other services as well. For Facebook we have GNU social, pump.io, Diaspora, Movim, and others. For Salesforce, we have CiviCRM. For Google Docs, we have Etherpad. For media, we have GNU MediaGoblin. But all of these projects need more help, and many services don't have any replacement contenders yet."
It is interesting that John mentions finding free software equivalents for common applications and services today. The FSF maintains a list of "High Priority Projects" that are designed to fill this gap. Unfortunately the capabilities of these projects varies tremendously and in an age where social media is so prominent, the software is only part of the problem: the real challenge is getting people to use it.
This all begs the question of where the FSF fit in today's modern computing world. I am a fan of the FSF. I think the work they do is valuable and I contribute financially to support it too. They are an important organization for building an open computing culture, but all organizations need to grow, adjust, and adapt, particularly ones in the technology space.
I wanted to get a better sense of what the FSF is doing today that it wasn't doing at it's inception.
"We're speaking to a much larger audience than we were 30 years ago, and to a much broader audience. It's no longer just hackers and developers and researchers that need to know about free software. Everyone using a computer does, and it's quickly becoming the case that everyone uses a computer."
John went on to provide some examples of these efforts.
"We're doing coordinated public advocacy campaigns on issues of concern to the free software movement. Earlier in our history, we expressed opinions on these things, and took action on a handful, but in the last ten years we've put more emphasis on formulating and carrying out coherent campaigns. We've made especially significant noise in the area of Digital Restrictions Management (DRM) with Defective by Design, which I believe played a role in getting iTunes music off DRM (now of course, Apple is bringing DRM back with Apple Music). We've made attractive and useful introductory materials for people new to free software, like our User Liberation animated video and our Email Self-Defense Guide.
We're also endorsing hardware that respects users' freedoms. Hardware distributors whose devices have been certified by the FSF to contain and require only free software can display a logo saying so. Expanding the base of free software users and the free software movement has two parts: convincing people to care, and then making it possible for them to act on that. Through this initiative, we encourage manufacturers and distributors to do the right thing, and we make it easy for users who have started to care about free software to buy what they need without suffering through hours and hours of research. We've certified a home WiFi router, 3D printers, laptops, and USB WiFi adapters, with more on the way.
We're collecting all of the free software we can find in our Free Software Directory. We still have a long way to go on this—we're at only about 15,500 packages right now, and we can imagine many improvements to the design and function of the site—but I think this resource has great potential for helping users find the free software they need, especially users who aren't yet using a full GNU/Linux system. With the dangers inherent in downloading random programs off the Internet, there is a definite need for a curated collection like this. It also happens to provide a wealth of machine-readable data of use to researchers.
We're acting as the fiscal sponsor for several specific free software projects, enabling them to raise funds for development. Most of these projects are part of GNU (which we continue to provide many kinds of infrastructure for), but we also sponsor Replicant, a fully free fork of Android designed to give users the free-est mobile devices currently possible.
We're helping developers use free software licenses properly, and we're following up on complaints about companies that aren't following the terms of the GPL. We help them fix their mistakes and distribute properly. RMS was in fact doing similar work with the precursors of the GPL very early on, but it's now an ongoing part of our work.
Most of the specific things the FSF does now it wasn't doing 30 years ago, but the vision is little changed from the original paperwork—we aim to create a world where everything users want to do on any computer can be done using free software; a world where users control their computers and not the other way around."
A cult of personality
There is little doubt in anyone's minds about the value the FSF brings. As John just highlighted, its efforts span not just the creation and licensing of free software, but also recognizing, certifying, and advocating a culture of freedom in technology.
The head of the FSF is the inimitable Richard M. Stallman, commonly referred to as RMS.
RMS is a curious character. He has demonstrated an unbelievable level of commitment to his ideas, philosophy, and ethical devotion to freedom in software.
While he is sometimes mocked online for his social awkwardness, be it things said in his speeches, his bizarre travel requirements, or other sometimes cringeworthy moments, RMS's perspectives on software and freedom are generally rock-solid. He takes a remarkably consistent approach to his perspectives and he is clearly a careful thinker about not just his own thoughts but the wider movement he is leading. My only criticism is that I think from time to time he somewhat over-eggs the pudding with the voracity of his words. But hey, given his importance in our world, I would rather take an extra egg than no pudding for anyone. O.K., I get that the whole pudding thing here was strained...
So RMS is a key part of the FSF, but the organization is also much more than that. There are employees, a board, and many contributors. I was curious to see how much of a role RMS plays these days in the FSF. John shared this with me.
"RMS is the FSF's President, and does that work without receiving a salary from the FSF. He continues his grueling global speaking schedule, advocating for free software and computer user freedom in dozens of countries each year. In the course of that, he meets with government officials as well as local activists connected with all varieties of social movements. He also raises funds for the FSF and inspires many people to volunteer."
"In between engagements, he does deep thinking on issues facing the free software movement, and anticipates new challenges. Often this leads to new articles—he wrote a 3-part series for Wired earlier this year about free software and free hardware designs—or new ideas communicated to the FSF's staff as the basis for future projects."
As we delved into the cult of personality, I wanted to tap John's perspectives on how wide the free software movement has grown.
I remember being at the Open Source Think Tank (an event that brings together execs from various open source organizations) and there was a case study where attendees were asked to recommend license choice for a particular project. The vast majority of break-out groups recommended the Apache Software License (APL) over the GNU Public License (GPL).
This stuck in my mind as since then I have noticed that many companies seem to have opted for open licenses other than the GPL. I was curious to see if John had noticed a trend towards the APL as opposed to the GPL.
"Has there been? I'm not so sure. I gave a presentation at FOSDEM a few years ago called 'Is Copyleft Being Framed?' that showed some of the problems with the supposed data behind claims of shifts in license adoption. I'll be publishing an article soon on this, but here's some of the major problems:
- Free software license choices do not exist in a vacuum. The number of people choosing proprietary software licenses also needs to be considered in order to draw the kinds of conclusions that people want to draw. I find it much more likely that lax permissive license choices (such as the Apache License or 3-clause BSD) are trading off with proprietary license choices, rather than with the GPL.
- License counters often, ironically, don't publish the software they use to collect that data as free software. That means we can't inspect their methods or reproduce their results. Some people are now publishing the code they use, but certainly any that don't should be completely disregarded. Science has rules.
- What counts as a thing with a license? Are we really counting an app under the APL that makes funny noises as 1:1 with GNU Emacs under GPLv3? If not, how do we decide which things to treat as equals? Are we only looking at software that actually works? Are we making sure not to double- and triple- count programs that exist on multiple hosting sites, and what about ports for different OSes?
The question is interesting to ponder, but every conclusion I've seen so far has been extremely premature in light of the actual data. I'd much rather see a survey of developers asking about why they chose particular licenses for their projects than any more of these attempts to programmatically ascertain the license of programs and then ascribe human intentions on to patterns in that data.
Copyleft is as vital as it ever was. Permissively licensed software is still free software and on-face a good thing, but it is contingent and needs an accompanying strong social commitment to not incorporate it in proprietary software. If free software's major long-term impact is enabling businesses to more efficiently make products that restrict us, then we have achieved nothing for computer user freedom."
Rising to new challenges
30 years is an impressive time for any organization to be around, and particularly one with such important goals that span so many different industries, professions, governments, and cultures.
As I started to wrap up the interview I wanted to get a better sense of what the FSF's primary function is today, 30 years after the mission started.
"I think the FSF is in a very interesting position of both being a steady rock and actively pushing the envelope."
"We have core documents like the Free Software Definition, the GNU General Public License, and the list we maintain of free and nonfree software licenses, which have been keystones in the construction of the world of free software we have today. People place a great deal of trust in us to stay true to the principles outlined in those documents, and to apply them correctly and wisely in our assessments of new products or practices in computing. In this role, we hold the ladder for others to climb. As a 501(c)(3) charity held legally accountable to the public interest, and about 85% funded by individuals, we have the right structure for this."
"But we also push the envelope. We take on challenges that others say are too hard. I guess that means we also build ladders? Or maybe I should stop with the metaphors."
While John may not be great with metaphors (like I am one to talk), the FSF is great at setting a mission and demonstrating a devout commitment to it. This mission starts with a belief that free software should be everywhere.
"We are not satisfied with the idea that you can get a laptop that works with free software except for a few components. We're not satisfied that you can have a tablet that runs a lot of free software, and just uses proprietary software to communicate with networks and to accelerate video and to take pictures and to check in on your flight and to call an Über and to.. Well, we are happy about some such developments for sure, but we are also unhappy about the suggestion that we should be fully content with them. Any proprietary software on a system is both an injustice to the user and inherently a threat to users' security. These almost-free things can be stepping stones on the way to a free world, but only if we keep our feet moving."
In the early years of the FSF, we actually had to get a free operating system written. This has now been done by GNU and Linux and many collaborators, although there is always more software to write and bugs to fix. So while the FSF does still sponsor free software development in specific areas, there are thankfully many other organizations also doing this."
A key part of the challenge John is referring to is getting the right hardware into the hands of the right people.
"What we have been focusing on now are the challenges I highlighted in the first question. We are in desperate need of hardware in several different areas that fully supports free software. We have been talking a lot at the FSF about what we can do to address this, and I expect us to be making some significant moves to both increase our support for some of the projects already out there—as we having been doing to some extent through our Respects Your Freedom certification program—and possibly to launch some projects of our own. The same goes for the network service problem. I think we need to tackle them together, because having full control over the mobile components has great potential for changing how we relate to services, and decentralizing more and more services will in turn shape the mobile components."
I hope folks will support the FSF as we work to grow and tackle these challenges. Hardware is expensive and difficult, as is making usable, decentralized, federated replacements for network services. We're going to need the resources and creativity of a lot of people. But, 30 years ago, a community rallied around RMS and the concept of copyleft to write an entire operating system. I've spent my last 12 years at the FSF because I believe we can rise to the new challenges in the same way."
In reading John's thoughtful responses to my questions, and in knowing various FSF members, the one sense that resonates for me is the sheer level of passion that is alive and kicking in the FSF. This is not an organization that has got bored or disillusioned with its mission. Its passion and commitment is as voracious as it has ever been.
While I don't always agree with the FSF and I sometimes think its approach is a little one-dimensional at times, I have been and will continue to be a huge fan and supporter of its work. The FSF represent the ethical heartbeat of much of the free software and open source work that happens across the world. It represents a world view that is pretty hard to the left, but I believe its passion and conviction helps to bring people further to the right a little closer to the left too.
Sure, RMS can be odd, somewhat hardline, and a little sensational, but he is precisely the kind of leader that is valuable in a movement that encapsulates a mixture of technology, ethics, and culture. We need an RMS in much the same way we need a Torvalds, a Shuttleworth, a Whitehurst, and a Zemlin. These different people bring together mixture of perspectives that ultimately maps to technology that can be adaptable to almost any set of use cases, ethics, and ambitions.
So, in closing, I want to thank the FSF for its tremendous efforts, and I wish the FSF and its fearless leaders, one Richard M. Stallman and one John Sullivan, another 30 years of fighting the good fight. Go get 'em!
This article is part of Jono Bacon's Six Degrees column, where he shares his thoughts and perspectives on culture, communities, and trends in open source.