Few debates in the computer industry are as passionate as those around Free and Open Source Software (otherwise referred to as “FOSS”) and how it “benefits society” or promotes “ethics” or some other optimist outcome for the future of mankind. The problem is that this ongoing debate for the past 20 years has only served the purpose of making people look in the wrong direction.
During my first days in University, around 1994, a friend told me about a hobby operating system he installed in his computer, something called “Linux.” He had downloaded it from that venerable anonymous FTP site called “ftp.funet.fi.” He told me the long hours he had to spend to (I quote) “recompile the Kernel” and that he was happy because he was running “free software.”
I barely knew what a “kernel” was, even less why it would need to be recompiled, and I obviously thought that free meant “without cost.” When you are a cash strapped university student, free is always good; it does not really matter if it is a free lunch, a free beer, or a free kernel. Free is good™®©, or so they say.
I started my career writing software on Windows 95, and running it in Windows NT 4.0 boxes running SQL Server 6.5 and IIS; a rather closed source, “proprietary” galaxy. I kept on hearing stories of how Linux on Apache was better (read: faster, snappier, cheaper, etc.) than IIS; it might have been, but the business did not care. As closed as the whole Microsoft galaxy could be, the company I worked for could get a working website for a fraction of the infamous Total Cost of Ownership (aka TCO) of what it would have cost using Linux.
The tradeoff between both galaxies was simple: either buy sanctioned Wintel hardware with sanctioned Wintel software, pay a monthly fee for a MSDN subscription and get regular support and updates. Or get a Red Hat Linux CD-ROM from Walnut Creek, deal with non-existing Linux device drivers, and search for solutions in AltaVista, evolt, or Slashdot.
Everything changed in 1998. That is when Christine Peterson coined the phrase “Open Source,” deliberately masking the business-unfriendly moniker of “Free Software” which Wall Street abhored so much. More or less at the same time, Netscape decided to rewrite its own product. Such wise decision, widely celebrated by the “community” helped Microsoft win the infamous browser wars; this situation ultimately begat Internet Explorer 6 and its horrible crush on web development for the next decade.
The dot-com boom contributed to the success of FOSS; or did it? Well for a time it seemed so. Instead of using non-existent venture capital to bootstrap your company, just use FOSS. This is how AWS won over datacenters. This is how Ruby on Rails won over Java. This is how WordPress won over Movable Type. This is why 2005 is still remembered as the year of Linux on the Desktop. Well, not really, that’s actually 2020. FOSS became an interesting alternative mostly because proprietary options were quite bad by all standards; the whole Windows Vista scandal convinced most users that… Macs were worth considering again, which made Apple win a lot of money.
Apple; who was by then smart enough to open up key parts of their software, but not more than necessary. Of course, always taking whatever they needed to get things done. Hence Konqueror begat WebKit and much joy was had, and Apple’s share prices went up.
The pressure of the mob in the streets of Redmond right after the crash of 2000 was strong enough for Microsoft to do two things, more or less at the same time. First, Ballmer treated all FOSS as communism and later as cancer. I do not blame him; after all, even Robert Metcalfe said it and he knows a thing or two about technology:
The Open Sores Movement asks us to ignore three decades of innovation. It’s just a notch above Luddism. At least they’re not bombing Redmond. Not yet anyway.
The second thing Ballmer did was to release parts of the .NET Framework under a “Shared Source” license. To this day nobody understood exactly why, but thanks, why not.
Almost 15 years later, Ballmer resigned and shortly afterwards Satya Nadella bought GitHub for 7.5 billion USD. Then they decided to copy the open source AppGet package manager and release it as another open source project called WinGet, touted as a better option than the MS Store. Sherlocking is not just for Apple, after all; even better in the case of open source, you can actually read the code while building a competitor. Of course, in both occasions, Microsoft’s share price went up accordingly.
Microsoft was late to the game; Oracle had bought Sun Microsystems in 2009, paying more or less the same amount of money. Among all of the assets that Oracle got through that aquisition, there was MySQL; the “community” (whatever that is) reacted with anger, distrust, disgust, and dismay, and promptly “forked” the project, henceforth called MariaDB.
The Times They Are a-Changin’ and FOSS became big business.
It is such a big business, that IBM pulled a record 34 billion USD out of the pocket and bought Red Hat. That is, almost the double of the record 19 billion that Facebook paid for WhatsApp a few years earlier. FOSS is big business, because the truth is that people will use good tools, whether they are proprietary or not; the purpose of computing is to solve problems, and it turns out that people will happily pay actual money to get help from somebody.
The core of the issue is this one: if a tool is good at solving a particular problem, then others will pay for support contracts around them. That’s true whether they are “closed” or “open” solutions. This fact does not matter; they are solutions, and all considerations of goodwill and positive externalities of FOSS for mankind are simply discarded, because ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. I’m not screaming, the BSD license is.
Google is another great advocate of FOSS; hence WebKit begat Chromium, and much more joy was had. Chrome later crushed Firefox, and Microsoft ditched Edge’s own rendering enging to use Chromium instead, and Google’s stock price continued to rise. And Chromium is the new IE6, and websites now advertise themselves as “better viewed with…” just like in 1998.
In 2009, a Google blog article stated the phrase “Open will win” together with a long list of naive predictions. One year after, Andy Rubin famously defined what open means:
the definition of open: “mkdir android ; cd android ; repo init -u git://android.git.kernel.org/platform/manifest.git ; repo sync ; make”
Much fun was had by Apple pundits in the following decade about those perceptions. While Google might allow anyone to “compile the kernel,” handset manufactures must abide to quite strict Original Equipment Manufacturer (OEM) contracts. Do what I say, not what I do.
What neither the Google blog nor Andy Rubin explained is: what is won when “Open wins?” And by whom?
Michael Tiemann was right. Open always wins… money.
FOSS is an economic model of production based on unpaid work provided willingly by a certain amount of (sometimes starving) enthusiast developers. All impatient to add a line to their CV saying that they contributed to this or that project so that they can brag about it in Slashdot Reddit. Some of them, thanks to those pull requests, might ocasionally land a job at Red Hat IBM, GitHub Microsoft, Google, or some other FOSS company, because meritocracy and things like that. Better yet, they pat each other on the shoulder, happy to have “made the world a better place” thanks to “open tools” that they can “fork” if some “nasty corporation” messes up with them. Those who don’t get recognition rant about it on Medium, so that everybody gets angry online, corporations keep on being nasty, and the cycle repeats again.
As a production model, FOSS would have surprised them all: Adam Smith, David Ricardo, Karl Marx. Nobody could see this one coming, not even the most optimist of trillionaires.
