Open Sores

For a decade or two now, I've been talking about the many problems endemic to open-source development. Companies frequently exploit open-source software/libraries without respecting license terms. Cheap third party labor has traditionally been employed as a form of firewall, so that companies can benefit from source code without having to worry about license terms and while placing any potential liability on the third party. Certain types of highly skilled labor have become undervalued due to the expectation that the associated work product can be freely obtained.

Open-source culture itself, on the other hand, is often plagued with exactly the type of sociopolitical problems you would expect from the socioeconomic realities surrounding it. Thousands of people contributing to the devaluation of their own labor in a race to the bottom, all in the name of the common good, fed piecemeal on petty, ego-based incentives. The resulting cultural toxicity doesn't plague every open-source community, but it is prevalent, and it should be viewed as a symptom rather than the disease itself. It's the direct result of the conflict between open-source ideology and our underlying socioeconomic realities.

Those who enter into this world with a fresh perspective, fostering a simple sense of goodwill, are often shocked when they first encounter the culture and the disturbing infighting (often just to establish a meaningless pecking order) which is so prevalent in open-source communities. People may come to sate some sense of personal curiosity, to build reputation, to gain professional visibility, to entertain their own concept of altruism, or to simply derive satisfaction from solving problems. The kind of personality which lends itself well to investing days/months/years to freely maintaining an open-source project, however, more often than not also turns out to be fostering some form of insecurity. That insecurity can manifest itself rather insidiously as they carry out their communal role.

There are certainly counter-examples of the bad company behavior I've referenced; companies often "give back" by contributing patches upstream or even by financially backing open-source projects/organizations. However, this also harkens to a system of convoluted incentives; could there be a better way to devalue the labor I'm currently paying for than by making that labor freely available to all? This is always the consideration; in this particular instance, do we benefit more from privately capitalizing upon the labor or by simply using it to help devalue all labor of its kind? Look at where the lines of open/closed are drawn with any large software developer, and you'll see this methodology at work quite plainly. For them, it's always win-win.

This system also preys upon the good nature of humans; most of us want to contribute to the common good and see our work benefit others. This is the natural venue through which most people find meaning and satisfaction in life, and the open-source initiative would be perfect in a world not plagued by our socioeconomic realities. Unfortunately, we do all truly live in the relentless grip of oligarchy.

Whether under the guise of capitalism or communism, there is very little difference at this point, as we all find ourselves feeding into one corrupt hierarchy of wealth or another regardless of our geography. From the station of a software developer, any work of genuine value that one manages to produce is quickly seized upon by someone or something higher in the food chain. When we offer up that work product to the "common good", it's perversely leveraged against the common good by organically devaluing the labor that has gone into the work. We all suffer as a result. Our socioeconomic realities work directly against our nature as human beings; it's really quite fucked up.

Although we've seen software labor value grow in tandem with open-source projects over the last several decades, we've really been witnessing the explosive growth in demand for software across multiple industries. Open-source software has been an underlying factor, affecting labor value in some subsectors more than others, but its impact on labor value has been relatively limited. However, now we're coming up on the end of the long game; this is the capitalist's natural response to any explosive growth in labor value.

It was already slowly coming to a head, but we've recently entered into a new era of exploitation. Large Language Models have been trained on decades of free labor and a clear crisis in misattribution has been ongoing for years now. There have been plenty of people sounding alarm bells on this topic, but the reality is that it's not going to matter. There's simply too much power behind the organizations feeding on all of this delicious, free labor product. Legislation will move in favor of these powerful organizations as needed, data will be convolved, licenses will be ignored when convenient.

If you think these are overly pessimistic assumptions, it's worth considering that many of the socioeconomic mechanisms I'm describing under the framing of corporate exploitation have already been predicted by economists, but those same mechanisms have been more generously framed as part of an efficient way to share knowledge. (that paper is worth a read for a thoughtful analysis of open-source software in 2004) Indeed, generations of programmers have now engaged in these rituals, because this model of sharing feels right to us as human beings. That's precisely what makes the open-source model such an effective venue for exploitation under our present-day socioeconomic constraints. When I look around at the license carnage already occurring in the mass-digestion of open-source codebases, I find my "I told you so" reflex extremely hard to restrain.

I contributed to open-source projects myself in my early programming career, sometimes as myself and sometimes anonymously. I didn't contribute anonymously so that I could be "above it all", it was simply to avoid locking horns with the kind of petty gatekeeper who sits on top of many of these smaller open-source communities. As I became known in various circles, I realized that if the gatekeeper wasn't threatened by the person behind a contribution, they'd just take the contribution with no questions asked. If the gatekeeper is threatened, they'll work to the detriment of their own community to create arguments or disputes where none should exist. This is also part of the cultural toxicity running through so many of these communities, as they languish under a sort of socioeconomic oppression that they so often appear to be unaware of.

As my view on what open-source software really represents under our socioeconomic constraints evolved, I stopped contributing to open-source software almost entirely. This wasn't a self-serving decision, as I came to genuinely believe that all of that free labor was hurting the working class in my industry and creating negative reverberations even outside of software development. The bald-faced exploitation that has since unfolded around the aforementioned free labor has served to galvanize my conclusions.

Although my subjective experiences with cultural toxicity in some open-source communities didn't exactly encourage me, they also didn't play a role in my decision to stop contributing to open-source projects as a general policy. I'd be the first to point out that there are some extremely healthy and generally altruistic open-source communities out there, and that a lot of people benefit in a lot of ways from those communities. Those communities help create opportunities and help lower barriers to entry for people who really need those opportunities. I appreciate them for what they're trying to accomplish, but I also can't help but view the scenario as something ever-increasingly akin to "pulling drowning people onto a sinking lifeboat." Rather than being influenced by any subjective experience, my choice to generally isolate myself from open-source projects was entirely the result of my realization that open-source software was being leveraged as an insidious vehicle of capitalist exploitation.

I wholly anticipate that many people will take this as a criticism of Open Source, in large part because the cultural toxicity running through those communities creates incentives for lashing out at anything that threatens the established order. I'll be clear on that front, though; this isn't an attack on Open Source. If this is an attack on anything, it's an attack on our shared global socioeconomic order. Rather than a futile attack, however, this should serve as a source of uncomfortable truth to everyone making these observations and wondering if they're alone in seeing the self-injurious nature of the entire open-source ecosystem. No, my friend, you are not alone. Many of us see it, but far fewer take the time to sit down and articulate it.

Ultimately, I believe software labor value is now on an unavoidable trajectory; there's already enough free labor product for companies to exploit in the process of convolution and extrapolation. Make no mistake, however, we were already on this trajectory. The employment of LLM's has merely accelerated us on that same trajectory. With the assistance of LLM's or something else entirely, the natural forces of capitalism would be doing their best to pull us down this path.

My prediction here is that open-source datasets combined with large-scale datasets within large-scale organizations (e.g. source code for every component of Microsoft Windows) will quickly lead us to a situation where it's the "middle" talent that suffers the most. The top tier developers will remain in their roles to continue maintaining order and/or breaking new ground (guided by LLM assistance or not), and the "senior" roles will blend right into the entry-level roles; your job is to put the pieces together, and you're about to see your relative compensation plummet. As it begins to make more sense for companies to invest in infrastructure than middle-tier humans, you can suck it for whatever compensation you're offered or you can make way for more servers. It's harder to be precise about the longer-term impacts, but you can be sure that all of the same forces will keep pulling in this very same direction. The magnitude is the only variable, so just keep extrapolating as you descend into that abyss if you really want to see what's next.

Circling back to the role open-source software had in all of this, this dystopian bent simply wouldn't have been possible without decades of free labor for these organizations to feed upon. If open-source software weren't so abundant, these same developments would instead be creating a powerful advantage for companies with a wealth of privately owned data. Although that would have its own destructive effects, it wouldn't be accelerating the broad destruction of software labor value. One important lesson that every worker must learn under capitalism is that a large part of the capitalist's job is to seek the destruction of labor value. Open-source software, in all of the good and growth that it has given to so many individuals, has also served as a conduit for that destruction.

The contradiction here is that there's absolutely no argument from me against the idea that the open exchange of ideas and source code benefits everyone. This is the natural state of human collaboration; this is why Open Source is so attractive as both a practice and an ideology. It's just the right thing to do, and we all know it. Unfortunately, we're also living on top of a socioeconomic system that's truly and wholly corrupt. This system naturally corrupts every collective effort we make as a species. The original intent behind Open Source ideology is pure, good, and generally quite wholesome. What it has become in practice reflects what we have become in our global socioeconomic order; self-injurious and extremely contradictory.

On a more personal tangent, writing a new Killer Instinct emulator served as another unique exposure to this open-source culture. Although it's another entirely anecdotal account, it's worth relating my perspective there. The KI emulator came about because a company low-balled me with an offer to write a KI emulator for them; I still wanted to make the thing, but didn't want to contribute to facilitating the devaluation of labor in that sector. It's already bad enough (because of the contractors who shamelessly steal from open-source codebases), and taking the low bid hurts both you and your peers.

I wrote every component of the thing from scratch, going to great pains to avoid taking anything from open-source projects, and I took the trouble to implement correct timings and pipelining in my MIPS III implementation. I developed a pattern-training backend which allowed me to avoid any runtime code generation while still achieving speeds that are generally unheard-of even with a good dynarec implementation. In the end, the thing can easily run KI/KI2 without dropping a frame on any modern CPU, even if it's clocked at <300MHz. I was tempted to do a 3DS port, but that's still a sizable time investment. Maybe one day, just for fun.

My big incentive to make the emulator was in developing that backend, because the intent is to apply it to a much broader range of software and architectures, ultimately leading to something like a PlayStation 2 emulator which can achieve viable speeds on low-end hardware without any dynarec or software-specific AoT. Console vendors these days explicitly forbid any form of dynarec, even for first-party software, for security reasons. Which means there's still a lot of value in developing such a solution for the coming generation of "relive the PS2 game of your childhood" releases. I expect to be in the grave before we hit PS3 retro collections, so I'm not looking that far ahead.

In achieving speed and accuracy (replicating all of the original arcade timing quirks) that nothing in the open-source arena had achieved thus far, this attracted the attention and ire of one of the MAME developers. People were eager to give me links to this person shitting on the whole project and personally attacking me in various reddit threads. In justifying the existence of my project and talking about the things I was doing "better", I'd ruffled this person's feathers. MAME emulates thousands of machines, so it shouldn't be surprising that someone can come along and do a better job at any given one. I can also understand a default stance of mistrust, because so many people/companies don't share my scruples over "taking" from open-source projects. Unfortunately, this person went beyond mistrust and into the realm of personal attacks without even directly interacting with me beforehand.

I'll avoid the uglier details of that encounter, but the thing I want to highlight is how that person's behavior was "accepted" (to a degree, and to be fair, with some resistance from a minority of people) at the cultural level. It was akin to someone standing up in the middle of a cafeteria, covering himself in his own shit, and screaming obscenities. Everyone just carries on eating lunch. A few people look up, smirk and remark "lol, poop dude", then go back to eating. A few people come up and say "dude, you're crossing a line", then go back to eating. You know, screaming poop guys in the cafeteria are a pretty normal affair these days. It happens, and you know why it happens, so there's no reason to make a big deal about it. The real takeaway is "poop dudes are prevalent here", and that's a very bad sign for the health of any community.

In the end, we have this ecosystem that's produced a lot of great things and helped a lot of people overcome various barriers to entry. On the other hand, it's produced widespread cultural toxicity and has arguably been a major contributor to a software labor value apocalypse. If we had a reasonable, equitable underlying socioeconomic structure, we would have a beautiful software utopia instead of all of this. Again, pretty fucked up.

Because I believe there's very little to be done to change our present trajectory, I debated for some time whether it was even worth writing this article. I finally decided that it's worth adding a drop in the bucket to help validate this assessment for others. As we witness the unraveling of so many aspects of our society, and as we're subjected to so many nonsensical narratives to help sustain the status quo, these more grounded (albeit uncomfortable) narratives have a lot of potential value.

I do wish we lived in a better world, where everyone could freely contribute to the common good and everyone could benefit from those contributions. Sadly, we're being strangled and exploited by a relatively small group of people at the "top" of our socioeconomic hierarchy, and it's led to this tortured state of human existence. One might argue that because this is our reality, this is also our nature, but it's only as natural as the cancer I recently had cut out of my body.

On a personal note, if you're somehow still checking my web site and reading this organically, let me express my solidarity with you as a remnant of the Old Internet. I haven't been motivated to write much in recent years, going through a variety of personal tragedies. (I had cancer, my pancreas is actively trying to kill me, and we discovered my daughter had a brain tumor in August) As previously mentioned, I've also been spending a lot of my time getting into emulator development. Getting involved in that arena has meant that my stance on open-source software does automatically put me in something of a pariah position, which is understandable. As I stand by and watch hundreds of people shoot themselves in their own feet repeatedly, however, I have to admit that I feel a certain sense of hopelessness for my fellow humans.

Actually, I'm a little surprised to still be alive at this point, but my daughter having brain cancer took the wind back out of my sails. So if you hate me for writing an article which deconstructs the system from which you derive most of your self-worth, take comfort my pathetic little friend, for my suffering is already great.

Editor's note: Our friend Thomas is rather angry that Richard didn't provide references to back up his assertions in this article. We told Thomas that these are knowledge-based assertions written as part of an opinion piece, and that the author does not need to back up their body of personal knowledge with references in order to put their complex conclusions out into the world. As such, it's our recommendation that if you (like Thomas) are upset about the horrific lack of academic rigor present in this otherwise pristine work, you should ask ChatGPT to provide citations for any given assertion. It will do a great job, because that's what LLM's are actually good at doing. May your restless heart now find peace.

Comments

8 comments in total.
Post a comment

profsr_x

March 9, 2026 at 6:59 am (CST)
Thanks for writing this. I've felt a lot of this. If there is such a thing as good vibes or energy sent between humans, I'm sending it to you and your daughter.

grgr

March 9, 2026 at 4:27 am (CST)
well, i think it's time to separate the nerd from the consumer realms. it makes no sense anymore to strive a direct "commercial" use of open source efforts (i'll come to it, it's open source and non-free software). why? because 5 years ago, hard work in open software (say, another 20 hours per week on top of the regular 40) would buy you a petty shortcut of the assessment center. now it's not even that. why don't we stop feeding the beast in our free time(we continue of course to work for the industry during the day), and chose noncommercial open source (here comes the free/non-free thing), chose programming environments not seen in the industry (niche operating systems, other SCCM than git, not mainstream programming languages, what ever you prefer YOURSELF while ignoring peer and "must learn this new shiny" pressure) for our nerd efforts? we'll still have an interesting day job, and we will still be able to learn and grow in our free nerd time. but: we'll stop working for free.

Law

March 9, 2026 at 3:38 am (CST)
You're not alone in these beliefs. Ive been feeling like this a lot the last decade or so since it became obvious what cloud providers were doing and how little support open source was getting from the giants who profit from. GenAI is destroying so much with little societal benefit, and it's going to get much worse before laws catch up (if they even can). I hope your daughter makes a full recovery, as a parent I cant imagine what you're going through.

Rich

March 9, 2026 at 2:50 am (CST)
While the idea that resistance is not futile is a nice thought, if humans were capable of "standing against the scums of the world", we wouldn't be trapped in this corrupt hierarchy of wealth to begin with. It's hard enough to even get a reasonable percentage of the population to recognize that trash is trash, as evidenced by the exceptionally smelly dumpsters which we call "political institutions" across the world. Uniting people against established power is an even more impossible task, considering that established power already controls all the levers of public opinion and perception. All any of us can do now is try to find a way to survive without being part of the problem.

Ahmed

March 8, 2026 at 11:28 pm (CST)
hmm, well the thing is, doing something good/valuable whether for a benefit or not have the potential of facilitating an evil outcome. for example, a scientist discovers something that he truly believed would help people immensely, some evil scum uses it as a weapon. the solution isn't really for the scientist to not share his findings and live in a stone age but rather ACTIVELY STAND AGAINST THE SCUMS OF THE WORLD WITH EQUAL FORCE.

Rich

March 8, 2026 at 6:18 pm (CST)
I think any human attempt at altruism tends to be either coerced or self-serving in some way, even when we convince ourselves to be altruistic just for the sake of altruism. But to the extent that we are capable of it, I do think people have tried to contribute to open-source software under that premise. One might expect that such a premise is more likely deluded than not, and therefore also prone to creating more cultural problems as the "altruistic" individual may come to feel a bit more saintly than they really are. Come to think of it, I'm pretty sure I met that guy in one of the Quake communities. And yeah, I agree that our egos tend to lead us along by the nose, and that we live in a culture that goes out of its way to make that seem like that's something we should always embrace. We indulge and even encourage a great deal of terrible human behavior in the name of maintaining these venues for extracting productivity!

Anton

March 8, 2026 at 6:10 pm (CST)
Thank you!Just thank you !

Borko

March 8, 2026 at 3:28 pm (CST)
Yes. I doubt however that altruism is the incentive ever for "Open Source" contribution (I find it peculiar that we stopped talking about "Free Software" a couple of years back... I wonder what happened...) There are other industries where the insane human desire to be noticed has been exploited for a while. Academic research, professional sports, politics come to mind. And yes, the sweet dopamine hit we programmers get when we finally get it to work.... we are all addicts. I am recovering at the moment; and it is so wild to watch people around me (who are not even programmers!) start exhibiting the same addiction symptoms with GenAI. Fun times ahead even without the war.

Post a comment

Name:


Enter the following (refresh if you can't read it):



Comment:

Stream Tags

Projects

Everything

Personal

Analysis

Indie

Technology

AVALANCHE

R-Cube Engine

Modding

Game Design

Commercial

iPhone

Delight

Recent Comments