That mostly is a symptom of actually trying to offer a Product. Donations alone aren’t cutting the full time staff that you need to provide any kinds of guarantees. I mean, look at Lemmy. There was a CVE a few months ago and people were losing their minds that the two main devs didn’t take time to do a proper writeup and instead focused on fixing the issue while keeping an eye on the community discussing it and informing others. It was amazing that they were able to get it fixed so quickly but “NOT GOOD ENOUGH”. And… the way you get the resources to be “good enough” mostly involves monetization. And, shockingly, most of the people who get angry when something is added and say they would have donated otherwise aren’t speaking in good faith.

As for RedHad: Disclaimer, I am currently in the process of working with partners over the debacle of Red Hat seeming hell bent on killing Rocky. And… this is after they killed CentOS. As a developer, having a “free” version of the OS we are targeting is incredibly useful. It lets us nail everything down at low cost and run a limited subset of paid nodes for final debugging before we send it to the customer who requires RHEL.

But also? A LOT of end customers also run a metric shit ton of Rocky nodes. Unless it is outward facing or can’t tolerate any downtimes, use the “good enough” free version. Which then leads to discussions of “okay… if most of the fixes end up in Rocky a day or so later than in RHEL, do we really need to pay for a couple hundred licenses when we could instead split Fred and Chris’s time for internal support?”. Which rapidly leads toward most of the paid customers of RHEL not being paid customers anymore.

It is shitty but… that is the difference between consumer and enterprise. When a failure leads to an hour or two of my personal time to set Nextcloud up again because the underlying infrastructure hates the idea of containerization? Whatever, I genuinely find that fun. When a failure leads to a bunch of angry customers who now have lost faith in me and a bunch of employees working frantically so that we don’t lose a customer and, thus, revenue? We aren’t going to be risking anything to a hobbyist platform that is prone to catastrophic errors because there just isn’t time to fix it (also, Nextcloud has a paid version which makes this even funnier).

But as a consumer? Those fuckers are putting ads in the start menu and making it harder for me to remember what the GE proton update tool is called.

Most UBI solutions (which I very much support and voted for Yang in the primaries for…) tend to be built around the idea of providing cost of living for “free” but encouraging people to still suck capitalism good if they want more money on top of that. Which is “good” because it is how you get those rock star developers focused on major products.

But that more or less makes the same problem. Sure, there are going to be people who genuinely want for nothing more than three meals a day and spend the rest of their time doing hardcore development. But, even then, they likely are never going to be “challenged”. I’ve worked with some AMAZING developers over the years and have learned a lot from them. And I would hope they learned from me. Because, during a code review, you see how Nancy solved a problem and might try to incorporate that pattern into your own workflow and so forth.

But when you are more or less the sole “ninja” developer on a project and are mostly working with college kids who can remember what the various design patterns are called? You are likely not being challenged in the slightest and you “stagnate”.

And most people who live and breathe “awesome code” are doing so because it lets them do fun stuff on the weekend. Which, until we live in a post scarcity society, needs money/resources.

Hell, if I haven’t already pissed off more than enough people with this, I’ll add on that I have never met what I would consider a “good” software engineer who doesn’t “work for the weekend” as it were. Because if all you want to do with your entire life is code? You never stop iterating. You always want to make the code better and I need to regularly “check in” with you to make you push code to a repository or remove the WIP from your MR. Whereas the people who want to finish their job so they can go climbing or take a trip to the beach with their family or just blow money on hookers and blow? They are able to realize when something is “good enough for production” and they get a LOT more done.

The reality is that there is no one size fits all engine. Unreal and, to a much lesser extent, Unity come close, but they still favor very specific scales and styles of gameplay.

Like, for as dated as it feels, Bethesda’s fork of gamebrio or whatever it was is a REALLY solid engine for the kinds of games they make. Geometry interactions are still a bit funky, but it allows for massive scale and high fidelity because exterior regions are broken into cells in a way that favors (what we would now call) streaming. But if you want a game where you seamlessly go in and out of buildings (similar to the recent Yakuza/LAD games), it is laughable and, to my knowledge, still treats every building as its own world even as of Starfield.

And when you try to make one engine do EVERYTHING? you get star citizen where a refusal to do any form of load masking means that they need to be able to simulate space ships light seconds away from each other AND infantry centimeters away from each other and you basically see the physics engine explode every few seconds as a result.

Maybe you aren’t a fan of the in house engines. If it does one thing that “wows” then that is likely how they got their publisher

Also: it is obviously corporate backed, but Unreal Engine kind of is that. Sony has put >400M USD toward Epic/Unreal Engine and other studios/publishers put smaller amounts in.