Is Free And Open Source Software The Best We Can Do?

We've been with both Open Source and Free Software for over 25 years and much longer in the case of Free Software and as can been seen from the modern software landscape companies are abusing these systems so is it possible that something better can be done.

==========Support The Channel==========

==========Resources==========

=========Video Platforms==========

==========Social Media==========

==========Credits==========
🎨 Channel Art:
Profile Picture:

#OpenSource #FOSS #Linux #FreeSoftware

🎵 Ending music
Track: Debris & Jonth - Game Time [NCS Release]
Music provided by NoCopyrightSounds.

DISCLOSURE: Wherever possible I use referral links, which means if you click one of the links in this video or description and make a purchase I may receive a small commission or other compensation.

[tauiin]

I do agree that free software & especially open source has failed in some sense, but to fix it would require more.... social and political change than any single licence would be able to fix.

[kxuydhj]

this looks like something between an actually great idea and the "now there's n+1 competing standards" meme.

[joruffin]

Let's not go the way of music rights payments. That hasn't worked out well for anyone other than the distribution organisations.

[XH]

After an open source software, you have an open stream software, then open river software, and finally open ocean software.
Then, with evaporation, which can take million of years it becomes open cloud software, aka AGPL, then open rain software and finally the cycle is complete

[Crnobyl]

What I like about open source software is the fact that someone dedicated enough could port software to work on other machines.

Open source software is what really makes new ISAs like RISC-V viable to begin with. Without Linux, BSD and the open source ecosystem, RISC-V would be useless as what good is an ISA without software to run on it.

[KiraSlith]

Yeah, I've got an education in economics and the funding model he describes will result in a far more finite funding pool, and the bulk of that pool being moved to this theoretical entity, basically turning FOSS into a zero-sum game by trying to make companies engage with funding through a singular entry point. I guarantee it's going to result in a net reduction in funding across the board and drag corporations into funding and using a central set of tools approved for funding by said entity. He's basically making a decentralized corporation out of the concept of free software, one without programmer rights, and without technically "hiring" anyone, a "Web 3.0" Microsoft for Linux software. This is outright dystopian.

[gpturismo]

Contracts will have problems as well. When you have a behemoth like IBM, who can keep things in courts even with appeals, hire the top lawyers etc. We talked about this a lot in the 2000s with SCO, SCO becoming SCO Group, Novell buying SuSE and the collapse of many large Nix corps like SGi. I remember when IBM bought Red Hat and we warned they would restrict things and no one listened.

[wp]

Free software is also designed to be friendly to corporate interest, they just want corporations to use foss software

[bruderdasisteinschwerermangel]

I think an interesting case is also Prusa with their 3D printers, which are a complete anomaly when it comes industries that design and manufacteur actual hardware.
Not only is their software stack completely open source (Marlin, PrusaSlicer), the entire hardware is as well. The founder and CEO even has a Open-Source-Hardware logo tattoo on his arm.
But they're in a rough spot now. Their competition is greatly benefiting from all the development they do and I'm not sure if their business model is that sustainable anymore. Most notabe is Bambu Lab, a Chinese manufacteurer that heavily uses software (and I'd argue general hardware improvements), which was funded by Prusa. I'm curious as to how they'll continue, I suspect they will be forced to introduce some kind of change to their licensing just to stay competetive and profitable. There's also a blog post "The state of open-source in 3D printing in 2023" from them about this topic, can recommend.

[JeffRyman]

I believe IBM is violating the GPL, but we need someone with very deep pockets to take IBM to court and hang on through all the appeals.

[guss]

I'm not loving the "post open" barrier to entry. Most of these like minded thinkers are considering "people at home" vs "huge corporations with tons of money". I have a small company that builds and provides software - and like all modern software companies, 90% of the code we distribute is open source. My company employs 4 people (including myself) and we don't have money or time to spare - if a software component we want to use has a license that requires us to jump through hoops such as compliance monitoring - we can't do that so we'll find something else.

[pandapip]

I'm not opposed to software you have to pay for, as long as:

- The cost is a one-time fee for a perpetual, irrevokable license
- Updates can be paid as long as the versions you have access to remain supported
- You have access to the source code of all versions covered by the license
- No DRM, no license codes, etc. The software must assume it's licensed.
- If you're not making money from it, you don't have to pay

Probably missed a couple of things. I'm fine with paying for software, but if I buy sofware, I want to own it. Basically, the WinRAR model.

EDIT: Don't read my list of points as a list of requirements to nitpick. I recognize the wording is flawed (particularly point 2). So I ask: please don't point out loopholes. There are a lot of them. What I would prefer to see discussed are alternative wordings that better represent the underlying idea and/or the underlying idea itself.

[tgheretford]

The arguments that apply to companies paying their way in a "post-open" architecture also applies to individuals and non-profits. I know there is a desire in some quarters to monetise free and open source software beyond voluntary donations. And that doesn't just apply to software, it's just as applicable to YouTube content too. The problem is, there isn't an infinite money supply for creators from consumers and charging individuals and non-profits as well as corporations for everything that provides value to them is going to get very expensive, very quickly.

[pvb]

Creating this type of "post-open" organization is just asking to create a monster that'll create more problems than it solves. There's no reason to believe that this kind of single foundation or corporate entity that would compensate authors of post-open software would act in the interest of those authors if it reached critical mass. And also no reason to think developers wouldn't put up with it anyway because of a lack of options. People develop software for Apple platforms, Android, publish games on Steam, etc, even though the massive cuts those platforms take and the degree of control these platforms have on developer freedom.

[PygmySurfer]

I see a lot of issues with this, but I think the biggest one is the non-profit that's going to oversee this - all of the money that gets collected that's supposed to go to developers is going to have to go to the non-profit, so they can audit these companies and bill them.

Re: AI, Bruce is spot on. OpenAI even admitted as much (Ars Technica: "OpenAI says it’s 'impossible' to create useful AI models without copyrighted material")

[paherbst]

What rhrl is doing isn't a loophole. It's explicitly allowed in the GPL

[brandishwar]

So, in short, he wants... non-profit software companies... Okay...

[jeremyandrews]

This might be one of my favorite videos from you so far. I feel like it touches on something that bothers me about the Linux community (and FOSS in general, probably) that I couldn't quite put my finger on. What FOSS seems to have done, is incentivized companies to make their money on services rather than software, and in general software quality and diversity seems to have plummeted, because now all the money is in the services built on top of the software, while the software itself is just a worthless commodity to be used as a vehicle for pushing someone's service. What FOSS has done, perhaps unintentionally, is made software by itself worthless and forced everyone to make money on services instead, leading to poor code quality and software that is increasingly crippled without a support subscription or cloud infrastructure of some kind. Granted, FOSS isn't the only reason for this trend, but it definitely seems to have helped accelerate this transition. Another big downside of FOSS is that, like you said, it's designed in such a way that it serves the interests of developers and not end-users. It seems like it promotes a kind of creative destruction that you didn't see with the previous model, a creative destruction that most negatively impacts those with the least technical skills and money to adapt to changing trends and who are most likely to dependent on a past investment. Since I've criticized Wayland to death in other comments, let me use a different example for this one... Chromium.

Back when Firefox was competing with Internet Explorer 6, the main problem wasn't that Firefox was having trouble maintaining feature parity with IE6. It was that because all websites were designed around the limitations of IE6, no one could tell that Firefox was better or had a lot more to offer, was more standards-compliant, etc. IE6 looked good mostly because it was the default and everything was designed around it. But if there was one good thing about IE, it was that it set a "floor" for Web features a browser had to support that was fairly slow-moving and made it possible for multiple browser engines to exist that met at least that standard. Now, let's look at what Google has done with Chromium. Instead of killing off competitors by holding the web ecosystem back like Microsoft did, they kill off competitors with a firehose of stuff they can't reasonably keep up with. They help to aggressively push new web standards that make older or different engines increasingly useless without a huge commitment of money and resources, and there is so much code churn in Chromium that even if it is technically open source, maintaining a hard fork of it for longer than a couple of years without a rebase is nearly impossible. So, yes it's technically open source. No, it doesn't matter... you can't possibly build on Chromium fast enough to make a legitimate alternative unless you're huge. What you can do, though, is if you are a company inject your own services into the codebase to make money, as long as you don't try to do anything significantly different from Google. Therefore, they can effectively have a browser engine monopoly, without anyone being able to claim they have a browser monopoly. But then... think about stuff like Widevine, think about Google services, think about all the things that they don't have to make open source if they don't want to, and how they could eventually tighten the noose enough to pull a RHEL-like move after Firefox is finally gone and enough competing browsers exist that they can claim to be competing fairly.

I feel like the Chromium situation is a lot like how Linux/FOSS as a whole tends to be. Just constant churn that serves corporate interests primarily, and which makes open source kind of useless to anyone who genuinely wants to fork off and do their own thing with the code. You just can't... it's a big interdependent, interconnected mess where all the parts of it are moving too fast for any individual developer to realistically do their own thing, at best being able to "slow down" a move towards a trend/technology they don't like, while eventually having to embrace it themselves just to stay alive. And much like in the browser space with Chromium clones that don't change much, RHEL clones or Debian clones with lots of value-added services and a bit of extra branding on top that change as little as possible and do nothing different, are usually the most successful projects. What was envisioned was a bunch of individual developers doing their own thing and taking projects in different directions. What we wound up with was companies turning various pieces of software into something like industry standards that they can all use to create their own products and set ground rules. Linux is an industry standard for Unix kernels. RHEL is an industry standard for desktop Linux. Chromium is an industry standard for web browsers, etc. When you start to look at it that way, you realize that we've moved from a world where software is valuable and somewhat diverse, to one where it is essentially a spec for how a service can be provided that it benefits corporations to collaborate on and not allow much deviation from.

[filipDcve]

A license that is hard copy-left by default and looks closer to today's FOSS licenses for paying customers could be a natural way to distinguish between corporations and other types of users

[Georey]

When you are reading articles, please highlight the line of the paragraph you are reading or something. I find myself getting lost in what you are reading multiple times in each video you read articles. Thanks! Love your content.