"Locked Boot-loader..."

I am hoping that scope SteamOS grows far beyond its current goal, and it becomes the general "Windows-compatible flavor of Linux" which can seamlessly run any Windows application - gaming or non-gaming. I know a lot of people who only use Windows only to run a few Windows-only utilities (usually business applications) but would prefer to run Linux instead. If SteamOS could achieve complete compatibility for Windows applications in general, then I think it would start to break Microsoft's stranglehold on the OS market because "Linux can't run [insert Windows program]" won't be an issue anymore for people hoping to switch from Windows. Maybe this is just me wishing for more from SteamOS than we're actually going to get, but Valve seems to be determined to make advances in Windows to Linux compatibility tools so it will be interesting to see how far they go with things.

I will defer to your Windows knowledge in this matter, you know more than I do (I stopped using Windows before 2010).

Well, all OSes transition somewhere. I can't talk about recent history, but I can tell you that transitioning games from Windows 95 to Windows 98 to Windows 2000 to Windows XP wasn't exactly smooth sailing. I can't say I was impressed.

Ubuntu took a misstep with Unity, but there was a purpose behind it: They wanted a unified GUI for desktop and mobile. Once that didn't pan out, they moved back to tools with more community behind them. Going back was distruptive, it was the right long term decision.

Depends what you are talking about.

A lot of the system level software and bedrock tooling is rock-solid and has a strong community behind it (that is not tied to the fate/decision of a single company).

You are also fully in control about what to update and when. And If you have some resources and are desperate, you can fork.

You can take any version of software that is in the open ecosystem and maintain it if you have to. Legally, you cannot do that with Windows. Providing support for a legacy version of Windows is a non-starter. You simply cannot touch the software.

But yes, the most recent version for a lot of projects move fast. If you are not comfortable being at the forefront, don't update right away?

Admitedly, that works better on servers than end-user desktop software who want things to work with the latest version off Ubuntu/Mint/Pop OS/whatnot. In that layer, the Linux world is still fragmened (and yet, has more hope of bridging a universal standard than closed commercial platforms that will simply not legally allow it... their idea of universal standard is world domination).

Yes, we live in a culture where everyone can Google stuff up and that somehow makes them an expert.

How dare do people with years of experience in a field dare to claim that their hard work and dedication gives them more insight in their field of expertize. How arrogant.

Or maybe the real arrogance is assuming that knowledge is flat, that everybody can know anything without investing any time and that somehow, your knowledge of a particular topic is on equal footing with someone who dedicated a lot more of their time at it than you (the general you, not *you* you).

Or you know, we could just accept that we are not perfect, that we will all ruffle each others feathers a little and short of being downright insulting (ie, calling each other idiots and whatnot), maybe we should gracefully let a few minor details slide and other people will similarly return the favor.

Just a crazy idea to get along better.

I disagree there. I think were Linux really err is in not catering enough to non-technical end-users.

It doesn't cater enough to developers in the areas where they themselves are non-technical end-users.

I worked on software for the PSP years back and I can tell you, I did not enjoy working with their "development kit". I kind of wish they had made it closer to developing for windows, like Microsoft did for the Xbox (that was circa 2007 I believe).

They also didn't have full software emulation for their device which meant I had to get up from my station and test on a PSP regularly which was annoying. A few years before that, I developed for some mobile phones with full emulation and the development velocity was a lot faster.

Admittedly, at some point, you do need to test on the actual device, but developers work on a PC, not a console so the more you allow them to do things on their PC without context-switching, the better.

There is an important caviat here: Windows and Xbox are both from Microsoft. As I said, very cohesive experience, very nice, but as a dev, I'm not interesting in investing my valuable time in proprietary walled gardens.

I want to be able to cater to as many devices as possible with as little work as possible.

The closest we've come to this are web browsers, which rely on standards, across corporations.

Just because it doesn't have the community on the desktop. That's it.

Its VERY profitable and dominates in most other domains of software development.

From an economical perspective, it makes a lot of sense to pool your efforts globally together with open-source.

Just solve the darn problem already for everyone, free of charge and let the world move on to bigger and better things.

Everybody needs a kernel. You can rely on a kernel from a specific company that is proprietary (and have to ask for their blessing for everything you want to do with that kernel) or you can pull your resources together world-wide and have an open kernel that anybody can adapt to their needs (and surprise, that kernel dominates, because anybody can take it and adapt it without asking permission).

Similarly, everybody needs a programming language (which tend to follow the open-source model more often than not), most people needs a web server, and etc, etc.

The only place where the proprietary incentive really makes sense is at the very top layer: The application software which is the real business differentiatior, which is not generic and which anybody can't just reuse for their project.

All the building blocs underneat that should be open and widely available so that developers can focus the bulk of their resources targeting end users: Not giving away 30% of their profit to the apple Platform, not ripping their hair out because Microsoft stopped providing support for some old version of Windows and they can no longer legally support it without Microsoft's consent and certainly not focusing tons of energy becoming expert in a proprietary platform that will ultimately only target part of your userbase (I'm looking at all the consoles out there, though I'll have to partially exclude Nintendo, because they added some genuine value with their innovative controllers).

I'm not interested in being a Microsoft, Apple, Sony of whatever corp float your boat dev. I work to maximize the difference I'm making in people's lives and I'm dedicated to providing the best experience with the software I write to as many people as I can. This is what software development should be about.

Both of you, put down the monkey’s paw. If there’s one company I trust even less than Microsoft, it’s Valve. If Valve ever improve Steam compatibly to a decent level the first thing they’ll do it make a Steam client/ OS exclusive. That’s even worse than what we have now.

I trust Valve, Microsoft, AWS, Apple, practically every console company out there and a whole bunch of other companies in the technology sphere to do a single thing: Make whatever decision will help both maximize and secure their future profits, with the least amount of effort possible on their end (a rentier setup, where they can milk money just by leveraging their complete control over the captive ecosystem they setup would be the ultimate desirable goal for them).

Ideally, that means exclusive global dominance (both technical and legal) over as big of a chunk of the ecosystem as they can manage.

Failing that, they'll grudgingly play along with others given that compatibility with third-party solutions will be expected to do business (ex: the whole compatibility layer Windows has with Linux... given how dominant Linux is for servers, they had no choice).

The only counter to that is a vibrant open-source ecosystem and in some cases, government intervention to foster competition.

With all this talk about how "great" windows is at backwards compatibility, this was my first thought as well. It was terrible.

You might get Windows for free, but it certainly isn't free!


That's rich, coming from you. You're not willing to use WINE because you can't follow the steps at WineHQ, but anyone should be able to follow the steps to install Linux on a phone?

You've got wicked bad luck with wine. For me, games that don't work are the exception, not the rule. It's pretty rare these days that I have to hit WineHQ at all.

maybe , ive only tested a few games so far and they just ran not as good as on win
fe mortal online 2 had some graphics anomalies which made it impossible to see the character
or fallout 76 ran like -20 fps while still looking not as good as on win

While I will acknowledge that is a possible action that Valve could take, it seems like a highly unlikely thing for them to do for a three reasons:

1. One of the largest drivers of Steam's success has been their focus on maintaining a large audience by increasing the accessibility of digital games. So the idea of them suddenly restricting Steam client support (and thus their audience) to only their own obscure build of Linux (which has a small market share already) seems like a step in the wrong direction with regard to accessibility. I would doubt that many Steam users would be willing to replace their entire OS just to play games on Steam when there's alternatives out there that don't require their own special OS. So dropping support for all OSes except for SteamOS wouldn't make sense financially because they would be alienating a large portion of their audience. Dropping Windows support in particular would be financially catastrophic given the market share of Windows.

2. The entire point of Proton is that it is a Windows-to-Linux compatibility layer which makes it look like they don't seem to want to abandon Windows but instead work with it.

3. A lot of Valve's work on SteamOS is open-source. The base Linux distro is Arch (which is GPL'd), and Proton is a frontend for Wine (which is LGPL'd). With this in mind, anyone can fork Valve's open-source Proton work and make a version that works on Linux distros that are not SteamOS. So I don't see the risk there either because Valve has already released their compatibility "secrets" to the public.

So can Valve do what you're describing? Technically, yes. However, I can't see locking Steam down to SteamOS as a smart move given Valve's own actions that make such a decision pointless and even harmful to themselves.

You’re kidding right? We’re talking about gamers here. These are the same people who through money at companies despite repeatedly getting screwed over by them. If Valves increases their dominates high enough they will have no problem migrating their user base. Linux people are always talking about the threat of Microsoft, but it’s really Valve you should be afraid of.

Unless they manage to take that market share away from Microsoft which I think is part of the plan.

‘compatibility layer’ is just code for WINE. If Proton is using WINE it stands the reason it’s going to have the same problems. Such as not working very well with gog games. I can’t wait for the day someone makes a proper emulator for Windows instead of this half backed half measure.

Why the hell are they using Arch? And not Debian or Ubtuntu. Also doesn’t Steam OS use APT? Anyway that doesn’t explain why Proton doesn’t work with GOG games.

If they do it now yes, if they achieve full market dominance, the next move will be to go to full on monopoly. The thing with monopolies is that in the end, every company wants to be one.

Once again, I understand the scenario that you are describing, but it assumes that SteamOS can somehow establish enough market dominance over Microsoft that people start developing games for SteamOS and not Windows. This is something that I don't believe is possible - especially considering that Valve is primarily focused on video games which only a portion of the overall Windows market. (Though I would like to see a dent made in Microsoft's dominant position.) Otherwise, people will continue to play their Windows games on the copy of Windows that they probably already have. So I'll have to disagree based on that point.

I don't know specifically why SteamOS is using Arch, but they appear to have switched to it from Debian starting with v3.0 in order to "make development easier" or something. Arch uses pacman as its package manager instead of APT, but since SteamOS was Debian-based prior to v3.0, then it most definitely used APT.

Regardless, my key interest is Valve's open-source work that they're doing in the area of making Windows compatibility more seamless and how that work can benefit Linux in general.

Wait. Why wouldn't Proton work with GOG games? What's so special about GOG games? Are you talking about having Wine integrated in the client for ease, i.e. in Galaxy? Cause that's not that convenient at the end of the day, if you know what you're doing with WIne.

Also, you can get custom proton/wine versions to work outside of Steam. You lose some nice shenanigans like Valve's Fossilize vulkan shader pre-compilation, but it's doable if wine-staging or a custom wine-staging like TkG's isn't enough for some reason.

I don't know exactly what problems with which games you had, and this is not an attempt to troubleshoot stuff, especially here, but I think you'll find there's a ridiculous amount of support from Linux gamers encountering similar issues. ANd through some trial, error, perseverance and optionally, diligent bug reports so stuff gets fixed, chances are you can get your games working fine. I still haven't encountered a game I couldn't play on Linux in the past year, with or without some tinkering. Increasingly more without tinkering, mostly thanks to Valve and their push for compatibility due to Steam Deck.