Gabe Newell: Linux is a “Get-out-of-jail free pass for our industry” | Hacker News Reader

Gabe Newell: Linux is a “Get-out-of-jail free pass for our industry”

geekwire.com

124 pointsrandywatkins55 comments

I suppose this is why the FSF has gone on and on about "free as in freedom", and not about the practical benefits for developers in releasing source code to the public.

I understand that Gabe does not want to play by the app-store rule, be that Apple or Microsoft. They want to be free to develop their business, their business model, free from dictated rules of a third party.

In a world where every user device is locked down under DRM, there won't be a get-out-of-jail free pass called Linux. If all obvious and commonly used aspects of computing are patented, then we won't have a Linux that can be shipped by something like Steam. Anti-DRM and anti-patent clauses, like those in GPLv3 and the Apache license are not anti-business. They are pro-freedom for companies to decide their own fate.

If we want to keep this get-out-of-jail free pass for the industry, we need to maintain it and banish DRM and patents to the dustheap. It's that simple today. It won't be that simple tomorrow.

Minor nitpick: there are no anti-DRM clauses in the Apache license and the one in the GPLv3 only refers to DRM preventing replacing of the GPLv3 software.

Thanks. I included Apache license because FSF copied the text regarding patents from Apache and copy pasted it directly into GPLv3 (they later added a trailing sentence about patent agreements).

For all my love for Steam, Valve, Linux and Newell himself, I cannot help but gag a little every time he opens his mouth as of late.

Not a single fuck was given about Linux prior to the release of Win 8. Then, suddenly, once Valve's profit margins and business model is threatened the company shows some interest. This is all--of course--perfectly rational behavior, but there's no need to be all righteous about it and feed into our superiority complexes as Linux users. We don't want an 'us vs them' mentality here. What we want is for all the Windows users to get curious and come explore what Linux has to offer, not dig in and defend their OS just because they were pissed off by smug flamebait rethoric.

"Not a single fuck was given about Linux prior to the release of Win 8. Then, suddenly, once Valve's profit margins and business model is threatened the company shows some interest."

Well, I don't think so. It has been a continuous move for no dependence on MS for a long time. It is just that the backend code for making their code multiplattform is not as visible as their recent announcements.

If I have to point milestones, they are: Android as a commercial(linux) viable product, companies that make Linux compatible hardware (Xi3), and of course the tremendous amount of demand for mobile devices(that will surpass desktop computers installed base this year), not just MS fear.

Not really. Years ago people mentioned they found extra code for what looked like a linux version somewhere in some steam package or development kit or whatever it was. This was years and years ago. Valve downplayed it. Of course now it's obvious they've been planning and experimenting with linux this whole time. They've gotten higher frame rates in Left 4 Dead 2 on Linux than on Windows on the same hardware so, they have quiet a vested interest in this thing.

I think they're in this for the long run. The biggest threat to consoles is Steam and consoles after all, are just crippled PCs.

I don't care what his motivation is as long as he jumpstarts the movement of supporting AAA games on Linux, and getting GPU makers to make drivers for Linux, and game engine builders to support it as well.

Linux adoption is a chicken and egg problem right now. If Valve fixes the developer side of that, then the consumers will want it on their laptops, and manufacturers will start building such laptops, too. Well except Dell now, I guess.

Rumors of Valve's Linux-Based console have been floating around for years.

Phoronix have basically reported on Valve's work on "Steam for Linux" since 2008, usually backed up with some kind of evidence.

http://en.wikipedia.org/wiki/Phoronix

It is possible Windows 8/RT was the deciding factor in making Steam for Linux a product, but the interest in Linux is not a new thing.

Sorry, but that's simply not true. I think the real truth is that developing major applications / games on Linux is still sucktastic. No universal package management system, no universal set of dependencies, no guaranteed binary runtime environment, no unified platform API for the desktop, etc.

Windows developers would also like a debugger and IDE equivalent to Visual Studio.

The sad fact of the matter is that Linux isn't ready for the desktop, and I think Valve has decided to gamble that if they motivate developers to come to the platform, it will become ready.

The explosion of Android as a platform has shown quite clearly both the power and flexibility of the Linux kernel, and the usefulness of hiding the complexity and providing a platform-neutral API. The VM-backed programming environment is what Valve perceive as the new gaming market here: they don't want just a console, they want their own iDevice or Kindle Fire.

Mr. Newell mentioned that Valve's games were shifting to a free-to-play scheme. The Steam Box might be something like this: a Linux-based OS, some sort of runtime/VM, a game-writing API (Source-based API?) and an online, on-console marketplace. Whether the VM is Android or even based on Java at all would remain to be seen.

The basic Source (or successor) engine might be distributed built into the Steam Box, with expansions ranging from full-on games to amusing hats available for immediate purchase and automatic delivery. Probably some back-catalog goodies will ship with the box/be registered to your Steam account, too.

Would game developers really want to make games for a VM? Don't most big games usually try to squeeze as much out of the hardware as possible? (Genuinely curious... I know next to nothing about the gaming world.)

And what would the point be? Massive loss of performance by going through a VM - only to save 30 bucks on licence cost for e.g. Windows? Any game dev wants the most performance they can get their hands on.

It all depends on the type of game.

Not everyone does AAA games.

If the gatekeeper (apple/ms/whatever) decides to tighten the screws, taking not 30% of your revenue, but, say, 70% revenue - then fsck the performance, it may be quite reasonable to do major changes in art, gameplay and coding practices just to fit to a more lucrative market model.

And, as is well known, simply the existence and theoretical threat of a plan B (in Gabe's case, Linux) ensures a much better negotiation position and a better (more fair?) plan A.

Game developers are fundamentally human. They care about people playing games, nothing more nothing less. You don't need to necessarily make your game have the latest and greatest graphical glitz. If they can make their game run smoothly on your hardware, they won't care if it's in a VM inside a Virtual Box on a cloud server, under the sea.

Minecraft is written in Java. There were games written in (at least partially) Haskell/Erlang/Ruby/Lisp. Unity3D is based on C#, etc. If tradeoffs are worth it, they'll do it.

While developers surely are, game development companies are not fundamentally human. Now that Bobby Kotick has finished running Blizzard and the FPS genre into the ground, Valve is pretty much the only major publisher left that hasn't been overrun by IP lawyers and MBAs. Minecraft is a great game, but it's not a good representative of the average successful game. Developers make games because they want people to play them, but unfortunately, the only ones who have any control over what they are doing are the indie devs.

We already have indie / casual games on Linux. That's not enough to make Linux a legitimate platform.

That usually only applies to performance-intensive code like rendering, physics, pathfinding, etc.

The actual gameplay is often implemented using scripting. For example, with Unity you write games in C# (with Mono) using their engine, and it's still fast enough to work on cellphones.

There used to be a time when game developers didn't use "high-level" languages like C++ at all. But in the end they moved to them because of other benefits they got, and most of them slowly started to realize that with the hardware getting faster and faster, it's a good trade-off to make. They might think that again.

Android NDK. http://developer.android.com/tools/sdk/ndk/index.html

There are certain restrictions thought.

If you need to interact with Android, all APIs are done via JNI and certain audio APIs have only Java APIs available.

I'm not sure if you have used that to do any kind of intensive code. I've used it for real time video processing (H264 encoding/decoding and playing from a custom protocol) and it's not a good development experience in the least. It's extremely hard to get good debugging working, eclipse is a giant pile, most of the other IDE's for Android don't support C/C++ ndk, and there a mountain of little special compiler tricks and options to learn. Not to mention the JNI tricks and hidden issues. It's far from ideal. I REALLY!!! wish google would step up and provide a real solution to the problem there. Compared to how easy the same things is on iOS, Windows mobile 8, or bb10 is fairly frustrating.

They want to maximize performance and installed user base.

A huge installed user base like Android maximizes one variable so much that the other gets a free pass.

You are correct. The mainstream gaming industry would have absolutely no interest in developing for a VM, because developing games is kind of like a nuclear arms race.

Unless someone finds a way to get Next-Gen 3d Engines to run in a VM without a performance hit of even 1%, its not going to happen.

Some of your statements pretty much contradict what Gabe has been stating in talks and interviews the last few months.

I'm not particularly into video games, but I thought his whole talk about thinking of their games as a platform to enable their users commerce was very interesting.

I also now understand their decision to hire an in-house economist. They're looking at creating a much more sophisticated market.

I'm an IT pro,so I run Linux servers for my job, but I'm also a gamer, and dual booting is a pain in the ass. Trust me, I would love to make Linux my primary desktop, and every few months when I have some time, I give it another serious shot. But I always end up back to dual booting.

Years ago, I was happy with OSS audio system, but that is controlled by one company so the FOSSies have done their best to replace it with a far inferior ALSA. This caused me so many headaches back in the day and still bothers me with its high latency. But lately my issue has been with the graphics subsystem. For my X1900 I have to chose between open source drivers with terrible performance and bugs or proprietary drivers that don't work with modern kernels.

Finally, X windows and gaming don't seem to mix. Here is a case, where I think Linux needs to change the interface. I think Wayland may be the answer. If not, at least X11 with decent full screen support would be a godsend.

So from my perspective for Linux to make it, it would have to use :

1) OSSv4 Audio Subsystem (FOSSies it's GPL'd already get over it)

2) Stable graphics driver interface ( or FOSSies stop breaking ATI and Nvidia's drivers )

3) Modern Display Server - maybe Wayland or throwing out all the kruff in X11 and fixing Full Screen graphics in games

1) You can actually install it and use it. Personally I had some difficulties with some missing card-features, but other then that it worked fine.

2) Both NVidia/ATI and Foss is at fault: one should go open source, the other should should do as you say :)

3) X11 is deemed beyond fixable. But you can run Wayland today. (It has X11 server-for-wayland.) The reason I didn't try yet is because it won't run my WM. (Well as far as I know anyway.)

All good points. You're right, MS did remov hardware accelerated sound support in Vista, but that doesn't exuse Other OS's for lacking it. Back when I was composing and Mixing the "upgrade" to ALSA caused me a major headache, eventually leading me to hacking it out. Where windows has always just worked.

For the video drivers, I'm running Visa driver under Win7. The kernel dev's could take a cue from MS on this one. By breaking the proprietary drivers, they've ensured I use windows for gaming on this setup.

1)No need for that, ALSA works fine now. Windows sound system is not better than ALSA in latency terms and it works just right, they(MS) even removed some hardware acceleration for sound as they considered it was not worth the complexity pain.

2) This will come naturally if Linux becomes an option for gamers and 3D professionals(like architects and engineers). It is not that FOSS are breaking the drivers, but that Windows graphic drivers budget are tens or hundreds of times bigger than Linux.(e.g. nouveau drivers are made with less than 2 full time workers who have to reverse engineer everything, and Nvidia creates drivers with just their special customers[big studio animators] in mind).

3)This is basically dependent on 2. It is way easier to code a display server when your drivers work. They could only use open source drivers for this through, as companies like Nvidia had not collaborated with the Wayland people.

Do you write code that uses ALSA? I do. It is not "fine". It is maddening. By comparison, Core Audio is a dream. Windows is a bit of a zoo but all the pieces are there.

Replacing X isn't so dependent on drivers as you might think. A big part of what makes X so horrible is unrelated to how X interacts with the graphics card -- most of the nasty bits actually sit between the X server and the client application. The protocol was invented in 1987 and was designed around assumptions that just simply aren't true any more. These days, you want to shuffle pixmaps around or get a GL/DirectX context, and you want to do it locally. X is optimized for sending pixel operations over a network.

Let me make an analogy for X: imagine designing a protocol for making phone calls over an email system. That's what doing anything modern on X feels like, most of the time.

Agree on the ALSA comment. I wouldn't wish ALSA development tasks on my worst enemy. Developers these days should target PulseAudio, and tell users without it to get lost. The Pulse API is a poorly thought-out mess, but it at least behaves fairly consistently. ALSA's behavior is a complete crapshoot.

Strongly disagree on the ALSA comment. PulseAudio is the primary cause for most latency in Linux audio. I've written sound playback and recording code in ALSA and achieved low latency without difficulty. I wouldn't wish PulseAudio on my worst enemy.

I'm going to have to agree with this. I was the one complaining about ALSA to begin with, but PulseAudio is just another layer of indirection around ALSA. In other words, ALSA sucks, but it's more widespread than OSS, and it at least gives you some control.

What I found works for me is having Windows as primary installation and running Linux in the VM. Then I can just alt-tab between these two OSs as I desire.

Agreed. I take it a bit further: I have a dual-boot Windows/FreeBSD box, and use VirtualBox to run one OS inside the other, depending on what I'm doing (using a raw VMDK[1]). For me, this is helpful because using Windows as the primary gives me access to games (via working graphics drivers), and FreeBSD as a primary gives me access to a desktop environment that doesn't drive me crazy to work in (XMonad+xterm).

I'm probably risking huge dataloss, though.

[1] http://www.virtualbox.org/manual/ch09.html#idp13208480

We should all applaud the huge advance gaming will bring to Linux. Just the side effects from the engineering are huge—nevermind the enormous political and financial benefits that gaming brings to Linux.

Next up, naturally: Steam for Android.

They already have an app, now they just need to actually sell stuff with it.

They do sell stuff with it, just not for Android.

I'd love to see Valve bankroll some hefty Wine development and/or buy out Cedega/WineX/Transgaming, in my opinion that's the holy grail for what needs to be done.

Getting 99% of the games on Steam working out of the box with DirectX compatibility layer with in-house testing can and should be done, and is a better step than providing a VM layer right now.

I'd also like to see them contribute to OpenGL and get it up to DirectX standards in terms of API ease of use and maturity. It simply isn't as good, and until it is developers will continue to target DirectX.

While it's all well and good that Linux is getting more press in regards to gaming these days, I'm still not seeing a massive push on the game developer side. Yes, there are the indie developers who will jump on it, especially those developing in Unity since it deploys to Linux and the others natively. But look how long it takes the major developers to even publish on OSX - and that has a much bigger (and much better known for customers willing to pay extra) market share.

Personally I'm working on all three platforms and I have reasons (and personal preferences) for each. I want my web servers to run Linux or BSD. I want my mobile computer to run OSX. And I want my work machine to run Windows. In each cases it's a mix of a need for certain applications and a preference for a certain workflow. Sometimes it's even something ridiculous like hating (HATING!) the way mouse acceleration works in OSX....

So, anyway, the way Gabe Newell tries to force an OS-war with his rhetoric is getting on my nerves. Steam is making him shitloads of money. And it'll continue to do so, regardless of Windows 8. And I love that he's opening up the market for Linux, but let's face it, Linux still isn't quite there as an end user desktop system. And unless it is, why should people switch and start purchasing their games on Linux, when they still do their everyday stuff in Win and OSX.

This is a hen-and-egg-problem. No games for Linux are published, hence no market exists. And since no market exists, no games are published.

However, the Humble Bundles strongly suggest that there is in fact a market for Linux gaming, considering the Linux users consistently paid the most for the bundles[1], often significantly more than the Windows users and more than even OS X users usually perceived to readily pay a premium.

[1]: http://cheesetalks.twolofbees.com/humble/

And still the Linux ports on the Humble Bundle are buggy and painful to play (some examples here: https://bbs.archlinux.org/viewtopic.php?id=154932)

That makes it even more interesting, doesn't it? They are willing to pay for having something at all, even if it is somewhat bug-ridden or not quite as fast or not quite as perfect as it could be.

That should emphasise even more that it might be worth pouring additional resources into development. Whether it will be done is another question, steered by corporate politics, risk-assessment etc. etc.

I have heard a number of times about how great Gabe is for bringing Steam to Linux. Isn't it just the case that the only reason Steam is on Linux is because Valve are making a console? Would Steam for Linux really exist without this console?

Windows 8 is a fine gaming platform and the Windows store is... useless. It brings crappy tablet games to the PC. I think he is over reacting to the damage Windows 8 will do to his business. I think his motivation towards Linux only came into existence when he saw the Windows Licensing cost on a console.

> I think he is over reacting to the damage Windows 8 will do to his business.

He's not worried about everyone buying Windows 8, he's worried about no one buying it:

http://www.xbitlabs.com/news/other/display/20130204235238_Gl...

Windows 8 isn't an existential threat to Valve because there are some UI tweaks, it's a threat because Valve is tethered to Microsoft, and for some, Windows 8 has not inspired much confidence.

Thankfully, we don't have to guess about this particular metric, or use silly results based on web-surfing of certain websites. Almost 9% of Steam users are already on Windows 8:

http://store.steampowered.com/hwsurvey#cat0

8.76% is much higher than 2.26%.

The salient point isn't "Win 8 adoption is x%."

> silly results

Please don't do this.

For your AAA full-fledged games to work on Linux, it's not just Valve - the Big-iron game engines such as RAGE, CryENGINE, FrostBite 2, etc also need to be ported and supported. Only then will AAA game makers be able to support Linux games.

An open platform with the accessibility and raw power of a living room console can only lead to bold new forms of gameplay elements both in hardware and the games. A common argument made in favour of strongly controlled console markets is that it ensures quality, which i don't understand. I think good content will always find its way to the players no matter how crowded the market is with crappy titles. And Steam is not abandoning it's own storefront.

I feel like the title of this article was a little misleading, it makes it sound like he's chiding the industry for relying on use of linux and open-source software to make them look good (android is linux-based? yay it must be the best!), which I would kind of agree with. He's actually saying that it's there as a fail-safe for the possibility of Microsoft and Apple walling off their ecosystems from tinkerers and developers (which I also agree with).

Isn't the Sony PS3 operating system largely based on Linux?

This is much more about what you let customers do with machines. There's a tension between "anything goes" (and rampant piracy and cheating) and total lockdown. Big companies seem to go the route of sealing everything up. It's unclear that this is totally necessary.

AFAIK no, but it did have an option to install a gimped Linux OS for some time, before Sony took that away.

Nope, it's a custom OS. So is the Wii (U)'s or xbox's; they might use some BSD's parts (network stack ?) but I'm not sure.

See http://en.wikipedia.org/wiki/OtherOS to clear the confusion.

BSD wikipedia says. I thought it was linux too!

>>The native operating system of the PlayStation 3 is called CellOS, which is believed to be a branch from the FreeBSD project

Reply on news.ycombinator.com