They'll probably all be running Linux. Torvalds always said Linux will not succeed on the desktop until it comes preinstalled. 10% here we come!
Torvalds is correct!
The only reason why Windows has been so successful, is because every computer comes pre-installed with Windows, unless its a MAC. Believe it or not, there are still a lot of people out there, who still have never heard of Linux, cause they never used Linux a day in their life.
Microsoft and Apple rigged the market, so that they would always be first, in the consideration of computing. Now, I don't know anything about thees Risc=V computers, whatever that means. The last major leap in computing I ever heard of was quantum computers.
But since quantum computer's are so big, they take up the room, this is why they are still only used by major corporations, and they are used for speeding up computational tasks, that take normal computers, much longer to do.
Unless quantum computers can be downsized, to the size of a full tower desktop, I don't see them in the home anytime soon. Also, were reaching a limit to the X86 current era platform. We can't keep pushing frequency anymore, as Intel had to learn the hard way, in the P-4 era, circa early 2000's.
So instead, they make CPU's with many more cores and threads, so the CPU can split the individual tasks to cores, so that we don't experience computer lag, like we used to in the pre-multicore chips.
RISC-V Security Issues
RISC-V, an open-source instruction set architecture, has been designed with security in mind, but it still faces potential security issues. Here are some key points regarding security concerns with RISC-V:
- Open-Source Nature: While open-source architectures like RISC-V can be more transparent and allow for community-driven improvements, they also introduce risks such as vulnerabilities being intentionally introduced by malicious actors or unintentionally by contributors. Verification and due diligence are crucial when using open-source components.
- Third-Party IP Integration: Integrating third-party intellectual property (IP) into RISC-V designs can introduce security risks, especially if the IP is not fully mature or well-tested. This is more challenging compared to established commercial IP like Arm cores, which have been integrated into SoCs for many years.
- Security Features Implementation: The security features of RISC-V, such as 4-ring trust, secure interrupt processing, and physical memory protection (PMP), need to be properly implemented in hardware. Implementation flaws can lead to security vulnerabilities.
- Formal Verification: Formal verification of security properties in RISC-V processors is essential to detect and remove design bugs, which can be exploited by attackers. This process helps in hardening the hardware against malicious attacks.
- Secure Boot and Storage: Ensuring a secure boot process and secure storage of keys are critical for protecting against unauthorized software execution and data breaches. However, these features must be robustly implemented to be effective.
- Debugging Security: Debugging units in RISC-V processors need to be secured to prevent attacks that exploit debugging interfaces. This includes implementing lightweight, firmware-controlled security measures.
- Security Analysis and Mitigation: Continuous security analysis and the development of remediation strategies are necessary to address emerging threats. This involves identifying areas of risk and working with RISC-V working groups to mitigate these issues.
- Community Contributions: The RISC-V community can contribute to improving security by reporting and investigating potential issues. This collaborative approach can help in enhancing the overall security of RISC-V implementations.
While RISC-V offers a solid foundation for security, ongoing efforts are required to address and mitigate potential security issues.
So as in all Systems, human design can still have flaws, just like other systems.
In terms of high end Intel Processors from Intel, a Portable Appliance test operative told me he once had to do testing of appliances at an AMD factory in Manchester. An operator there was spending all day slicing through Intel Processors that would never see a motherboard!
Intel chips being diced at an AMD factory makes some sense, or was that a typo?
No, not a typo. I was gobsmacked when he told me. He asked the operator if he could have one intact one but he was not allowed.
Of course not, this is human society we are talking about, the most wasteful species on the planet. Let you have an intact one? Nope, thats against regulations. But lets tear chips apart, then slate them for recycling.
I still remember when I used to do volunteer work, for a local brick and mortar, resale store. People would bring their old computer's in, and pay the resale store, to recycle them.
Due to the loophole of regs, I wasn't allowed to take any home with me, cause people paid to get them recycled. I even said, if it was about data, we could leave the hard drives behind to get recycled, nope, still nogo.
Most people haven't ever seen a full sized recycling center, or full sized garbage company dump pile yards, to even get a clue, as to just how wasteful our species really is.
IMHO, we are far far far, behind StarTrek, and their level of recycling technologies. I have little faith in our species to do better, as we currently stand. But that whole thing of, hack CPU chips up, for the job, wasting every single one of them, I am not a fan of that.
And before you say, well, they got to hack them up to see them inside. No they don't! We have developed high tech scanning devices, that started with X-rays, now we got mass spectrometers, and lots of ways to see inside a chip, without needing to destroy it.
Were just a wasteful species, pure and simple. When oh when, will we get out of the dark ages? Anyways, trying to get back on topic, RISC-V sounds interesting, but security wise, sounds like an utter nightmare. I won't be using it!
I'm gonna lay down some bitter truth. My shield is up; feel free to hate me.
I strongly believe macOS and Windows still provide the best 'customer support' for your average computer user (who doesn't know how to properly Google, doesn't know how to troubleshoot WiFi issues, doesn't know how to set up a printer, and doesn't know what an operating system is).
I expect some pushback on my comment, but all of us here, to some degree, enjoy tinkering with computers that most average consumers don't. They just want to get work done. If someone buys a new Mac or PC, they have dedicated support from Apple or Microsoft. (How good that support is is a topic for another day). But they know where to go to get help.
The overwhelming majority of your average computer user doesn't know how to install Linux or fix most common issues that still happen today.
Linux, for the most part, is a community thing. And not everyone has the time, willingness, or the curiosity to fix computer things and get help from the community. Most people want to get stuff done, close their Excel sheet and go do their thing outside of their computer.
This is the biggest blocker in Linux's desktop marketshare.
Torvalds is right, but it needs to go one step further. Now that gaming on Linux is a thing, manufacturers like Asus, Lenovo, Dell, HP, and others need to come up with their own version of Linux-based distro (or back one of the existing ones, whichever that may be) and offer their full support in addition to Windows.
Far too many times I've been disappointed by a cool laptop when I thought, "Yeah but it'll run Windows."
You can install Linux on any laptop, but the moment you run into WiFi, sound, display, biometrics and other issues, most people would just go back to Windows (and it makes sense why they would).
Imagine a future where all laptop manufacturers have their own distro (or certify full support for any existing distro). If something doesn't work, customers can turn to the manufacturer for Linux-related support. That's the day Linux desktop marketshare will truly begin to skyrocket.
Valve has: SteamOS. It's based on Arch Linux, and currently only supports the Steam Deck and an upcoming third party handheld (I forgot which). One of the biggest reasons for the limited support is that drivers aren't where Valve feels they need to be from Intel and Nvidia, but your point stands and is correct: anyone serious about a broadly adopted Linux for people who want to grab their system and do stuff needs a Linux designed for that. Steam Decks stay easy to use because by default, you only interact with Steam, not the OS. (You can absolutely switch to desktop mode, in which case you get Plasma 5, but from there you're on your own. Still won't run into hardware issues because Valve knows exactly what's in every Steam Deck, and provides the drivers.)
The problem there is that maintaining that level of user friendliness also means locking down the experience a lot. SteamOS runs on very specific hardware only and is immutable (that is, you can't alter the system itself, even with admin rights). It does allow you to turn off that immutability, but if you change things and it breaks, it's on you.
You and I are in agreement there. If someone's PC use is completely task oriented and they would rather trade flexibility for getting things done and logging off, Linux isn't the best fit.
Macs tell you you're doing it this way, you will and must; options are limited. Windows is slightly more flexible and much less of a walled garden, but for basic functionality, every Windows PC working the same basic way means people can sit down, do the thing, and get up. In fact, relatively minor changes, like moving the start button, irritate people. Then you have Linux. At the operating system level, GNU/Linux generally does all work one way, regardless of distro. You've got multiple filesystems, but they're all going to present the same directory structure because that's how Linux works. The kernel will go in the same place, user directories go in the same place, etc. But the user experience is wildly different by desktop environment, much less distribution, and that's before you get into fixing actual problems, like wifi misbehaving.
I'm increasingly convinced that there will not be a "year of the Linux desktop" unless SteamOS gets its broad hardware release and people start using it for more than just games. Ubuntu made broad strides in approachability, but it seems to have run out of momentum. Linux Mint is now widely considered a good starting point for Windows users, but I don't see it scratching at any significant chunk of Microsoft's install base. The main reason I think SteamOS could is sadly the same reason Windows and Mac are quick and comfortable for most people: a major corporation with money to throw at problems defines it, provides a single, one-size-fits-all experience, and a reason to switch. (Presumably. Running better on lower end PCs, not hurting framerates with Windows' drek, etc.)
The thing is, that'll create SteamOS users more than "Linux" users. That's still good for us, in the way that Valve's Proton has been good for us, and advanced users likely will break the immutability and experiment, but what makes for adoption by those who just want to accomplish a task isn't power, or customization; it's familiarity, and that comes from a dictated user experience. The "Linux desktop" will grow in that the Steam desktop arrived, not wide adoption of Linux the way it looks today.
Here I disagree unless they all use the same desktop environment and can run Microsoft Office (so they can work in a familiar environment). Have you tried getting Windows support from Microsoft directly? Corporate customers can get a lot of help from them, but end users are largely stuck with forums filled with other end users and occasional volunteers, not so different from this.
Right! I have a Steam Deck, and I used it as my primary (non-work) computer for weeks without any issues. If I wanted, I could work off of it, too. (I used the USB-C to my display that has a USB mouse and keyboard plugged in, so it was pretty much plug-and-play.)
I agree that some lockdowns need to happen, but I don't really see any issues in it. Most people who don't want things locked down are already not in the target audience (like you and I). The target audience, the average Joe, so to speak, doesn't care about it.
Valve's limitation aside, it goes to show that it's possible if manufacturers really want to. I feel like in many years, more companies will realize Windows is a bottleneck, and they might start doing something similar.
I believe this, too, for the sole reason there are so many distros. If you look at the YouTube comment section arguments, people are always divided on topics like Mac vs. Windows, iPhone vs. Android, and Windows vs. Linux.
Even if you happen to side with Linux, the "fight" doesn't end there. Then we have Ubuntu vs. Linux Mint. Fedora vs. Ubuntu. Do you think the argument stops the moment you choose a distro? Not really. You go with Fedora, now there's argument about which spin is better. Gnome vs. KDE. On Mint, Cinnamon vs.... you get the idea.
Choices are good, but too many choices will always confuse the average guy. (Dell got rid of the XPS and other lines in order to make it easier for people to choose. I don't think their new lineup is any less confusing, but their research surely proves the point.)
All that is to say, Linux is way too fragmented, but on the plus side, if people can get used to one flavor, getting used to another (with similar UI) will probably not be a big deal.
If SteamOS is big one day, people might be like, "Oh, Zorin is also Linux-based? Okay, I can try that on my next PC."
There will be some things to learn, but if hardware compatibility improves, the rest shouldn't be too difficult.
If Linux has enough marketshare, I think Microsoft itself will release a version of O365 that works natively. Right now, there's just not enough incentive for them to do so (in fact, quite the opposite).
I live in a country where we don't get support from any big tech, so I have no personal experience. But even if end users don't get support directly, they can usually go to the shop for aftersales service, and most provide some level of Windows troubleshooting. With Linux, you're completely on your own. I don't know if it's different in other countries.
They certainly have become less adamant that everything run on Windows only, but my intended meaning is that Office is necessary before it can gain that kind of market share. Somewhere around here (*waves hands at the forum in general*) is a lengthy thread discussing and debating various office suites and their degree of compatibility with and ability to match formatting of Microsoft Office. Even setting aside Microsoft's convenient-but-invasive Office ecosystem and interoperability, getting your document to look the way you intend in Office is pretty difficult when starting in most Linux suites, and the opposite is true, too.
The counter argument to that is you are inviting Free Software to become proprietaary and you might just as well go back to the gaol/jail of Mac and Windows. Somethings I can live without. Choice isn't one of them.
I guess that's the fundamental difference between the average computer user and Linux enthusiasts. As long as we don't understand this, we'll be left wondering why Linux desktop marketshare is so far behind.
No, it's a question of education. We need to be better ambassadors. That's why on my website I've turned Apple's "Think differently" into "Don't think different, Think better!"
While I don't disagree with many of the points discussed in this thread, from my own experience I think one of the main reasons most people don't swiich is ignorance. By ignorance I mean they are not aware of what Linux can do and for many they are not even aware Linux exists. Even If they have heard of Linux they think it is something so far out there it is not for them.
Over the past year I have put Linux on both of my brother's desktops and my daughters laptop and my daughter-in-laws laptop. I installed MXLinux on all of them with KDE de. All of them were surprised that the menu and desktop were very similar to windoze. None of them have had any problems using Linux. And I might add none of them are computer geeks. They just use their PCs for email and browsing and other normal tasks.
I really think many people would gladly use something other than windoze but they just don't know there are alternatives.
It's not impossible that, in a scenario where SteamOS becomes the big contender, Valve throws their weight around and starts going proprietary. It's not hard to point out ways in which they've not been the best possible stewards of their power, and the high publishing fees are among them. That said, I think it's unlikely.
- Valve doesn't own Linux and it can't be bought. There will always be the choices of other distributions, and being closer to something people are used to will do more to spawn interest and exploration than the comparatively big leap from Windows.
- It would take a tremendous amount of work to move from being based on Arch and relying on open source software to going proprietary. For all their influence, Valve is a company of fewer than 400 (as of 2021), and would have a very hard time doing enough development for proprietary replacements in a sane time period.
- Valve has gone to great lengths to promote choice in its Linux efforts. While the system is immutable by default, a single command is all that's required to break out of that.
- They've made no effort to stop third party tools like Decky, which work even without undoing immutability, or installing competitors like GOG, EGS, or even Xbox (though as far as I know, only streaming games works for the last there).
- While their market share is enormous, they've not historically acted inappropriately toward competitors: whether that's EGS, GOG, or farther back to Stardock's Impulse. (Impulse would still be going strong, but the CEO of Stardock didn't like that so much of the business had become their store, sold it to GameStop, and they killed it.)
- Valve actively, if slowly, provides drivers for Windows for Steam Deck. (No, you don't want Windows, but choice isn't choice if only the options you like are available.) You can do that.
- Valve strongly discourages hardware modding and voids the warranty if you mess with the insides, but that's primarily a CYA. They also posted video tutorials on exactly how to take the Deck apart safely, enabling people to do things like replace the SSD. iFixit receives official Steam Deck parts for user repairs.
- They continue to keep a lot of their work open source.
I know better than to blindly trust a corporation. Gabe Newell could die and his replacement could be a complete monster. They've done all the above though, and kept the company private, allowing them to avoid answering to the demands of shareholders and their constant insistence on short term gains. Given how Valve has carried itself in the Linux space and broader market so far, I don't see them as a threat in the near future.
FYI, I think Gabe Newel is a monster. I've been following the company since its founding, I know all about the power dynamics, and the toxic atmosphere of that company, and it was all fostered, by that man Gabe Newel. I have absolutely no love for that man whatsoever.
I honestly think the company could benefit from new leadership. I think anyone with a degree in business, could do a better job then him. And that company could use a good HR department, cause its clear that the power dynamics, toxic atmosphere, are real serious problems.
If you want the company to do better, it must first be fixed from the inside out. Its really that simple! Valve is nothing but a hardware pusher now, its all about the money with them. I really wish they would go back to making games, and not VR only games either, thats a cop out to sell VR Index headset, I aint stupid, I know what he's doing!
Get back to making games Valve, and maybe I will consider your company again. I take pride I haven't bought anything from their company, since Half Life 2, which is circa 2004, through 2008. If they want my money again, then they have to do better.
Most people with a business degree would have held an IPO, made bank on investors, and had things get even MORE toxic trying to meet shareholder demands. I say this as someone who works at a game company that was private, got bought by a public giant, and has suffered ever since, to the point that we unionized, which until lately was unheard of in the industry. There wouldn't be any improvement to games, either, if that happened. Think of the change in Bioware pre- and post-EA buyout. That's what most people with business degrees would bring Valve.