Microsoft has been trying for years to developed a streamlined version of Windows that isn’t bogged down by decades of legacy code, as a way to better compete with modern operating systems like Android, iOS, and ChromeOS.

After scrapping several earlier projects, the company is now said to be working on a new project code-named CorePC that could allow Microsoft to divide Windows into a set of modular components, making it possible to release different versions of Windows for different devices. Another key part of CorePC? According to a report from Windows Central, it will bring state separation to Windows, which means that core operating system files will be stored to a different partition from hardware drivers, program files and user data.

Microsoft says state separation brings security and speed improvements

Among other things, this can make the operating system more secure by placing system files in a read-only partition that’s separated from program files and user data.

It an also lead to quicker operating system updates by enabling support for A/B partitions like those used by Chromebooks and recent Android devices. This allows a device to download operating system updates in the background, write those changes to a separate partition, and then switch from the A system partition to the B partition the next time you boot your computer rather than writing a bunch of changes to disk the next time you reboot.

Microsoft had planned to bring State Separation to Windows with the launch of Windows 10X a few years ago, and the company explained some of the benefits in a presentation about that operating system (which was eventually canceled before it was ever released to the public).

There’s no guarantee that this new CorePC project will ever see the light of day either. Microsoft has canceled similar projects in the past, like the ill-fated Windows Core OS.

But unlike Windows RT, Windows 10X, and Windows 10 and 11 in S mode, CorePC won’t necessarily strip away support for legacy Windows apps and force users to install Windows Store apps.

Some versions of Windows based on CorePC might do that – Windows Central says that Microsoft is testing a version of Windows that “only runs Edge, web apps, android apps, and Office apps” that would be designed for cheap PCs for the education market, and which could take up as little as 1/4th as much space as Windows 11 SE.

That could make future Windows devices truly competitive with ChromeOS, which can run reasonably well on inexpensive hardware with as little as 4GB of RAM and 32GB of storage.

But not all versions of Windows based on CorePC will be lightweight. The idea is to separate out different parts of the operating system into modules, allowing Microsoft to offer multiple versions of the OS that are appropriate for different devices.

Microsoft is said to be starting with the idea of delivering a full desktop experience, complete with support for Win32 apps, Microsoft Store apps, Progressive Web Apps, and support for features like the Windows Subsystems for Linux and Android.

But the modular nature of CorePC means that you can strip out some of the features you don’t need for entry-level hardware, devices meant for the business or education market, or depending on whether a PC is a laptop, tablet, or desktop computer, among other things.

In other words, if CorePC pans out… you might never really notice it, since your next computer may support all the apps and features as your current one. And that’s why the thing I’m most intrigued by is the move to put system files in their own partition, as that could have a big impact on security, reliability, and the speed of installing OS updates, which is something you really might notice.

Of course, if it’s not implemented properly, it could also introduce all sorts of compatibility with older Windows apps if they aren’t updated by their developers. So we could have that to look forward to as well.

via Windows Central

Support Liliputing

Liliputing's primary sources of revenue are advertising and affiliate links (if you click the "Shop" button at the top of the page and buy something on Amazon, for example, we'll get a small commission).

But there are several ways you can support the site directly even if you're using an ad blocker* and hate online shopping.

Contribute to our Patreon campaign

or...

Contribute via PayPal

* If you are using an ad blocker like uBlock Origin and seeing a pop-up message at the bottom of the screen, we have a guide that may help you disable it.

Subscribe to Liliputing via Email

Enter your email address to subscribe to this blog and receive notifications of new posts by email.

Join 9,545 other subscribers

Join the Conversation

11 Comments

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.

  1. Windows could be a more efficient and a minimalist operating system if Microsoft would go back to making Windows purely a operating system and not add in all the bloated features and services. Anyone who has used Windows over the decades has seen Windows grow into much more then a operating system that runs apps. But most operating system have bloated with age even Chrome OS has become much more then what it started out as. I don’t know that it really has hurt anything except that Windows runs poorly on very low powered chips. Which I think is more what Microsoft wants to address then anything else.

  2. Well, look up Microsoft Pluton. It’s a way to create vendor lock-in. CorePC is nothing new, but the two together would be a great way to lock people in to Microsoft. Run only “trusted” apps from them.

    It’s time to get away from x86 and find alternatives, like RISC-V or something.

    I do disagree with you Some Guy that this means the end of linux. Linux has survived a lot of attacks against it (and opensource in general) over the years and has survived.

    1. This really isn’t the end of Linux.
      It’s a prerequisite for the end of Linux, which Google has JUST TODAY committed to the Chromium Source Code: Remote Attestation. They’re calling it Web Environment Integrity. They’ve been talking about this for the past three years on a small github group that no one saw until recently, and now they’re moving the discussion off github because anyone who knows anything about computers is rightfully upset.
      You need a state separated OS because otherwise the owner might tamper with the attestor program, which is useless in establishing that “this computer is not running any browser automation, macro programs, or scrapers right now, and the browser binary was hashed and signed correctly” if you can spoof the token, and websites have plenty of incentives (DRM, crawlers, scrapers, mandates from google ads and governments, DDOS) to start using remote attestation immediately.
      The requirements from mainstream websites and anything behind Cloudflare can only increase from there.
      Linux will never be accepted by any websites if it can’t prove these things, and I see no effort to make it provable, and I can’t imagine how it could be done without compromising the user’s freedom. So even if a solution is made for Linux, we’re still probably screwed.

  3. Cool, now I know the name for this concept I hate. State separated operating systems suck and hate their users and consider you unworthy of doing anything remotely out of the ordinary with them, even fixing their creator’s mistakes.
    Not that windows cares to let you do that anyway, but the worst part is, they might actually not screw this up this time. It might do so well that using state separated operating systems, and requiring that your users use remote hardware attestation to prove the integrity of their state separated operating systems, could end up being a requirement of cybersecurity standards. And thus, using an operating system that respects your freedom is going to be looked on with more disdain than ever.

    1. It’s also worth noting that a state separated OS’s read only partition is only secure if it’s either encrypted and another OS can’t decrypt it, and/or the bootloader is locked.
      So if this takes off, we might be looking at the end of the line for actual choice in operating systems. And thus, the end of the line for community development of Linux.

      1. Maybe Microsoft will still allow us to run a Microsoft approved Linux from inside of Windows?

        1. That would be completely missing the point (for me). I switched to Linux over concerns about telemetry. I don’t want to have to worry about whether my OS is keylogging me or whether it’ll restrict me from saying certain things to certain people or accessing certain information. In theory, windows can intercept and prevent anything microsoft wants to coming out of WSL, it’s all a question of when they can be bothered to implement that.
          Not to mention the windows GUI is far worse than KDE.

    2. “State separated operating systems suck and hate their users and consider you unworthy of doing anything remotely out of the ordinary with them, even fixing their creator’s mistakes.”

      I’ll take things that aren’t true for 200.

      Also Linux has some level of state separation and performs better with it.

          1. No, it’s an issue with how these operating systems conduct themselves. Just about every state separated proprietary desktop and phone OS relies on stopping you from booting, and thus installing, other operating systems to keep itself secure. Theoretically, you could have a state separated proprietary OS that doesn’t do this, but in practice, Android, iOS, ChromeOS, and now Windows are doing this. MacOS on Apple Silicon is only kind of an exception, since if you want to use another OS, you can’t go back to MacOS.