Nothing. It's very well possible that in future, we have Apple devices that only run iOS, Google devices that only run Android/ChromeOS/Whatever and Microsoft devices that only run Windows 10 - and a large server market with a tiny consumer segment of devices that run linux.
We'll all be fine in that world. We just won't have any control about our devices anymore.
Perhaps Linux, BSD, etc. users should be creating their own PC standards doc and asking OEMs to adhere to it instead?
That's actually a good idea in theory - so far that "standard" had been the x86/x64 IBM compatible PC. Except that I fear they lack the market power to actually make such a standard more than a stack of paper. You'd need an OEM network the size of Microsoft or enough capital and brand trust so you can make your own devices. I see neither for linux.
If I buy some hardware, I've done research on whether all of its chipsets and devices have Linux driver support just as a matter of course, because I don't expect that Linux will run on every PC. But I feel like I have a right to know if a company puts a non-standard artificial restriction on their hardware. Apparently in this case, Windows 10 doesn't even have the driver included by default, so I can't even reload the software that the machine was designed to run, on my own. That sounds like a defect in the product, to me.
Where? I'm looking at http://shop.lenovo.com/us/en/laptops/yoga/900-series/yoga-90... and the specs page lists only Windows 10 Home and Pro in the operating systems section.
> and effort was expended specifically to disable that option
Lenovo has issued a formal statement denying it. http://www.techrepublic.com/article/lenovo-denies-deliberate...
I was imprecise. The chipset supports a Linux-compatible option, were it not explicitly disabled in the firmware.
> Lenovo has issued a formal statement denying it.
They deny purposefully disabling Linux, and I don't have any evidence against that denial. They don't deny disabling AHCI, because they clearly did. In the disassembly of the code: https://lnv.i.lithium.com/t5/image/serverpage/image-id/76843...
Intel's RAID is implemented in their driver. Switching the option from AHCI to RAID switches the disk controller's ID so it doesn't bind to the OS's standard driver (whichever OS that is), and binds to Intel's instead (assuming it's present). Intel provides its driver (and only for Windows), Lenovo locks the option so that hapless users mucking about in the UEFI settings don't make their machines unbootable.
Lenovo can deny "deliberately blocking Linux", but they can't deny "practically blocking Linux".
I expect any computer I own to run anything I put on it.
Locking out software is like a car maker trying to control what brand of gas you put in your car.
None of those other examples are sold with the intention of the user changing the system in a way unexpected by the vendor.
PCs are supposed to be open-ended, and for this you typically pay more for the hardware capability than a product which serves as a gateway into a walled-garden ecosystem (phones, gaming consoles, etc.)
Any cpu in my household is mine (presumably.) I don't care what peripherals it has, it's just a computer.
From my POV all those devices are just computers with various peripherals. I don't see any way in which attaching a specific peripheral changes the universal nature of the cpu.
How else am I going to avoid the crapware/malware that lenovo is famous for?
Happens that the same BIOS tweak that prevents windows from working also prevents linux from working.
I don't expect lenovo to spend time/effort accommodating linux, but I also expect them to not artificially limit when I can do with a laptop I purchase.
A offline (PDF) version is available at the bottom-most link on the page.
A serious question: What gives Linux users the right to expect that Linux will run on any given PC / laptop?
Think about it. These PCs/laptops are designed to adhere to Microsoft's Windows PC hardware standards doc (which was never, ever an open standard) and, with few exceptions, only tested against various versions of Windows. Linux users are failing to receive a feature that they were never promised in the first place so it's unclear that they have any valid cause for complaint.
Perhaps Linux, BSD, etc. users should be creating their own PC standards doc and asking OEMs to adhere to it instead?