I have used Windows since a neighbour handed me a demonstration floppy for version 3.0. So when Plug & Play arrived, I watched it closely, and before long I had my own name for it: Plug & Pray. I went on calling it that for as long as I kept using Windows. The name stuck because it kept being accurate.
It was not that the idea was wrong. The idea was excellent: you attach a thing, the computer works out what it is, and it starts working. The trouble was what happened when the computer got it wrong, which was often. It guessed. When the guess was bad, the device half-worked, or worked until an update, or did nothing at all — and the machine told you almost nothing about which of those had happened, or why.
Linux has an opportunity Windows no longer has: to do this properly, from the beginning, rather than patch over three decades of guessing. That is what this project is about.
Linux Already Proves It Can Be Done
Printers are the example to point at. Plugging a printer into a modern Linux desktop generally just works. The system finds it, identifies it, and prints — frequently with no driver download at all, because the printer and the computer agree on a common protocol and get on with it. No installer, no vendor software bundle, no thirty-megabyte download that also installs a toolbar.
This is genuinely better than what most people are used to, and it is worth saying out loud, because the reputation Linux carries for hardware trouble is a decade out of date in this particular area.
So the capability exists. The question is why it stops at printers.
It Should Be That Good For Everything
Every USB device, however obscure, should get the same treatment. Not just the popular ones, not just the ones with a large enough user base that somebody volunteered to package the support. All of them.
That is a large ambition and it will not be finished quickly. But the direction matters more than the pace, because the alternative is what we have now: common hardware works, unusual hardware is a lottery, and the user has no way to know which they have bought until they get it home.
And When It Cannot Work, It Should Say So
This is the part nobody does, and it is the part I care about most.
Today, when you plug in a device Linux cannot support, what happens is nothing. No message. No error. No indication that anything was attempted. The information exists — the system can see the device, and the device announces who made it and what model it claims to be — but none of that reaches the person standing there wondering whether the cable is loose.
A technical user can go and look. The kernel log will have something to say, and so will the journal. But an ordinary person will never open a terminal to find out why their scanner did not appear, and it is unreasonable to expect them to.
What should happen instead is a plain message. Something a person can read and act on:
- What the device is. The system already knows what the hardware reports itself to be. Say it.
- That no driver is installed for it. Not a silence — a statement.
- Where to go next. The manufacturer’s support page, the project that would maintain such a driver, or a page that explains that support does not exist yet. Anywhere is better than nowhere.
That last item is the whole point. A user who is told “this device needs a driver that is not installed, and here is where to ask” has a problem they can solve. A user who is told nothing has only a computer that does not work, and no way to tell whether the fault is the device, the cable, the operating system or themselves. The second case is where people give up and go back to Windows.
Why This Is Not An Abstract Interest
I convert people from Windows to Linux for a living. That is the other half of my week — onsite work through Mobile PC Clinic, largely for people whose perfectly serviceable computers were left behind when Windows 10 stopped receiving security updates.
Device support is what decides whether those conversions stick. Not the desktop, not the software, not the learning curve — people manage all of that. It is the printer, the scanner, the webcam, the odd USB gadget somebody has owned for eleven years and relies on. When one of those fails silently, the conversation stops being about an operating system and starts being about whether I have broken their computer.
So this is not a theoretical complaint about user experience. It is the single most common obstacle I meet, in houses, with real people, most weeks.
Where This Stands
At the beginning. This page describes what the Foundation intends to work on and why, not a body of finished work — there are no patches upstream to point at yet, and I would rather say that here than imply otherwise.
If you maintain something in this area, or you have a device that fails silently and you have documented what it reports, that is genuinely useful and I would like to hear about it. Write to Support@JCArtes.Org. Concrete cases of hardware that goes quiet are worth more at this stage than agreement.