I’m sorry, but they do. They are proper keyboards following the HID spec. Plug them in anywhere and they work. Win, Mac, Linux, even an Android or Apple phone.
It’s Windows weirdness that doesn’t allow Alt codes to work with a separate numpad. It could absolutely work, but microsoft didn’t care to do it. QMK could certainly work around Window’s stupidness. But it will take someone who knows qmk, uses Windows, and cares about alt codes / doesn’t just use layers or other easier ways to type characters. Also, if you’re not willing to compile, you might have to ask someone to do it for you. Most people just share code, since running code from some rando on the internet on your keyboard blind, without seeing it, is a risk.
But still, I’m willing to bet someone will write a Windows alt code workaround, eventually. If no one does for long enough, I might do it. I believe I can think of how to work around Windows. But I don’t actually use alt codes, and I don’t enjoy booting into Windows.
People have already mentioned how they use a separate keypad on windows devices and can still use alt codes, so it’s very much not that, in fact i use a Bluetooth numpad as does my wife on our smaller windows laptop and Altcodes work flawlessly, and it’s not even a qmk pad, so it’s not using that to work around windows supposed limitation
I see others say that separate keypads do not work for Alt-codes on Windows. Just googling, I see people complaining about it. You’re pressing alt on your separate keyboard to do alt-codes on the numpad?
For sure, but if some have the issue, and others dont, its likely not windows and the individual devices being used, and in this case it would have been doable for them to program these to read as a single device, the os doesnt decide how hardware appears, thats what abstraction layers are for, thats how usb drives have been made to appear as keyboards to perform malicious actions on computers
Yeah I get that, but it’s also not much of an investment to learn.
Of course I touch type, both on the normal keyboard as well as the numpad.
Admittedly, I mostly use the num row for typing in numbers while programming for example. I rarely use the numpad, only for otp codes and similar things where only one hand is free.
My main goal is to switch to a 35 key split column staggered layout soon (charybdis nano), with thumb keys to trigger these layers. One of the reasons why I love these layers so much is that it’s kind of like vim movement, but it works everywhere as almost all applications/websites support navigation with arrow keys, pos1/end etc.
They are separate USB devices, you can’t program two separate USB devices to appear as one. USB drives which act as keyboards are entirely different, it’s reverse. One device can present as multiple, think of USB hubs, they allow this. Those malicious USB drives have a microcontroller that present as a HID keyboard and whatever else they wish.
What we could do is program a key combo on the numpad to be a left-alt, such as + and - held at the same time, so we dont lose any individual key functionality, and can still do alt-codes, it will be far less retraining than using the staggered keys, or as others have shown, i might look into the deeper qmk stuff that might be able to solve it, it is dissapointing that framework didnt just set them to read as one device, this is actually the first time i have been truly dissatisfied with any of their choices
i don’t use alt codes (i also don’t run windows) so that stuff isn’t something i’m directly concerned about. i do think it’s unfortunate you ran into this, but i also know that it’s highly nontrivial to have usb devices automatically unify in the fashion you describe.
it basically requires highly nonstandard drivers and modes of operation, which is contrary to framework’s design
+ and - at the same time for L.Alt would be a QMK combo. It’s a standard function. Sadly, Via, the GUI that Framework choose doesn’t expose combo configuring. So you’d have to compile to add combos.
Fair, to everyone their own
My goal was to show alternatives to solve this issue in a different way (that’s imo better, but it does require some initial effort).
I might also be spending too much time optimizing my environment and too little actually getting stuff done. My train of thought is that I want my environment to be as good as possible before I had to university, where I won’t have that much time to spend on this kind of stuff and need everythign to Just Work™. That’s why I switched to Colemak, learned vim stuff and Linux in general ^^
I mean, in this case its two of their own decices communicating through the same hardware, it shouldnt be any more difficult than any of the split keyboards, or keyboard seperate numpad combos etc, and you can get those for dirt cheap, so it cant be that difficult or every knockoff brand from here to china wouldnt be selling unified split usb modules for under 30 bucks
I saw that, when i get my batch 6 i am for sure going to play with it, mine isnt ansi, but i have enough experience with qmk that i feel confident in altering that for my us english. I would like to hear from framework about this, just to know why they designed it this way or if it was even intentional or whatnot, but thats just my curiosity, its nowhere near a big enough issue to cancel, or really even be that upset at then about, though picking a qmk interface that doesnt allow combos adds to the annoyance.
I’m not a fan of Via myself, but I think the choice was understandable. Via is popular and more well known. Mass-market mechanical keyboards go with Via, so more people will already be familiar with the arrangement. And the keyboards are open source. Someone can (and will) port it over. The keyboards use the RP2040, it’s very easy to reprogram. Just put it in bootloader mode, it will show up as a flash drive, then you drag-and-drop your new program to it.
I’m pretty sure the US English keyboard uses the ANSI firmware.
From what I can see, there’s no particular known design choice that causes this issue. The most common “mistake” is programming the numpad to emit top row number keys instead of numpad keys, but it doesn’t look like Framework did that.
I keep meaning to throw a question in about this, but i assume its not an issue on linux? As linux is my daily driver for my personal, its a work laptop we use the seperate numpad for with windows, i never buy laptops that dont have numpads to have ever worried about it for personal laptops
