Permission is one. (A pretty important one, at that.)
A second is funding, the prospect of becoming profitable. This hurdle, Framework seems to have taken (for now), but it also seems to be stopping competitors (for now).
Would be great if something grew from hobbyist / enthusiasts.
Can you elaborate on the permission comment? The designs in their github allow commercial use with attribution. They even link to a Google form for submitting your product to their marketplace. Is there a submarine restriction somewhere?
>Would be great if something grew from hobbyist / enthusiasts.
Making PCBs to support modern x86 and even high-end ARM CPUs with DDR5 RAM and also PCI-express is beyond the skillset of a vast majority of hobbyist and enthusiasts. It gets so so far up the creek of high frequency signal engineering under electrical engineering. It's not impossible, just very time consuming and costly (between equipment required for testing and PCB manufacturing).
But yea all the other stuff in a framework could get hobbyists and enthusiasts making their own elements like the expansion cards and keyboard significantly more easily if there was a need for it.
I did a DDR3 layout, sure there's a lot to think about if you want it to work in extreme temperatures or with lots of EMI. But for hobbyist use its essentially get the right trace width for the PCB material using an online impedance calculator, and then length match and match the count of vias for all your traces of the same bus and don't exceed the length restrictions. Don't forget to have a good ground plane. This will work 99% of the time.
Its time consuming and annoying but not rocket science.
A second is funding, the prospect of becoming profitable. This hurdle, Framework seems to have taken (for now), but it also seems to be stopping competitors (for now).
Would be great if something grew from hobbyist / enthusiasts.