I find it amusing that every single one of our high-speed standards (ethernet, HDMI, PCIE, displayport, USB) are basically just the same high-speed differential pairs under the hood with different framing.
I don't know exactly what we can do about this situation, but at some point it probably makes sense for all of them to converge into a multi-lane, differential-pair standard (or just support nesting each other arbitrarily).
Thunderbolt actually includes a lot of other stuff that isn't just "pcie" in a cable - like USB and DisplayPort streams in addition (at the same time) to the PCIe links.
There is also U.2 for shorter links. There are relatively trivial risers (not sure if it needs a chip or not, but either way they're under $30) that let you turn regular PCIe x4 or M.2 x4 slots into U.2 cables. It's intended for the enterprise space but if you just want "pcie in a cable" that's what it does.
ExpressCard actually also was a PCIe 1.1x1 or 2.0x1 link as well, and people used risers to break out to external GPUs just as people do with thunderbolt. There are various eGPU setups "GDC Beast" is one I remember?) that are still available for ye olde thinkpad if you want more/better display connectors.
Also obviously miners use those shitty USB-based riser cables (they aren't USB, but use USB wiring?) with x1 connectivity.
I dunno whether any of those are "legal" according to the PCIe signalling requirements but in practice you can get short runs pretty reliably especially at lower speeds, and good PCIe riser cables can even do 3.0x16 or 4.0x16 capability.
Nah, we cant go with 1.0. Bad for adoption. Since we can push 10Gbps easily, and maybe up to 40Gbps if we try hard, lets call the first version usb "ultraspeed" and let the marketing guys work with that!
You can have arbitrarily many, so long as the serial lanes are independent of each other. PCIe can have lots, for example. Because they are independent, they won't suffer from clockskew problems that made parallel signaling impractical at high speeds and it's often not required to connect all of the wires (e.g. an x16 card hanging out of an x8 or smaller slot) .
I don't know exactly what we can do about this situation, but at some point it probably makes sense for all of them to converge into a multi-lane, differential-pair standard (or just support nesting each other arbitrarily).