3 ms·

Agreed and more. To claim we're churning through solutions is absurd, just off the wall perposterous. USB Micro was introduced in 2007, pretty much as a respin

Agreed and more. To claim we're churning through solutions is absurd, just off the wall perposterous. USB Micro was introduced in 2007, pretty much as a respin of the physical USB Mini connector- introduced in 2000. USB Micro is pretty much technology from 2000, with a couple small connector fixes. Whether we want to accept USB Micro as pretty much essentially 21 year old legal-drinking-age-in-any-country or only 14 year old tech, this connector has been around for a long long long long time, and in 2010, when it was "standardized" was a pretty obvious sensible choice with little on the horizon. The big characterization I'd make between USB Micro and USB-C (2014) is that USB Micro was designed for one specific task: connecting 5W power (expanded (unofficially?) to 11W) and USB2.0. USB-C is a much more abstract connector, with gobs of general purpose data & a huge range of power ranges. For data there's two pairs of highspeed differential lanes (so, 8 pins), plus some other lines, for power there's way more amps (5A) & scalable voltage (really scalable with USB EP's 48V) meaning tons of juice. These are all opt-in for use. There's a very flexible power delivery range, there's lanes designed for general high speed connection use not just USB. There were upfront designs for how to run multiple protocols, not just USB, over the connector. With USB4 there's packet level sharing, so net bandwidth can be shared in more interesting mixes than 50/50 or to many destinations at once. USB-C can do just about anything. We can keep changing our uses of it to do whatever. It's almost hard to imagine what else we might possibly want. Intel's mobile chip's onboard thunderbolt/USB-C ports offer 40 Gbps system-to-system connectivity, while desktops are still slowly marching towards 2.5Gbit ethernet. 10Gbit connectivity is still semi exotic to anyone but hardcore compute enthusiasts, but this connector has proven itself capable of way way more already, at a clearly mass-market price point. If there's one innovation I would like to see from USB-C and USB in general, it'd be longer connections. I'd love to see a 10Gbit USB4.x mode that works over a 6m cable, for example. I don't think the connector has much to do with the range difficulties faced by modern, high speed USB. And similarly I don't see much push to create a higher-lane count connector, but that's another possible place where USB-C proves less-than-ideal long term.

daurnimator 5y ago
All of those usb specs are quite young and none have really lasted. Compared to other connectors that people may know: - RJ45/8P8C was standardised in 1987 - the 3.5mm audio jack in 1979 - the PCI express slot from 2003 These are all still in use today. People expect to change connectors maybe twice in a lifetime, not every decade.
- samatman 5y ago
  This is a brilliant illustration of survivor bias. I sometimes wish I hadn't thinned my cable collection in the late teens, I could show you fifteen connectors which haven't lasted as long as, say, Micro-B.
- rektide 5y ago
  I'd love to have something new & interesting to look forward to. It'd be so neat. I really want a fast, simple, low cost, decent power alternative, that isn't a huge jumble of old legacy specs. Even USB4, which I love, still has dedicated old USB2 wires running around, because it's just too pre-modern to encapsulate adequately. This fact is a shot across your bow: even though USB-C is from 2014, it subsumes the previous connector. It's backward connectively compatible. The RJ45 & TRS jacks you cite are so primitive, so simple, so single-minded that of course they had no competition. There was so little to compete over, to adapt to, to grow into. A different implementation would have been irrelevant. Except, oh, even TRS/3.5mm audio had to evolve & grow to become TRRS, to add a mic channel: a case where even your immutable forever connectors aren't. You're right that the time-scale isn't long enough to make USB-C the obvious forever connector, that the time-scales talked so far aren't that long. But digital connective technology isn't that old. We didn't have anything before, and Micro-B basically served & would keep serving, but we want more. I feel like there's an obvious huge colossal truth batting at you & you're ignoring the enormity of how clear, deliberate, & sequential this evolution has been. The fact that USB-C emerged out of the previous connector feels to me like further evidence that this is part of one continuum, that your greivance that connectors can only change once or twice is missing the clear upfront truth of the matter. And now we're here. We spent 30 years going from crap at consumer digital systems to great, and I have a hard time seeing this as anything but the semi-final destination for us. I don't see good change as likely or probable in quite a long long long time. Perhaps lifetimes. USB-C is the final destination of mainstream computing, as far as it looks, and I merely wish it wouldn't last as many lifetimes as it seems probable to.
  - kube-system 5y ago
    > I don't see good change as likely or probable in quite a long long long time. Perhaps lifetimes. USB-C is the final destination of mainstream computing, as far as it looks, and I merely wish it wouldn't last as many lifetimes as it seems probable to. We’re already well on our way to removing connectors and using Qi/WiFi/Bluetooth. I bet we’ll see phones without connectors soon.

daurnimator 5y ago

All of those usb specs are quite young and none have really lasted. Compared to other connectors that people may know: - RJ45/8P8C was standardised in 1987 - the 3.5mm audio jack in 1979 - the PCI express slot from 2003 These are all still in use today. People expect to change connectors maybe twice in a lifetime, not every decade.

samatman 5y ago
This is a brilliant illustration of survivor bias. I sometimes wish I hadn't thinned my cable collection in the late teens, I could show you fifteen connectors which haven't lasted as long as, say, Micro-B.
rektide 5y ago
I'd love to have something new & interesting to look forward to. It'd be so neat. I really want a fast, simple, low cost, decent power alternative, that isn't a huge jumble of old legacy specs. Even USB4, which I love, still has dedicated old USB2 wires running around, because it's just too pre-modern to encapsulate adequately. This fact is a shot across your bow: even though USB-C is from 2014, it subsumes the previous connector. It's backward connectively compatible. The RJ45 & TRS jacks you cite are so primitive, so simple, so single-minded that of course they had no competition. There was so little to compete over, to adapt to, to grow into. A different implementation would have been irrelevant. Except, oh, even TRS/3.5mm audio had to evolve & grow to become TRRS, to add a mic channel: a case where even your immutable forever connectors aren't. You're right that the time-scale isn't long enough to make USB-C the obvious forever connector, that the time-scales talked so far aren't that long. But digital connective technology isn't that old. We didn't have anything before, and Micro-B basically served & would keep serving, but we want more. I feel like there's an obvious huge colossal truth batting at you & you're ignoring the enormity of how clear, deliberate, & sequential this evolution has been. The fact that USB-C emerged out of the previous connector feels to me like further evidence that this is part of one continuum, that your greivance that connectors can only change once or twice is missing the clear upfront truth of the matter. And now we're here. We spent 30 years going from crap at consumer digital systems to great, and I have a hard time seeing this as anything but the semi-final destination for us. I don't see good change as likely or probable in quite a long long long time. Perhaps lifetimes. USB-C is the final destination of mainstream computing, as far as it looks, and I merely wish it wouldn't last as many lifetimes as it seems probable to.
- kube-system 5y ago
  > I don't see good change as likely or probable in quite a long long long time. Perhaps lifetimes. USB-C is the final destination of mainstream computing, as far as it looks, and I merely wish it wouldn't last as many lifetimes as it seems probable to. We’re already well on our way to removing connectors and using Qi/WiFi/Bluetooth. I bet we’ll see phones without connectors soon.