14 ms·

Reading through this, I'm left unable to see the underlying forces. None of the people working on this spec are idiots, so how did they arrive here? What are th

Reading through this, I'm left unable to see the underlying forces. None of the people working on this spec are idiots, so how did they arrive here? What are the forces and unstated requirements behind the scenes that I don't understand that have led to a USB4 spec with this many potential flavors and absurd naming conventions to express them? Is it something about cost? Or patents? Or some random company who won't budge on something but needs to be included?

AceJohnny2 7y ago
I suspect the underlying force is "A Committee".
mtgx 7y ago
With Type-C and USB 3 they obviously tried to cater to everyone's mother and their dog from the USB consortium. 99.99% of consumers will never know what speed their type-c cable will have or what type of connections it actually supports. And then the whole fiasco with USB 3.1 and 3.1 gen 2 and 3.2 or whatever that whole mess was. That was beyond idiotic - it was just pure corruption. The group obviously worked with the OEMs so they can both lie to customers about getting a "next-gen" connection. https://arstechnica.com/gadgets/2019/02/usb-3-2-is-going-to-make-the-current-usb-branding-even-worse/ https://arstechnica.com/gadgets/2019/02/usb-3-2-is-going-to-...
throwaway156503 7y ago
> None of the people working on this spec are idiots There are a lot of very smart people in the world that lose sight of simple concepts and their value, however.
- AceJohnny2 7y ago
  I've witnessed some Standards negotiation in a previous job, involving the Khronos OpenMAX standard (which is DirectShow/GStreamer for mixed HW/SW multimedia codecs). My company and another had opposing solutions to a problem, which made opposite tradeoffs and opposite implementations (ours worked better for a more HW-focused solution, theirs for a more SW-focused solution) How does one choose? In the end, it's backroom negotiation, diplomacy, and compromise. In other words, politics.
  - rswail 7y ago
    In other words, and as the IETF demonstrates, consensus.
iagooar 7y ago
One smart person will find the most elegant solution to a problem. A committee of smart people will find the most elegant solution to fuck up a problem.
- imtringued 7y ago
  In a committee of smart people everyone has a different problem.
- ineedasername 7y ago
  That kind of perpetuates the myth of the solo inventor as the only true innovator, when in reality it tends to be groups of people, formally or informally affiliated, that move the ball forward on science, engineering, etc. Like the teams of people employed by Edison, who was as much a project manager as he was an inventor. Or look at something like the Manhattan project, where a vast team of very smart people worked together. Or the Bletchly park folks, of whom Turing may have been the most prominent but by no means the only smart person to help solve the problem of decoding German communication. He worked closely with others that offered their own significant improvements to the bombe, including three Polish crypt-analysts (most notably Rejewski) that Turing worked with to further development.
  - patentatt 7y ago
    Perhaps it’s more relevant to consumer facing products that require a balance of features, performance, and user experience to be good. I think that’s harder for a committee to do than scientific work. While USB isn’t a singular product, it is a component of consumer products, however. I wonder if much of the success of Apple was due in large part to Steve Jobs’ heavy personal involvement at the design stage, and his famous rough-edged character.
r00fus 7y ago
Profits are the reason. With so much confusion the safest bet for the consumer will be to use the cabling from the vendor of the device. Vendors in turn can charge more.
msbarnett 7y ago
> Is it something about cost? It's hugely about: cost, and the desire of the USB forum to ship a single "does everything" protocol that keeps a large chunk of the industry from choosing to stay on some old version of the spec forever. Tons and tons of manufacturers of cheap low end devices out there aren't willing to replace a $0.10 USB controller with something that costs 30x as much when they have no need for the high-end speeds of Thunderbolt 3, fast charging, DisplayPort alternate modes, etc. Consequently the spec can either contain a lot of optional features they can choose to ignore, or it can mandate support for all of the 40 Gbps active cable fast charging etc options, and watch as devices partition themselves between the few the need all of that stuff, and the majority that ship simple micro-USB 2.0 hardware. If they did the latter, then USB 4 is essentially pointless: that's the exact situation we were already in 5+ years ago, except the cheap side was called "USB 2" and the expensive side was called "Thunderbolt". All we've done is rename the expensive side to "USB 4". If they do the former, they ship a "spec" so full of options that it's already nearly impossible to tell exactly what devices or features any given USB C port will support without recourse to a specifications sheet. Fundamentally, "one port and protocol to rule them all" is probably just a bad idea. Trying to serve the needs of both cheap slow devices and expensive fast ones puts the forum in a contradictory position, and the only "benefit" from combining the standards that serve each niche is opaque ports and cables that consumers can't easily reason about the compatibility of.
- rowanG077 7y ago
  I feel that all of this would be MUCH less of a problem with simple code that are required on each USB cable with the exact specs it supports. This code could be put into a tool which then decodes to show what exactly that cable supports.
  - Kalium 7y ago
    This works until some mgf decides it's easier to claim to support everything, ships a simple controller that does that, it gets widely used, and a lot of harware becomes mysteriously incompatible as a result. Trusting vendors to be honest in declaring capabilities is not always the best of all possible approaches, historically.
jjoonathan 7y ago
Back when I wanted to do Computer Engineering, I more-than-skimmed-less-than-read the USB 1,2,3 specs and the PCIe 3 specs. The contrast was startling. The USB specs were a complete mess. The PCIe specs were not. This extended from the highest top level decisions (understanding the benefits of packet switched networks) to the lowest (missing figures, ambiguous and redundant terminology, etc). Something was systematically different between the USB-IF and the PCI-SIG, and whatever it was persisted over many years... and likely continues to this very day, if outward appearances are any indication. If anybody has insight into what makes these groups different, or even just speculation, I'd love to hear it.
- flyinglizard 7y ago
  That's not a very good comparison. USB covers up to the application layer, while PCIe is much more low level than that. USB is a great standard. Or at least, was until 2.0. It was ahead of its time in performance and the same interoperability principles carried it forward for over 20 years. If you compare the first USB standards to other standards of the era, the amount of thought that has gone into it is fantastic. The fact it has no rivals even today is telling. Compare it to the clusterfuck that's Bluetooth, which took well over a decade to get to a point where its kinda reliable (and it can still vary between implementations). Unfortunately USB became too complicated with the various modes and backward tidbits it now supports. Also, the rates it now supports have taken it to a whole new level of implementation complexity (ICs, board design, power options, etc).
  - cjbprime 7y ago
    We're really going with a protocol that involves constant polling (rather than interrupts) as being ahead of its time? :)
    - flyinglizard 7y ago
      The USB hardware "polls" by acting as a master. It's a master-slave protocol which is just fine for simplicity and bus utilization. The polling is abstracted from the software with the concept of hardware endpoints (mailboxes, really). How are you going to implement interrupts over a serial bus, unless it's multi-master (very complicated implementations, slow)? You don't have direct interrupt lines like in PCI.
colejohnson66 7y ago
> None of the people working on this spec are idiots, so how did they arrive here? I’m convinced it’s so they can advertise as supporting USB4 when they really only support the slowest (aka cheapest) method
xvector 7y ago
Can someone in the USB4 working group actually, well, speak out?
- StreamBright 7y ago
  I am not sure they are on HN.
  - dbuder 7y ago
    Really? It seems like everyone is here. Got an old IRC story? The other party will pop up. Got a question about some old software? The first reply could be from the person who wrote it.
- Murkin 7y ago
  I can tell you war stories of USB-IF, but only from back in 2006. Prob nothing changed.
madengr 7y ago
Design by committee at it finest.
Murkin 7y ago
As someone who been to USB-IF committee back around 2006 There are some smart people there, unfortunately some are old-timer in large orgs - they were pushed there "to do least harm". Didn't care less about how bad some of the proposals were, as long as they were according to the required format. We tried to fight a bit, but it was a lost cause :\

AceJohnny2 7y ago

I suspect the underlying force is "A Committee".

mtgx 7y ago

With Type-C and USB 3 they obviously tried to cater to everyone's mother and their dog from the USB consortium. 99.99% of consumers will never know what speed their type-c cable will have or what type of connections it actually supports. And then the whole fiasco with USB 3.1 and 3.1 gen 2 and 3.2 or whatever that whole mess was. That was beyond idiotic - it was just pure corruption. The group obviously worked with the OEMs so they can both lie to customers about getting a "next-gen" connection. https://arstechnica.com/gadgets/2019/02/usb-3-2-is-going-to-make-the-current-usb-branding-even-worse/ https://arstechnica.com/gadgets/2019/02/usb-3-2-is-going-to-...

throwaway156503 7y ago

> None of the people working on this spec are idiots There are a lot of very smart people in the world that lose sight of simple concepts and their value, however.

AceJohnny2 7y ago
I've witnessed some Standards negotiation in a previous job, involving the Khronos OpenMAX standard (which is DirectShow/GStreamer for mixed HW/SW multimedia codecs). My company and another had opposing solutions to a problem, which made opposite tradeoffs and opposite implementations (ours worked better for a more HW-focused solution, theirs for a more SW-focused solution) How does one choose? In the end, it's backroom negotiation, diplomacy, and compromise. In other words, politics.
- rswail 7y ago
  In other words, and as the IETF demonstrates, consensus.

iagooar 7y ago

One smart person will find the most elegant solution to a problem. A committee of smart people will find the most elegant solution to fuck up a problem.

imtringued 7y ago
In a committee of smart people everyone has a different problem.
ineedasername 7y ago
That kind of perpetuates the myth of the solo inventor as the only true innovator, when in reality it tends to be groups of people, formally or informally affiliated, that move the ball forward on science, engineering, etc. Like the teams of people employed by Edison, who was as much a project manager as he was an inventor. Or look at something like the Manhattan project, where a vast team of very smart people worked together. Or the Bletchly park folks, of whom Turing may have been the most prominent but by no means the only smart person to help solve the problem of decoding German communication. He worked closely with others that offered their own significant improvements to the bombe, including three Polish crypt-analysts (most notably Rejewski) that Turing worked with to further development.
- patentatt 7y ago
  Perhaps it’s more relevant to consumer facing products that require a balance of features, performance, and user experience to be good. I think that’s harder for a committee to do than scientific work. While USB isn’t a singular product, it is a component of consumer products, however. I wonder if much of the success of Apple was due in large part to Steve Jobs’ heavy personal involvement at the design stage, and his famous rough-edged character.

r00fus 7y ago

Profits are the reason. With so much confusion the safest bet for the consumer will be to use the cabling from the vendor of the device. Vendors in turn can charge more.

msbarnett 7y ago

> Is it something about cost? It's hugely about: cost, and the desire of the USB forum to ship a single "does everything" protocol that keeps a large chunk of the industry from choosing to stay on some old version of the spec forever. Tons and tons of manufacturers of cheap low end devices out there aren't willing to replace a $0.10 USB controller with something that costs 30x as much when they have no need for the high-end speeds of Thunderbolt 3, fast charging, DisplayPort alternate modes, etc. Consequently the spec can either contain a lot of optional features they can choose to ignore, or it can mandate support for all of the 40 Gbps active cable fast charging etc options, and watch as devices partition themselves between the few the need all of that stuff, and the majority that ship simple micro-USB 2.0 hardware. If they did the latter, then USB 4 is essentially pointless: that's the exact situation we were already in 5+ years ago, except the cheap side was called "USB 2" and the expensive side was called "Thunderbolt". All we've done is rename the expensive side to "USB 4". If they do the former, they ship a "spec" so full of options that it's already nearly impossible to tell exactly what devices or features any given USB C port will support without recourse to a specifications sheet. Fundamentally, "one port and protocol to rule them all" is probably just a bad idea. Trying to serve the needs of both cheap slow devices and expensive fast ones puts the forum in a contradictory position, and the only "benefit" from combining the standards that serve each niche is opaque ports and cables that consumers can't easily reason about the compatibility of.

rowanG077 7y ago
I feel that all of this would be MUCH less of a problem with simple code that are required on each USB cable with the exact specs it supports. This code could be put into a tool which then decodes to show what exactly that cable supports.
- Kalium 7y ago
  This works until some mgf decides it's easier to claim to support everything, ships a simple controller that does that, it gets widely used, and a lot of harware becomes mysteriously incompatible as a result. Trusting vendors to be honest in declaring capabilities is not always the best of all possible approaches, historically.

jjoonathan 7y ago

Back when I wanted to do Computer Engineering, I more-than-skimmed-less-than-read the USB 1,2,3 specs and the PCIe 3 specs. The contrast was startling. The USB specs were a complete mess. The PCIe specs were not. This extended from the highest top level decisions (understanding the benefits of packet switched networks) to the lowest (missing figures, ambiguous and redundant terminology, etc). Something was systematically different between the USB-IF and the PCI-SIG, and whatever it was persisted over many years... and likely continues to this very day, if outward appearances are any indication. If anybody has insight into what makes these groups different, or even just speculation, I'd love to hear it.

flyinglizard 7y ago
That's not a very good comparison. USB covers up to the application layer, while PCIe is much more low level than that. USB is a great standard. Or at least, was until 2.0. It was ahead of its time in performance and the same interoperability principles carried it forward for over 20 years. If you compare the first USB standards to other standards of the era, the amount of thought that has gone into it is fantastic. The fact it has no rivals even today is telling. Compare it to the clusterfuck that's Bluetooth, which took well over a decade to get to a point where its kinda reliable (and it can still vary between implementations). Unfortunately USB became too complicated with the various modes and backward tidbits it now supports. Also, the rates it now supports have taken it to a whole new level of implementation complexity (ICs, board design, power options, etc).
- cjbprime 7y ago
  We're really going with a protocol that involves constant polling (rather than interrupts) as being ahead of its time? :)
  - flyinglizard 7y ago
    The USB hardware "polls" by acting as a master. It's a master-slave protocol which is just fine for simplicity and bus utilization. The polling is abstracted from the software with the concept of hardware endpoints (mailboxes, really). How are you going to implement interrupts over a serial bus, unless it's multi-master (very complicated implementations, slow)? You don't have direct interrupt lines like in PCI.

colejohnson66 7y ago

> None of the people working on this spec are idiots, so how did they arrive here? I’m convinced it’s so they can advertise as supporting USB4 when they really only support the slowest (aka cheapest) method

xvector 7y ago

Can someone in the USB4 working group actually, well, speak out?

StreamBright 7y ago
I am not sure they are on HN.
- dbuder 7y ago
  Really? It seems like everyone is here. Got an old IRC story? The other party will pop up. Got a question about some old software? The first reply could be from the person who wrote it.
Murkin 7y ago
I can tell you war stories of USB-IF, but only from back in 2006. Prob nothing changed.

madengr 7y ago

Design by committee at it finest.

Murkin 7y ago

As someone who been to USB-IF committee back around 2006 There are some smart people there, unfortunately some are old-timer in large orgs - they were pushed there "to do least harm". Didn't care less about how bad some of the proposals were, as long as they were according to the required format. We tried to fight a bit, but it was a lost cause :\