4 ms·

>When someone seriously suggests to 'reinvent the wheel' it is time to re-focus. Goodyear released spherical wheels. It's not that stupid at all to reinvent th

by wut42 9y ago

>When someone seriously suggests to 'reinvent the wheel' it is time to re-focus. Goodyear released spherical wheels. It's not that stupid at all to reinvent the wheel, sometimes: https://www.sciencealert.com/watch-goodyear-unveils-the-future-of-tyres-with-spherical-maglev-wheels https://www.sciencealert.com/watch-goodyear-unveils-the-futu...

jacquesm 9y ago
You're missing my point. It may not be stupid to reinvent the wheel at all. But that's not a project that you should mix with inventing a self driving car unless you conclude that spherical wheels are the only way a self driving car will ever become a reality. Similar to how every software application project could theoretically result in a new database, a new operating system and if you really want to go all out a new CPU it is much more efficient to concentrate on your application if you actually want to see it go to market. > Goodyear released spherical wheels. No, they didn't, that's just some fancy CG stuff, not an actual product. Also, it just so happens that Good Year is in the wheels-and-tires business so for them such research makes good sense, and of course it is a theory of a concept, not an actual proof of concept much less something that they are ready to roll out. It's more like serious science fiction even than actual science, it's how things maybe could work, not how they actually work.
- coldtea 9y ago
  >You're missing my point. It may not be stupid to reinvent the wheel at all. But that's not a project that you should mix with inventing a self driving car That's only true if the team that reinvents the wheels and the team that does the self-driving tech is the same, and you're stealing members from the latter to the former. Obviously those are not the same teams, nor they need to be. And when you get out with your self-driving car, as 3-4 companies will do at more or less the same time or within a few years of each other, having other great new innovations on top of that will give you an edge.
  - jacquesm 9y ago
    The degree to which globular wheels would complicate such a design is such that it would leave you being passed by your competitors right out of the gate, on top of that it would complicate all kinds of things such as certification and potentially reliability and safety as well as supply chain issues after delivery. Solving too much is a fantastic way to crash a project.
    - coldtea 9y ago
      >The degree to which globular wheels would complicate such a design is such that it would leave you being passed by your competitors right out of the gate We don't know that. We just know that we're not familiar with them. If anything, globular wheels could make some kind of automated steering easier. And of course, it's obviously not like Apple only tried that design: they tried globular wheels, fat wheels, thinner wheels, wooden wheels, and what have you to see what works better -- like they did when they designed the iPhone, etc. https://www.cultofmac.com/181782/every-iphone-prototype-apple-ever-made-before-released-the-first-iphone-gallery/ https://www.cultofmac.com/181782/every-iphone-prototype-appl... https://www.theverge.com/2017/1/6/14188624/apple-prototype-iphone-ipod-click-wheel https://www.theverge.com/2017/1/6/14188624/apple-prototype-i...
      - jacquesm 9y ago
        > tried globular wheels, fat wheels, thinner wheels, wooden wheels, and what have you to see what works better Yes, and 'standard wheels' would have saved them time, money and other resources and would have increased the chances of bringing a self driving car to market.
        coldtea 9y ago
        >Yes, and 'standard wheels' would have saved them time, money and other resources Time from what? It's not some team considering other kinds of wheels was holding back their self-driving researchers... You seem to have an overly linear and heavy-first concept of how such a development should go.
skgoa 9y ago
They made a little PR thing. That's far from "releasing" anything.

by wut42 9y ago

jacquesm 9y ago

You're missing my point. It may not be stupid to reinvent the wheel at all. But that's not a project that you should mix with inventing a self driving car unless you conclude that spherical wheels are the only way a self driving car will ever become a reality. Similar to how every software application project could theoretically result in a new database, a new operating system and if you really want to go all out a new CPU it is much more efficient to concentrate on your application if you actually want to see it go to market. > Goodyear released spherical wheels. No, they didn't, that's just some fancy CG stuff, not an actual product. Also, it just so happens that Good Year is in the wheels-and-tires business so for them such research makes good sense, and of course it is a theory of a concept, not an actual proof of concept much less something that they are ready to roll out. It's more like serious science fiction even than actual science, it's how things maybe could work, not how they actually work.

coldtea 9y ago
>You're missing my point. It may not be stupid to reinvent the wheel at all. But that's not a project that you should mix with inventing a self driving car That's only true if the team that reinvents the wheels and the team that does the self-driving tech is the same, and you're stealing members from the latter to the former. Obviously those are not the same teams, nor they need to be. And when you get out with your self-driving car, as 3-4 companies will do at more or less the same time or within a few years of each other, having other great new innovations on top of that will give you an edge.
- jacquesm 9y ago
  The degree to which globular wheels would complicate such a design is such that it would leave you being passed by your competitors right out of the gate, on top of that it would complicate all kinds of things such as certification and potentially reliability and safety as well as supply chain issues after delivery. Solving too much is a fantastic way to crash a project.
  - coldtea 9y ago
    >The degree to which globular wheels would complicate such a design is such that it would leave you being passed by your competitors right out of the gate We don't know that. We just know that we're not familiar with them. If anything, globular wheels could make some kind of automated steering easier. And of course, it's obviously not like Apple only tried that design: they tried globular wheels, fat wheels, thinner wheels, wooden wheels, and what have you to see what works better -- like they did when they designed the iPhone, etc. https://www.cultofmac.com/181782/every-iphone-prototype-apple-ever-made-before-released-the-first-iphone-gallery/ https://www.cultofmac.com/181782/every-iphone-prototype-appl... https://www.theverge.com/2017/1/6/14188624/apple-prototype-iphone-ipod-click-wheel https://www.theverge.com/2017/1/6/14188624/apple-prototype-i...
    - jacquesm 9y ago
      > tried globular wheels, fat wheels, thinner wheels, wooden wheels, and what have you to see what works better Yes, and 'standard wheels' would have saved them time, money and other resources and would have increased the chances of bringing a self driving car to market.
      - coldtea 9y ago
        >Yes, and 'standard wheels' would have saved them time, money and other resources Time from what? It's not some team considering other kinds of wheels was holding back their self-driving researchers... You seem to have an overly linear and heavy-first concept of how such a development should go.

skgoa 9y ago

They made a little PR thing. That's far from "releasing" anything.