5 ms·

I took a class in cognitive design (fancy name for controls design) with Donald Norman (he wrote Design of Everyday things) and his standard assignment was to a

I took a class in cognitive design (fancy name for controls design) with Donald Norman (he wrote Design of Everyday things) and his standard assignment was to ask us to make the controls for a clock radio television. This was a long time ago. One of the points of the exercise was to make us realize that combining different devices and their controls didn't always make sense. In the same sense in a car their is a lot to be gained by keeping the controls for audio and the controls for temperature separate. The only good reason is to save money on buttons and knobs.

sizzle 3y ago
Awesome was this class at UCSD in the cogsci department?
- georgeecollins 3y ago
  Yes
robomartin 3y ago
> The only good reason is to save money on buttons and knobs. That’s not quite right. At scale the buttons and knobs are cheap. Having designed many control panels for a wide range of applications, I can tell you that good design is an excruciatingly long and painfully expensive undertaking. You cannot recompile your way out of a poorly working physical control panel or bad design. You can’t just throw something together for testing and finish it while other teams complete the design of the rest of the car, machine, etc. As a real example, you might have to evaluate hundreds of button designs for tactile, audible and visible feel. Once sub-selected, you might evaluate over a dozen full scale control panel mock-ups before design lock. Etc. Touch screens allow you to eliminate a huge portion of what goes into creating a user interphase.
- georgeecollins 3y ago
  Good to know! I always thought it was to save money.
  - robomartin 3y ago
    Well, it does save money…just not necessarily due to hardware cost differences.
  - Summershard 3y ago
    It actually is. The design is just a small part of the control's cost. Then it has to go through a lot of compliance tests (the controls can't be flammable, can't break easily into shards, and so on). You have to do those tests for every car's interior part. When you squeeze everything into a touchscreen, you have to make the whole screen compliant only once.

sizzle 3y ago

Awesome was this class at UCSD in the cogsci department?

georgeecollins 3y ago
Yes

robomartin 3y ago

> The only good reason is to save money on buttons and knobs. That’s not quite right. At scale the buttons and knobs are cheap. Having designed many control panels for a wide range of applications, I can tell you that good design is an excruciatingly long and painfully expensive undertaking. You cannot recompile your way out of a poorly working physical control panel or bad design. You can’t just throw something together for testing and finish it while other teams complete the design of the rest of the car, machine, etc. As a real example, you might have to evaluate hundreds of button designs for tactile, audible and visible feel. Once sub-selected, you might evaluate over a dozen full scale control panel mock-ups before design lock. Etc. Touch screens allow you to eliminate a huge portion of what goes into creating a user interphase.

georgeecollins 3y ago
Good to know! I always thought it was to save money.
- robomartin 3y ago
  Well, it does save money…just not necessarily due to hardware cost differences.
- Summershard 3y ago
  It actually is. The design is just a small part of the control's cost. Then it has to go through a lot of compliance tests (the controls can't be flammable, can't break easily into shards, and so on). You have to do those tests for every car's interior part. When you squeeze everything into a touchscreen, you have to make the whole screen compliant only once.