Not surprised to see a concrete example of the Tesla iPad-control being dangerous. I first test-drove a Tesla Model S like 2 years ago (some friends of mine were curious so we pretended to be rich...
Not surprised to see a concrete example of the Tesla iPad-control being dangerous. I first test-drove a Tesla Model S like 2 years ago (some friends of mine were curious so we pretended to be rich and called a dealer). It's an impressive car but wtf is that touch screen thingy? How is that a solution for a high end car?
My head was buzzing with things that could go wrong, I mean that thing clearly was running on some complex framework with more than a few failure points. And you lose all the haptic feedback of physical buttons so you actually have to look for every little adjustment. All the buttons needed to physically operate the car should be hard-wired, physical dials and buttons. The iPad is fine for things like music, nav systems or phone connection. But putting the defroster in a touch screen submenu is pure insanity. If they think that is a good idea, it makes me question the whole UX even more.
Indeed. From a comment from the reddit thread linked to in the post: The idea that you would need to reboot to keep your car running smoothly seems absurd, especially considering how much...
I mean that thing clearly was running on some complex framework with more than a few failure points.
Indeed. From a comment from the reddit thread linked to in the post:
After a 1.200 miles in the car yesterday, I loathe the V11 update. Beyond the usability issues (hiding the trip odometers and getting rid of tenths of a mile/km 🤬), I found that the UI slowly lags. Takes about 4 hours of driving, but there is clearly a resource leak.
It starts with stuttering in the AP display, then audio drops outs. When I pulled to an SC and tried to back up, 3 times the camera view appeared, but then the UI locked up. It takes several tense minutes to reappear from a black screen after a 2-button reboot.
The idea that you would need to reboot to keep your car running smoothly seems absurd, especially considering how much functionality is locked away in the UI. I can't imagine any other company deeming this road-worthy.
Yeah this is pretty much correct. The recent V11 update put this terrible UI into all of the cars that requires more clicks for common things than the previous UI did. It seems to be mostly hated...
Yeah this is pretty much correct. The recent V11 update put this terrible UI into all of the cars that requires more clicks for common things than the previous UI did. It seems to be mostly hated on Reddit, so hopefully they get around to actually making it better.
When I look at the Tesla touch console, I genuinely don't understand who would want to use it as opposed to just tolerating it because it's a Tesla. I get that I can get used to the touch screen...
When I look at the Tesla touch console, I genuinely don't understand who would want to use it as opposed to just tolerating it because it's a Tesla.
I get that I can get used to the touch screen just like how I get used to my phone's keyboard and I can roughly type accurately without looking and only get a handful of spelling mistakes. But with a car - a big lump of metal and rubber wheels that is hurtling down the road at very likely high speeds, a flat surface just does not compare to a physical buttons and surfaces where you can identify each button by feel.
It just feels incredibly dangerous to have a flat surface touchscreen used to mimic real physical buttons when driving, especially at high speeds. Another downside mentioned in this article is that Tesla can quite literally remove these software buttons as they please and put them in hard to reach places. Why would anyone want to have this experience? I can understand for phones and other devices it can make sense but for cars you really want the buttons to remain the same all the time and not change at all during the entire ownership of the car.
Or at least that's how I feel - I might be an out of touch fart who is too used to the "old ways" of vehicles where a software update on the car rearranges buttons that users could and would potentially use frequently.
I've mentioned before that I'm an automotive engineer for a major, non-Tesla, OEM and this specific issue is something I used to get in major fights about. I say used to, because now I work in a...
I've mentioned before that I'm an automotive engineer for a major, non-Tesla, OEM and this specific issue is something I used to get in major fights about. I say used to, because now I work in a different area of the same company.
Basically part of the problem is that when you interact with these screen you can't rely on haptic feedback like you can with analog controls, so you have to balance looking at the controls and looking at the road. Even purely physical controls create some cognitive load, but you want to minimize that.
Problem 2 is there was a big push towards simplification. That was about 5 years ago on my side, so cars that came out within the last year or two. Sleek designs look good and sell cars, and the team that does the control designs are artists. My team had some ability to push back, but management always wanted to know the absolute minimum number of physical buttons which is an extremely difficult question to answer. And management-types don't like the idea that just because something is hard to quantify doesn't mean it's not extremely important.
Which brings us to problem 3. I haven't kept up with that field, but at the time the metrics to try and simulate that sort of activity didn't exist. There were a few groups working on developing it, but most UI/UX testing took place in stationary test setups, rather than ones that simulate the cognitive load of driving. Which is an extremely important distinction. So it created a false confidence in how easy it was to operate those controls while driving.
So this is very much a hot button topic for me. I should see what the current state of the field looks like
Edit: The push for simplification was explicitly to copy Tesla. At this point Musk and Tesla were still very much seen as visionaries.
Edit 2: Becoming an engineer has really hammered in the truth of the phrase "regulations are written in blood". Sigh.
Not surprised to see a concrete example of the Tesla iPad-control being dangerous. I first test-drove a Tesla Model S like 2 years ago (some friends of mine were curious so we pretended to be rich and called a dealer). It's an impressive car but wtf is that touch screen thingy? How is that a solution for a high end car?
My head was buzzing with things that could go wrong, I mean that thing clearly was running on some complex framework with more than a few failure points. And you lose all the haptic feedback of physical buttons so you actually have to look for every little adjustment. All the buttons needed to physically operate the car should be hard-wired, physical dials and buttons. The iPad is fine for things like music, nav systems or phone connection. But putting the defroster in a touch screen submenu is pure insanity. If they think that is a good idea, it makes me question the whole UX even more.
Indeed. From a comment from the reddit thread linked to in the post:
The idea that you would need to reboot to keep your car running smoothly seems absurd, especially considering how much functionality is locked away in the UI. I can't imagine any other company deeming this road-worthy.
Yeah this is pretty much correct. The recent V11 update put this terrible UI into all of the cars that requires more clicks for common things than the previous UI did. It seems to be mostly hated on Reddit, so hopefully they get around to actually making it better.
When I look at the Tesla touch console, I genuinely don't understand who would want to use it as opposed to just tolerating it because it's a Tesla.
I get that I can get used to the touch screen just like how I get used to my phone's keyboard and I can roughly type accurately without looking and only get a handful of spelling mistakes. But with a car - a big lump of metal and rubber wheels that is hurtling down the road at very likely high speeds, a flat surface just does not compare to a physical buttons and surfaces where you can identify each button by feel.
It just feels incredibly dangerous to have a flat surface touchscreen used to mimic real physical buttons when driving, especially at high speeds. Another downside mentioned in this article is that Tesla can quite literally remove these software buttons as they please and put them in hard to reach places. Why would anyone want to have this experience? I can understand for phones and other devices it can make sense but for cars you really want the buttons to remain the same all the time and not change at all during the entire ownership of the car.
Or at least that's how I feel - I might be an out of touch fart who is too used to the "old ways" of vehicles where a software update on the car rearranges buttons that users could and would potentially use frequently.
I've mentioned before that I'm an automotive engineer for a major, non-Tesla, OEM and this specific issue is something I used to get in major fights about. I say used to, because now I work in a different area of the same company.
Basically part of the problem is that when you interact with these screen you can't rely on haptic feedback like you can with analog controls, so you have to balance looking at the controls and looking at the road. Even purely physical controls create some cognitive load, but you want to minimize that.
Problem 2 is there was a big push towards simplification. That was about 5 years ago on my side, so cars that came out within the last year or two. Sleek designs look good and sell cars, and the team that does the control designs are artists. My team had some ability to push back, but management always wanted to know the absolute minimum number of physical buttons which is an extremely difficult question to answer. And management-types don't like the idea that just because something is hard to quantify doesn't mean it's not extremely important.
Which brings us to problem 3. I haven't kept up with that field, but at the time the metrics to try and simulate that sort of activity didn't exist. There were a few groups working on developing it, but most UI/UX testing took place in stationary test setups, rather than ones that simulate the cognitive load of driving. Which is an extremely important distinction. So it created a false confidence in how easy it was to operate those controls while driving.
So this is very much a hot button topic for me. I should see what the current state of the field looks like
Edit: The push for simplification was explicitly to copy Tesla. At this point Musk and Tesla were still very much seen as visionaries.
Edit 2: Becoming an engineer has really hammered in the truth of the phrase "regulations are written in blood". Sigh.