We can agree that the article is sensationalist and not particularly interesting. I personally don't get hung up on that because it's what I expect. Just like I'm not shocked when a used car dealer describes a vehicle in unscrupulous terms. It's what I expect because it aligns with their incentive structure.

But note that you did your own editorializing by using the word "marginally." You've staked a claim that the recall is insignificant. But that view is in direct conflict with the regulators, and it's their opinion that matters.

It is marginally; they indicators are very tiny bit too small. Regulators operate in a binary -- either something is compliant or not. Those ideas are not in conflict.

The size can be marginally different, while the impact may not be. Just like when material specs are only “marginally” lower and cause a part failure. While measurements are always continuous, failures often are not.

That's true in general. But in this specific case the impact is also continuous because it's a visual indicator.

Right, which is why there has to be a definite threshold in the standard. Otherwise, you can just keep making it "marginally" smaller, on down until it's a single pixel, and still claim it's okay. That would mean there's no standard, and every/any UI design is equally valid. Standards exist, in part, to avoid such rules-lawyering by having a consensus on what determines a valid design.

Of course no one is talking about making it continuously smaller. It was just one-time marginally smaller. Literally a tiny mistake. If it had been mistakenly rendered as a single pixel, someone probably would have noticed the issue internally.

That misses the point (again). The point is you need some threshold if you’re going to have a standard, or else it’s all subjective and there is no standard. So the standard is 1/8”, and you claim 15/128” is only “marginally” different so it should be allowed. There’s nothing to stop someone from claiming 7/64”, or 15/128” are also only “marginally” non-compliant. You either is or you ain’t compliant, there’s no “kinda pregnant”.

So if you think they should be safety standards, there has to be hard thresholds for compliance. And (again) the bigger issue isn’t this particular incident, it’s that it brings into question how they are managing all hard safety requirements if some apparently slip through the cracks.

I never said it should be allowed. Please quote me where I said anything like that. I just said it's so minor, the fix so easy, that it doesn't demand any attention from us. The regulators told Tesla to fix it. Tesla fixed it. Most people wouldn't even notice. End of non-story.

> it’s that it brings into question how they are managing all hard safety requirements if some apparently slip through the cracks.

Again there is no evidence for that. The fact that all automakers have dozens to hundreds of recalls every year means that this one instance from Tesla is entirely meaningless.