There's a world of difference between being overly optimistic about how easy writing self-driving code will be, and "nobody bothering to try basic functionality at all"
Yes, this is a trivial point -- i.e., I agree -- and yet has nothing to do with the paragraph you cited from me in conjunction with your response. I don't believe I claimed that "nobody bothered to try basic functionality at all" and if I did, then I certainly retract that claim. I think you read all that I had written, digested it, and it seemed to you as if that's what I said.
What I wrote was, "some poor design decisions".
As I just mentioned, Model 3 does not use "rain sensing algorithms". It uses a dedicated rain sensor. A piece of hardware specifically for detecting rain, same as other vehicles use. This is different from the S and X which did it with cameras (a much harder task to get right).
Yes, my bad, I conflated sensor and algorithm. However, you appear to be assuming that a dedicated rain sensor somehow replaces a "rain sensing algorithm". But there's always an algorithm of some kind going on in the code monitoring the sensor, plus whatever hardware and/or software algorithms may be in the sensor itself. The result may or may not be an improvement.
Not being on their engineering team myself, and quite possibly not being as familiar as you are with the content of Model 3 and/or engineering processes (although I do have a bit of an engineering background), I have to infer that you are comparing Model S/X as "sensor=camera + monitoring algorithm" versus Model 3 as "sensor=dedicated rain sensor + monitoring algorithm".
So I would say to you in return: (1) poor design decision to use camera as sensor in Model S/X, as possibly evidenced by (2) switch to dedicated rain sensor in Model 3.
Windshield wiper behavior is a basic drivability issue.
That's actually a much more difficult issue. You can't just hang a duct from the ceiling; you're not driving a warehouse.
For the sake of this discussion, please assume that I am familiar with ceilings, warehouses, ducts and automobiles and need no further education on the distinctions between them.[/QUOTE]
It's highly probable that the engineering team specifically decided during the design phase that it wasn't worth the sacrifices (increased hardware cost, increased manufacturing cost, reduced space, worse aesthetics, sacrifices to aesthetics, or whatnot) versus the level of added satisfaction it would give the percentage of people who actually have children ride in the jump seats.
Yes. IMHO, this is called a "poor design decision" by people who disagree with the result. Having had my children in the jump seats, and having had to remove them after 5-20 minutes due to overheating and poor ventilation, repeatedly, over a period of years, I am qualified by training and observation to render my opinion that this was a "poor design decision". They got the tradeoff wrong. And the affected individuals are ages (roughly) 2-11 years. Even with heavy tinting on the rear hatch window, the usability of the jump seats are vastly reduced. That's why on my second Tesla (Dec '16), I did not repeat my decision to purchase the jump seat option. Even though my children have aged out of the jump seats, I would have otherwise gladly purchased the option ANYWAY simply for the pleasure of including OTHER children of appropriate age and height.
This one is actually fairer. But if you think "grab handles on the ceiling" is on par with "basic driving usability", you're kidding yourself. And again, I strongly suspect that it was debated by engineers, and that there was a reason why it wasn't included. There almost always is.
OK, I will agree that grab handles are not "basic driving usability".
I will agree that it was a decision likely debated within the relevant team.
This is still a poor design choice.
That's not how it works. As I'm not o the engineering team, I can't say the reason they were omitted. Just to pick a random example: perhaps where they'd need to go didn't have enough structural strength (on an aluminum panel rather than a structural beam, or a beam that can't take sideways torque, or may not be penetrated or welded for mounts), for example, so to add them you'd have to add an unjustifiable amount of structural mass versus the benefit. These things are rarely so simple.
Yes. I am aware of "analysis" and "tradeoffs", also "architecture", "design" and "implementation", along with "verification", "validation", "quality assurance", and so on. When I don't like the design decision that results from the process, I call it a "poor design decision". I am entitled to do so even in the absence of complete information about the process.
To quote XKCD: "You can look at practically any part of anything manmade around you and think "some engineer was frustrated while designing this." It's a little human connection."
For the purposes of this discussion, please assume that I have a minimally adequate knowledge of a branch of engineering.
None of this is, again, even remotely comparable to basic driving functionality. Not even close. You're talking minor inconveniences. Nobody is going to decide, "let's trade off the ability to operate windshield wipers in order to save $0.50 by buying a cheaper wiper stalk"
This discussion began with my concern about the impact of wiper functionality on drivability by my 80-year old parents. If they can operate variable-speed wiping behavior without having to refer to the center console and take their eyes off the road, then I'm a happy camper. If they instead have to hit a button on the center console -- which requires eye-hand coordination due to the lack of tactile feedback for the control involved -- then it's a serious drivability issue from my standpoint that is causing me to ponder whether to purchase a $45K car for them. They will have to take their eyes off the road while operating with some level of precipitation. Not a good idea.
In closing, let me point out that the tradeoff GM engineers evaluated for the key switches used in a number of their products over a period of years in the 2000s was something around 50 cents. If you have followed that saga, you know that a number of fatalities resulted from that design decision, which from the public information I have seen was indeed cost-based. I can imagine that the engineers involved didn't see the key switch as being a basic drivability issue until some number of people died because their engines shut off suddenly in traffic. I certainly wouldn't have imagined it to be an issue of such severity; it's a humbling realization.
Alan
