Major GM restructuring underway


Not anywhere near me. As I said, they sell it in California, but not in most of the US.

More on Toyota electric cars: why should they get into it now? They have a successful business in hybrids, and don’t need full electric cars at this time. The hybrid batteries are not capable of supporting full electric operation. Maybe as the market for electrics grows, they might go through the certification procedure to sell in the US in general. They certainly need new generation batteries like the Bolt and Tesla have.


Absolutely nothing. But that’s a completely different argument for only keeping a car for 3-4 years. Buying a car because you want to is one thing…buying a car because you perceive it to be cost effective are two completely different arguments.


This is the reason I sold my 84 GMC S-15 pickup in 1990 with only 90k miles. I was doing consulting work at the time and I estimated I lost well over $10,000 in lost wages because of problems with that truck.


This is exactly what caused the housing crash of '07/08. I guess the only good news is that consumers won’t be as much on the hook, because most cars don’t cost $250,000 and up, but when the big automakers go bankrupt (again) it’s gonna have some pretty serious ripple effects on the economy.

That’s true, but wages have not risen along with inflation over the last 30 years, and so the cars are more expensive on a cost-to-income ratio basis to the vast majority of the population.


As far as I can tell that’s incorrect. Lower-income wages have been flat, but not dropping when adjusted for inflation:


An unfortunately that’s the vast majority of population earns under $50,000


Not the vast majority. Looks like about 50% from the above plot.


Here’s the median income:


Yes. About 2/4 below 50,000 and 2/4 above. That’s half and half.


There are thousands of combinations of phone hardware and operating system versions. Car makers put a lot of effort into determining compatibility but it is a daunting task. Most car manufacturers will publish a list of which phones and which operating system versions they have verified are compatible with their hardware and software. I’d be willing to bet it’s more of a testing oversight than a conscious decision to ‘crash the dash’.

Right. But again, car makers do not support every single version of phone made. Look at any of them. My Honda has a specific list of phones and operating systems it supports. It’s not just Bluetooth hardware interface and Bluetooth communications protocol. The car’s software needs to play along as well. We had a phone that wasn’t supported by Honda. I had to update it if I wanted to use the feature.

Well, I certainly agree that it shouldn’t take down anything critical to operating the car as a fundamental feature. But if the Infotainment console goes black because of a software crash due to a device connected to it, that’s unfortunate but unlikely to be an immediate safety issue. Again, if done by design, I would say that’s pretty egregious fault mode to experience. If it’s an implementation bug due to some strange interaction with that particular phone and it’s operating system and not approved for use then that’s a real bummer but not indicative of malice or indifference on the part of the BMW designers. Unless you have access to some inside information that says they did this purposefully.

Das machine und ekwipment ischt nicht fer gewerken by das dumkophen. Ist easy schnappen der springenwerk, poppencorkin mit spitzandsparken. Das rubbernecken sightseerin keepin hands in das pockets relaxin and watch das blinkenlights…


In other words (and this is using newer statistics from the Bureau of Labor than the article used because the article is older), you need $8 today to buy what $1 bought in 1965. Most people who are not CEOs have not seen 800% wage increases in that time span, and so things are indeed more “expensive” today when considered alongside actual wages. Cars are actually a little better - their prices haven’t risen as fast on average as other things we buy, but it’s still bad especially when you consider that the other things we buy still need to be bought and therefore cut into the pool of cash available to buy cars.

The average wage in 1965 was a little under 7 grand a year, which works out to around $3.50 an hour. You could easily earn that as a high school graduate. The average wage today for college graduates is just under 20 an hour, which works out to about $40,000 a year. This, of course, means that many of those average wage earners are paying off tens of thousands of dollars in student loans that the average high-school-graduate wage earner in 1965 did not have to concern himself with.

The original MSRP of a Ford Galaxie in 1965 was $2,730.

The MSRP of a 2018 Focus sedan, which is probably the closest equivalent to a Galaxie, is $17,950.

If you were making 7 grand a year in 1965, and you bought a new Galaxie, it cost 39% of your income if you were making the average.

If you’re making 40 grand a year today (which is somewhat misleading because that average counts CEO salaries which skew the data higher, and CEO salaries have increased by an astonishing amount since the 60’s) and you buy a new Focus, it costs just under 45% of your income.

And that’s not factoring in the idea that today in many markets, a $250,000 house is considered a starter home and apartment rents are well north of $1,000 for a 1 bedroom dump. It’s not at all unusual for people to have to spend 50% of their income on housing - a big change from the old days - which means that 45% to buy a car stings even more because after you pay for the other stuff you have to buy, you have even less left over to put toward a vehicle.

I never said it was a conscious decision to crash it. I would, however, say it’s a conscious decision not to care if the dash crashes due to use outside of the parameters intended.

Yes, my Acura is the same way, and mom’s ancient flip phone isn’t on that list. Seeing as that’s the phone that crashed her BMW, I paired it to my car for fun. Know what happened? Nothing, because the car refused to pair with it. That’s how it should work; if pairing with the phone will cause problems, then pairing should be rejected

I mostly agree, although a lot of drivers tend to get distracted easily, and trying to figure out why everything to your right just went black is a real good way to be distracted. But no, it’s not going to send the car out of control, and that’s great, except that it goes toward my point that, not being a strict safety issue, it’s an engineering decision that can and does get past regulations requiring safe engineering.

Red herring. It’s already stipulated that the phone isn’t going to work with the bluetooth module. That’s fine. It’s even, sort of, fine if the bluetooth module crashes when trying to deal with the phone. What’s not fine is that the bluetooth module crashing causes collateral crashing. Other car makers have managed to figure out how to avoid the situation entirely. I would expect The Ultimate Driving Machine™ to not be worse than other cars.

It doesn’t matter. Hardly any engineering screwup is on purpose. Even the deadly ones. An airplane mechanic once accidentally used the wrong screw to secure the plane’s windshield. The windshield blew out when the plane was at altitude, and sucked one of the pilots halfway out the window. They couldn’t get him back in until the plane landed. It’s a miracle the guy survived.

That was entirely accidental on the mechanic’s part. He did not set out to almost kill a pilot. But it happened, and it’s because he screwed up.

I’m not accusing BMW of intentionally doing bad things. I’m accusing them of not caring if bad things happen as a result of people using the car in ways they didn’t specifically dictate.

This is funny. I’ve seen it before. As an engineer I’m sure you know that this joke is meant to be internal to engineers/IT professionals/etc, and the sentiment it expresses should not make it into design, because you know that your designs are going to be used improperly by idiots and smart people alike, and you need to at least make an attempt to keep the mess to a minimum when it happens.


But how do you know that? You’d have to test every single possible combination to know that not one of those would result in a crash. That’s what I’m pointing out. It could be (bad) luck of the draw. The ones on the list are definitive. They have been tested and shown to work. If it’s not on the list, then it may or may not work and the results are not guaranteed. Although it would seem possible to prevent a crash and abort gracefully, there are always instances of non-gracious exits when the number of possibilities is so high.

What I’m saying is some things are purposeful and others are “screwups” as you put it. If they said, “know what, we don’t care what happens if you pair a non-qualified phone, that’s your problem” is a lot different than “we made our best attempt to exit gracefully if you attempt to pair a non-qualified phone but results are not guaranteed”.


Actually, the data I posted show that this is incorrect. The average inflation-adjusted income for each group of earners has been flat for the lower 3 groups, increasing for the upper groups.


Here’s one site that says differently

But others are aligned with the 50%.

In many parts of the country like Boston area - under $100k is very low-end middle class.


No he doesn’t. Just has to know that SOME don’t crash.

I disagree with that. You don’t have to account for every combination. The software just has to account for the pairing it knows about. I’ve designed many many systems…with software technology and methods today you can easily prevent crashes. Ever hear of try-catch.

I have to disagree with that. These are NOT complicated systems (at least some systems I’ve worked on with millions of lines of code). Doing a graceful exit is NOT rocket science. It’s following the right coding standards and design patterns.


I haven’t done any coding since I was a kid fooling with BASIC and LOGO (and I started learning Assembly until I figured out how insane that particular rabbit hole is).

But even in the stupid piddly programs I used to write, it was trivial to include an error section that basically said “if something goes wrong, stop running the program.”

You could even get fancier and say things like “if something goes wrong, start the program over,” or “if something goes wrong, tell the user that something went wrong and go back to the step you were on before things went wrong.”

And it was a good idea to do that because the alternative was “ignore anything that goes wrong and keep going until things get so bad that the whole system freezes.”

It baffles me why I could pull that off plunking away at BASIC on my Commodore Vic 20, but BMW’s highly-paid software engineers couldn’t manage it in the bluetooth code. “If a compatibility problem occurs, reject the connection.”

And at any rate, when a program would error and stop working, it did not shut down the TV the computer was connected to. That wasn’t due to any particular design intentions, but it’s a feature of the fact that the TV was not daisychained to the computer.

In other words, if you have a string of lights, and the first light sends power to the next light, and so on, they’re daisychained together, and when the first light burns out, they all turn off and the room goes dark.

A smarter design is to run power to each light individually so that when one burns out, all the others still have power going to them and you don’t lose everything because one light (module) stopped working.


Mike, you can’t know for certain none will crash if you only know some will not.

Sure, quite familiar with that approach. It doesn’t always work. Fourth quadrant- things we don’t know that we don’t know. Ever hear of that? In the real world, stuff happens we didn’t anticipate or discover in testing.

I’ll refer you back to the fourth quadrant. If it worked as well as you say, there would never be any issues- but there are. Why? Because we’re human and we can’t envision or test every possible combination. Unless you’re the government with unlimited budgets and even then… schmidt happens!

I happen to write code for many years for safety and mission critical applications. I am very familiar with all of the mitigations to minimize fault occurrence and recover as gracefully as possible in the event of an unforeseen condition or situation. I am not arrogant enough to think my software can never fail in ways I did not anticipate…now downgrade the severity of the situation to a phone pairing and the effort put into testing every possible condition to verify and validate definitively that it works under all conditions- ain’t gonna happen…


And yet, the operating system could still hang up and stop responding to keys. There is no catch all for every possible scenario. There are many things you can do in coding to try and mitigate all possible scenarios and then one comes along no one thought about or ran across before and all heck breaks loose.

I have a saying, in every code set there’s a bug waiting to be found. The worst programmers are those that think it can’t happen to them…


100% certain - probably not…99.9999999999999999% - Damn right I can.

I’ve designed systems that have been running for several years that have NEVER CRASHED.

A crash is caused by not handling an unexpected situation gracefully. The software technique is called exception handling. How you handle exceptions determines if you system will crash. If you’re not doing exception handling then yea - sure you’re going to get crashes. That’s were good software design and implementation come in play. A simple software system that’s in a car’s audio system - if you can’t handle exceptions gracefully in that simple system then you have no business writing software. Systems my company designs and implements are hundreds of thousands times more complex and they don’t crash. There are grave financial and security penalties if they do. We take every precaution to ensure they don’t.

Because they’re not following best practices…that’s why. I see it all the time. We’ve bought 2 companies in the past 10 years for their technology. Both companies were NOT following best practices. And both companies had problems with reliability. They had a product that was needed, but were loosing sales because of problems. Within a year after buying those companies we had good solid products added to our solutions.


$50,000 in NJ doesn’t even get you into the middle class.