But we're not talking about GM forcing you to use their own system. We're talking about GM forcing you to use Android Automotive ("Google Built-in"). They are ditching their own system.

There's no fee for the service. There's no fee for the basic cell service, and my car came with 8 years of the higher tier service (that allows streaming), but you can tether the car to WiFi from a phone and be able to access those features.

GM DOES push OnStar still. In my car, there's an OnStar safety package thing (which is free for 3 years?), and you need to pay some sort of fee to use the built-in cell modem as a WiFi bridge after 1 year (IIRC). I believe that after 8 years, there's a fee for certain services (namely audio streaming).

I think pretty much every car company is now charging to use the cellular modem in some way or another. Typically the basic services are free, and "premium" services (real-time traffic and weather, and streaming services seem to be key). In the case of my car, GM charges to you enable the WiFi hotpost the same way my cell-service provider used to (but no longer does).

Google. We're talking about GM switching to Android Automotive, so GM only updates car-specific drivers. The OS and apps are all Android and updated by Google.

GM isn't moving to a proprietary infotainment system, though. CarPlay and Android Auto are third-party UIs designed to be hosted on top of a proprietary OS. GM is moving away from that to the Google Automotive platform -- same as Polestar, Volvo, Renault, Honda, Nissan, Ford, and Mitsubishi.

It may or may not be obsolete in a decade, but GM is explicitly moving from their own proprietary platform to the Google built-in platform. That's what people are complaining about. They are effectively turning over the infotainment to Google and dependent on that third-party's interest in maintaining and supporting it; Google might decide to sunset Maps, Google Assistant, or their automotive platform in that time.

People seem to think this is GM using something proprietary designed by them. It's not, though. They are moving to Google Automotive, which is effectively Google Auto, but instead of having a proprietary OS in the car have an app that hosts a UI exported from a third-party device, they are just running the Google Automotive version of Android on the car so no third-party device is necessary and they rely on Google to provide the infotainment platform.

The Pros: doesn't require a phone, more robust, can integrate directly with car hardware; the Cons: crappy Google Play support so far (no games or video streaming apps for when you are sitting in the car waiting for someone / recharging), limited options for customizing the UI.

I have a car with the new system. If you've ever used Android Auto, this is pretty much the same experience but more robust. The Google Maps navigation works very well. I installed Waze, and it had a hard time logging in at first. The UIs for many of the apps are clunky, but consistent with Android apps in general (mostly based on my experience with Samsung devices).

You really can't do it without regulation. For an electrical grid to work, there has to be lots of regulations that assure the power generation meets standards that make the power usable (including phase, frequency, preventing surges and brown outs, safety, etc). Electrical grids simply cannot function without regulation, at least to set and enforce standards.

That said, existing regulation is not so onerous. Anyone can stand up solar (provided space), wind (provided space and zoning), at fairly low cost. Hydroelectric is tough as it requires rivers, and damming those up affects lots of people. Nuclear is complicated for fairly obvious reasons, and burning "stuff" spews out harmful gases (and regulations impose reasonable limits on those; they probably should require capture and sequestration since the air pollution incurs huge costs that are passed on to consumers and non-consumers alike and in equal proportion).

FWIW, the Model 3 is considerably more efficient than the Model S. More like 350 miles on an 82 kWh battery, but the point is well-taken. Charge times are typically given for 10-80% (70% of the capacity), so 700 miles takes 164 kWh battery x 0.7 = 114.8 kWh charge. In 10 minutes, that would be 688.8 kW charging -- a LOT of power. That's not to say that there's not proposed megawatt chargers (mostly for trucks), but that may never be practical for passenger cars. Most people with BEVs do almost all their charging at home, and if you've got 10 kW coming out of that charger, then it's going to take a while to fill.

Large commercial vehicles use a different standard than cars: MCS.

Tesla has more >150kW chargers deployed than the rest of the industry right now. There are higher power CCS chargers out there, but relatively few. Tesla v4 chargers are 500V or 1000V with 615A. Max is 307kw for 500V and 615kw for 1000V. There are a few pilot sites, but these will be the replacement for the v3. Also, Tesla will not be the only supplies of NACS charging â" you still have EA.

NACS supports 1000V. Whether Tesla offers degraded charging to non-Tesla cars in the US at SuperChargers is unknown, but you wouldnâ(TM)t be limited to using the SuperChargers.

The question is who will have the most >400V chargers deployed by 2025 - Tesla or Electrify America? EA has the lead as theyâ(TM)ve started already (though have not deployed many) and Tesla has only a small number of v4 charger pilots.

Historically, Tesla has rolled out more chargers per year than the rest of the industry combined, so if they deploy more v4 in 2024, they should pass EA easily. They have a big advantage too in that their chargers cost about 1/3rd as much to install.

To be fair, the Tesla connector was rolled out >2 years before the first CCS spec was ratified. It was CharIN that introduced the incompatible connector.

The question is long, but mostly for the illustration, which makes it clear how it will be graded. I suspect the reason that so many got 0 points is that they used up all of their time on the preceding 3 questions and jut didn't get to question 4. That's a fault of calibrating the time to answer on the part of the test designers.

I remember in February of 2022 reading (Detroit Free Press, I believe), comments from GM to the effect that they expected to replace the Bolt with the Equinox in 2024 as part of their move to the Ultium platform. That was repeated in several other places in the first half of 2022. I guess people forgot, or that they felt there would be some overlap.

Touch displays make sense for a tablet, and UI's centered around that sort of interaction, but aren't particularly a good idea for the laptop form-factor. There's the ergonomic issue of holding the arm up and moving back and forth between touch pad, keyboard, and screen. There's the issue of finger prints. Most importantly, though, is that touch UIs have different considerations that non-touch UIs. Touch targets need to be bigger and consistently sized; they need to be laid out to accommodate the hand obscuring parts of the display during interaction. The touch UIs are designed not to use keyboard modifiers for the pointer interactions. The touch UIs also use up screen real-estate much differently.

I don't know that Apple has anything to gain from adding a touch screen to their Macs. For cases where it's appropriate, their tablets are appropriately suited for it.

This is, indeed, quite complicated, and right now most of the costs associated with it come from two things: the ad hoc nature of it (it's labor intensive and can only be done by a small number of groups), and licensing of technology (mostly various things related to CRIPSR). Patents on a lot of the CRISPR tech are already half-way to expiration, but there's considerable desire to monetize it ASAP - so somewhat reasonable licensing fees are expected.

With regard to the ad hoc nature of it... It needs to be reduced to practice. The software to perform the predictions needs to be validated. However, most of the technology used here is rather mundane contemporary molecular biology and could be automated with reagents packaged in kits. It would require companies to offer a service to do the analysis and prepare the cells, but the price will ultimately come down to how broadly applicable the approach is -- the volume of the treatments will ultimately set the price.

Keep in mind that whole-genome sequencing went from costing a few billion dollars in 2003 to less than $500 today through computer algorithm, automation, and advances in molecular biology.

The tricky bit here, though, is the regulatory hurdle. This is essentially a form of genetic surgery performed on immune cells. The closest thing we have to it is CAR-T. The regulators are still coming to terms with how to review, approve, and regulate these approaches.

I concur. Studio + decent 5K monitor is very pricey. I don't think that the M2 design would allow adding RAM slots, but it would be nice if they would add 2-3 user-accessible SSD slots, and move the SD card slot to the front or side edge (and maybe add a Thunderbolt port there too).