bhauer

> Model 3 in Europe with the so-called FSD

No Teslas in Europe offer the FSD Beta program.

If you have purchased the "FSD capability," your car should be upgraded to include FSD Beta (or "release") at some point in the future. In the meantime, what you have is essentially Enhanced Autopilot with some bonus features like automatic lane changing on freeways.

Plus it's a U-turn situation on ancient version of FSD Beta from two years ago (see the dotted lines in the display, or the YouTube upload date for that matter) that did not support U-turns. And it's not necessarily clear it wasn't going to brake more abruptly if the driver had not intervened.

Obviously, the article is microservice apologia, but...

> They were able to re-use most of their working code by combining it into a single long running microservice that is horizontally scaled using ECS...

No, it was no longer a microservice; it became a plain service, as in SOA. It was no longer micro. That's the whole point.

They could have saved time and money had they just made the thing a plain service to begin with. This thing was not sufficiently complicated to warrant the complexity of starting with a bunch of microservices.

The article says many hot takes have missed the point, but I think what we're seeing here is an example of the two sides talking past one another since the author hasn't appreciated the opposition's arguments at all.

Tron Legacy is an amazing visual spectacle paired with an enjoyable story.

It's a shame that this site uses Vimeo embeds rather than Youtube. On Youtube, you can pause and use the , and . keys to advance and reverse by single frames to truly enjoy the wonder of the animation. On Vimeo, no such frame-by-frame scrubbing seems to be available.

Incredible but true. In fact, many comments on this thread could apply to that context as well. For example:

> On paper, the incumbent mega corporations [legacy ICE manufacturers] should have been able to come up with an iPhone killer [a Tesla killer] quickly, considering the resources available to them... What they missed was the inertia and the lack of agility in these big companies. They kept insisting on their current plans until it was too late.

As we know, the theoretical ability of the incumbents to fiercely respond has been imagined every year since 2012, and yet the legacies are still struggling.

Note that your data includes PHEVs, and is not limited to BEVs. While BYD is growing fast, they are not yet ahead of Tesla in BEV units.

Looking at the Cox Automotive data for Q1 2023 (not just January and February as Jalopnik selected), it's difficult to spin the competition as anything but a bunch of fighting over scraps.

Chart here: https://pbs.twimg.com/media/Fth_aAtXwBQDUvK?format=png&name=...

According to Cox Automotive [1], through the entirety of Q1 2023, Tesla's US EV market share is 62.4%. I'm not sure why Jalopnik elected to use data from only January and February, but given their editorial history with respect to Tesla, it's not especially surprising they cherry-picked. You can see the Cox Automotive data rendered as a chart in [2].

In general, EV market share is a misleading metric since a decline for anyone holding a 100% market position is inevitable. Market share will drift downward as the number of deliveries from competitors grows; it doesn't seem rational to expect Tesla to retain 50+% of the US EV market share in the long term. Spinning that gradual decrease in EV market percentage as a dire concern reeks of an agenda.

It's also bizarre to focus on Model S, when Tesla's Q1 press release [3] says that Model S and X units are in transit overseas. S and X are manufactured in California and shipped to customers worldwide. Some of those customers have been waiting two years for their orders to be fulfilled. It's clear Tesla finally prioritized some overseas delivers of those two models to satisfy their shockingly-patient overseas customers. Contrary to the Jalopnik narrative, even with a large portion routed to other countries, US sales of S and X still exceeded popular media darlings (and good cars, mind you) such as the Rivian R1T (outsold by Model X) and F-150 Lightning (outsold by Model S).

[1] https://www.coxautoinc.com/market-insights/q1-2023-ev-sales/

[2] https://pbs.twimg.com/media/Fth_aAtXwBQDUvK?format=png&name=...

[3] https://ir.tesla.com/press-release/tesla-vehicle-production-...

Same problem with our 2007 Acura TL. And we fixed it the same way.

Exactly. For example, GFA Basic, mentioned in the article, was quite fast, especially when compiled.

When I lived in the Los Angeles area, I commuted to work three miles—six miles round trip—as a pedestrian. I certainly buy the data shown here, though it feels like it may merely reflect the number of people walking versus the number of drivers. I suspect in the central-northern part of the country, there just aren't a whole lot of pedestrians out and about to be injured in the first place.

Back in Los Angeles, I was most surprised by the resistance I received from the local city council when I asked for additional protections for pedestrians at obviously-problematic intersections. There was one crossing on my route where the cross-traffic speed limit was 40 MPH, meaning the real-world speed was 45 to 50 MPH, sometimes lower depending on traffic congestion. Only the intersecting street I was followjng had a stop, and as a result, the 45 MPH opposing traffic also had essentially zero expectation of having to stop for a pedestrian.

I asked the city to consider modifying the intersection into an all-way stop.

They promptly rejected the idea and instead simply painted thicker lines on the crosswalk. Thicker lines. That was what the city thought would save me and other pedestrians from cars traveling ~45 MPH on a street where no one ever expects a pedestrian to be crossing. Needless to say, I avoided crossing there for all the years I walked to work, even though doing so made my route less convenient. They claimed it had something to do with nearby train tracks, but that argument did not hold any water.

So it comes as no surprise that California has high pedestrian traffic fatalities.

Maybe I am misremembering, then. Perhaps I had a 9600 bps HST.

You might be right about the USR 9600 bps mode, but for certain their original 14.4k bps mode was proprietary and named "HST." I still fondly remember when I got my Courier HST and witnessed its intense speed boost versus my prior 2400 bps modem.

Fantastic.

The Atari ST was such a beautiful machine for its time. It still looks great today, with those diagonal lines and that crisp gray color, among a sea of beige peers.

When GP says "the update could be displayed on-screen," they are referring to the car's user interface, and not a secondary device like a computer or cell phone.

Incidentally, when you point out the issue of missing a theoretical electronic version of the notice because of spotty Internet connectivity, in such a scenario, they wouldn't be able to receive the OTA update either.

For Tesla's collection of data on accidents with Autopilot technologies, they considers Autopilot to have been enabled during the crash if it was on several seconds beforehand. These systems may hand control back to the driver in crash scenarios moments before impact, but that doesn't mean the data will report that the system was not enabled for the incident.

As posted on the Tesla subreddit [1]:

> Remedy: Tesla will release an over-the-air (OTA) software update, free of charge. Owner notification letters are expected to be mailed by April 15, 2023. Owners may contact Tesla customer service at 1-877-798-3752. Tesla's number for this recall is SB-23-00-001.

[1] https://www.reddit.com/r/teslamotors/comments/113wltl/commen...

You can see for yourself: https://www.tesla.com/model3/design

    Full Self-Driving Capability
    
    $15,000
    
    All functionality of Basic Autopilot and Enhanced Autopilot
    Traffic Light and Stop Sign Control

    Coming Soon

    Autosteer on city streets

    The currently enabled features require active driver supervision and do not make the vehicle autonomous. The activation and use of these features are dependent on achieving reliability far in excess of human drivers as demonstrated by billions of miles of experience, as well as regulatory approval, which may take longer in some jurisdictions. As these self-driving features evolve, your car will be continuously upgraded through over-the-air software updates.

It is semi-autonomous driving software, so the use cases are what you might expect. For example, I have seen many Youtube videos where people are using it for supervised semi-autonomous drives. Some say that once you're used to its quirks, it's less stressful/easier to drive this way.

In the end, who knows? Who cares? If a customer says they are happy with the purchase, why should someone else tell them they are plain wrong? I can disagree, of course, but I should not tell them they're just wrong for being happy.

I don't tell people they're wrong for paying extra for a special paint color.

FSD take rate has always been quite low, even when it was available for a much lower price than today.

A system that doesn't have the capability to make turns at all will not make an error when making a turn.

By the same token, walking will avoid the risk of being in a highway accident. So by this logic, walking is superior to driving.

This subthread is about the people who system16 criticized for being happy with purchasing FSD. The Justice Department is not acting on behalf of that group of satisfied customers.

If you pay $15K today, you get FSD Beta. For some people, the beta is worth the price, even not knowing precisely when it will be more than SAE Level 2.

We are talking about the group of people who system16 criticized for being "happy with having paid for FSD." Other peoples' needs and wants differ, and characterizing those needs and wants as if they are plain wrong is arrogant.

Yeah, and if you pay $15K today, you get FSD Beta. For some people, that is worth the price.

Again, arguing that you know better about what other people do with their money is arrogant.

People will pay $20K for sport brakes or $10K for an unusual paint color.

Acting like you know better than actual customers of FSD whether the $8K, $10K, $12K, or $15K they spent on the software is reasonable for them is the height of arrogance.

Also, characterizing the current FSD Beta system as "cruise control" is delusional.

Not to be flippant, but this potentially has not happened because Tesla has demonstrated it's an unnecessary burden to building a reliable charging network. If you build a charging network intelligently (for example, by putting the user interface in the car, not the stalls) and are motivated to keep it functioning, there's no need for a regulatory agency.

Customers should be selecting vehicles that don't have to deal with the deliberate incompetence of Electrify America.

If I were a company still on the low side of EV volume (GM, Ford, BMW, Mercedes, etc.), I would be seriously in negotiations with Tesla to adopt NACS and join the Supercharger network, allowing our customers to avoid the debacle of CCS.

I consider any manufacturer that isn't making moves in that direction to be demonstrating an unserious attitude toward EVs. CCS is a total shambles in the United States and its headwind should cast doubt on any advertised optimism from manufacturers using CCS.

That is twelve years from now. Even assuming they stick to that plan, which I feel is a naive assumption considering how often they've failed to meet EV plans previously, they can easily continue poisoning the EV waters today and for several years to come.

It is very much against the prevailing HN narrative to have an optimistic view on autonomy in general, and Tesla's efforts in that space in particular. The approved HN narrative is that autonomous driving is far too complex for computers and cameras, and Tesla's FSD is endlessly crashing into everything. Oh and it's somehow more dangerous to be near a Tesla on FSD than any other driver on the road, data be damned.

> Want to get data from inside a device onto your computer? Nope, it’s not a mass storage device. You need to install another app, that will be used just for this one device, to sync things. The app inconspicuously asks for location access while at it, and eats 85% of your CPU at all times, even when idle.

This one has to be an iPhone and iTunes. Absolutely nonsensical levels of obfuscation to getting data on and off. iPhones can't connect to network disks on the local network, they don't appear as mass storage devices if connected via USB, and the software they require--iTunes--is an absolute cluster. If only Android weren't its own complete disaster for different reasons.