I don't understand this "[reading] his email" thing. I can't look away from the road for more than a couple of seconds without the car alerting me to pay attention while FSD is enabled. It's currently both watching for attention (eyes) and nagging on the steering wheel. Autopilot at least does the steering wheel nag every 30-60s. Failing to acknowledge these cause the car to pull over and start blaring alarms in the cabin. I'm not sure how "99% of people" can use it the way you're describing, especially for FSD.

I've personally never felt like I couldn't take control in <1s.

I assume your car is newer. I have a 2018 3 and a 2023 X, and the X will nag me almost instantly if it thinks I'm looking at my phone. The 3 however happily chugs along.

I set it up for a team of ~30. This makes a huge difference for mounting things like `node_modules` with lots of inodes. The difference locally is pretty astounding.

I recently did a 70 mile drive on FSD with no interventions from local suburb roads, to highway, to urban roads, to some odd roads near a zoo.

That being said, often times I've had to intervene going <10 miles on local roads because it's so cautious that I'm afraid other drivers may react poorly and cause an accident. Rarely, it has done some dangerous actions that I'd characterize as "bad driving." So no, it's not there yet and it is dangerous because of the rare situations, but it's also occasionally perfect and usually "okay."

I agree, many of my interventions are because:

1. It hangs at an intersection too long that the person behind honks at me

2. It refuses to lane change out of fear leading me to miss highway exits or turns (sometimes but rarer, it tries a turn from the wrong lane)

3. It just gets bumpy some rides with a lot of phantom breaking that can be dangerous at times (I’ve seen the phantom breaking reduce in V11 though)

I don’t think the standard fear of if I didn’t intervene I would hit so and so obstacle is something I ever experienced in my FSD, though I did start using FSD in the much later versions of FSD 10+ where a lot of kinks were ironed out

Apple Card is very convenient, but it's not the best in the rewards arena.

For example, CapitalOne SavorOne (3% for many categories) and Wells Fargo Active Cash (2% on all purchases). There are many others that are competitive for specific situations though, including Citi Custom Cash (5% in a single category). Most of these cards also offer an intro APR and/or bonus after $X has been spent.

Definitely true, but this means that you have to spend some time and effort on using the "right" card for any given type of purchase, and staying on top of any changes to your rewards as well, to say nothing of having to juggle multiple due dates and automatic or manual payments.

A flat reward on everything definitely has its appeals. (The Apple Card isn't quite that, though – paying online without Apple Pay only gives you 1% rewards.)

As I mentioned in the comment to which you're responding, I never get less than 2% from my Apple card, so the Wells Fargo doesn't seem better to me.

I'll look into the Capital One option, thank you!

Not to defend SBF or suggest he did anything correct, but isn't that _negligence_? Given that the factors of the negligence benefited their operations, it still may be charged as fraud and he may be guilty in court.

Couldn't anyone committing fraud just claim negligence? I'd imagine it is pretty common for people doing fraud to try to cover it with negligence, like not keeping records of money going "missing"

Weird -- why not make the fine $604 million (or whatever they truly should owe in refunds) and allow all but, say, 4 million to be reduced?

That would create a weird dynamic: is it more beneficial to the airline to issue refunds to passengers, or is it better to give money directly to the DOT, which, as most government institutions, is probably underfunded and will appreciate the "donation"? Cost for the airline is the same if both cases, but indirectly funding DOT might win you some friends there :)

It's as though nobody realizes extensions can be created or purchased by sketchy actors and that this is a huge security risk when the extensions request "all access to all sites." OK, so when setting up an account's username and password and are provided 2FA codes or recovery codes -- those can all be compromised. How can you know an extension is compromised? It's almost impossible to tell with certainty.

Things like "The Great Suspender" incident get ignored and folks assume no other extensions have the same problems.

And what in MV3 solved all of that? It still allows enough to do a lot of damage.

Regardless of that, at some point you have to trust software. You can't expect everyone to read every line of code and compile all the software by themselves.

Yes, I trust plenty of software and I'm not suggesting that extensions are bad in theory. Extensions being able to silently inject code and ownership to change at any time is a pretty bad security model. We can agree that there are _bad_ security models, right?

It's the sum of the parts in changes from manifest V2:

- no arbitrary code injection via executeScript, must be a file now

- no more remote code

- no more arbitrarily getting selected text or highlighted text on a tab

- declarativeNetRequest instead of intercepting requests

- explicit listeners on the page to help detect bad actors (vs just arbitrary JS running on the page)

Even ignoring ad blocking - with those rules stuff like Tampermonkey which is totally legit can no longer work with execute script. You also lost a lot of functionality by losing DOM, having to rely on the broken lifetime of a Service Worker instead of persistent background page .

It took Google three to four years to acknowledge that the community is completely correct with its criticisms, and that MV3 is garbage designed by people who have no knowledge on how people write extensions and which abilities they actually use.

They're finally adding features that should have been there years ago. The new scripting API which brings back arbitrary scripts, in a new form, the offscreen documents API, and hopefully they'll eventually implement limited event pages which are somewhat solving the background page lifetime and DOM issues (which are already implemented by Mozilla and Safari I believe). Obviously, everything was decided hastily in the last second so all of the features are supposed to be completed by "around" October 2022, just two months before the original MV2 cutoff.

I've also read some of the extension working group transcripts, it's pretty sad how Google/Chrome has no accountability and almost zero transparency.

>actors and that this is a huge security risk when the extensions request "all access to all sites."

sure but that's my choice, that's why it's an extension. Paternalism of telling me what to do with my browser is silly merely because something is potentially dangerous. The entire internet is potentially dangerous. Clicking on a link or installing a piece of software is dangerous.

You're an adult, make responsible choices about whose extension to install instead of demanding that Google strangle you with security policies which at the end of the day serves only one purpose which is to extend their control over the user experience.

These are a bunch of straw man arguments against what I said. There is a difference between clicking a link and an extension being able to read the contents of pages you visit -- like your bank records or credentials.

Some of these "choices" aren't actually _made_ by anyone. Even with trust of an author, if remote code is being used and a domain or server is hijacked, then the remote code could be replaced. It's a lose-lose problem for Google and not addressing this problem means worse security for casual users. The boogeyman that they will remove useful extensions is antithetical to their behavior so far.

I think the contention is that hosting is a service. Just as you are suggesting Facebook is offering a defective service (that you said should not be offered), then so are hosting providers as they similarly can't moderate granularly.