Really surprised that Autosleep performed so poorly. I've been using it for over a year and I'm pretty satisfied, and chose it because it was the only apple watch sleep app I could find that didn't require a monthly subscription. Admittedly I don't look at the 'score' so perhaps that's it, I just look at the overall 'quality sleep' hours and have found that tracks pretty well with how off-my-game I am that day.
Wow this is really cool. I just tried the example on the homepage, that's magic! No email, username or password. Can someone explain what is happening?
On iOS this seems to use the iCloud Keychain which is slick but how would I then login to sites using Firefox or any computer that doesn’t have access to my keychain? The reason I use a 3rd party manager is precisely this reason.
Typically for web authentication the websites rely on the browsers which by default will back into the platform.
But any level of that may take responsibility - for instance, 1Password and Dashlane replace the browser/platform support by default by altering the implementation of the javascript API via their Web Extensions.
There are ramifications to this approach, such as having to fall back to the browser/platform UX to support hardware security keyfobs.
The platforms (and browsers using their API) also support or plan to support a cross-device option, where you should be able authenticate within a desktop browser using your cellphone via QR code and radio proximity checks. The vision is that some websites will see that the location browser _could_ have supported authentication directly, and offer to help the user register it as a second (more convenient) option.
Sites should likely let you enroll multiple such passkeys from different vendors (add a Microsoft Account passkey from your PC, a Google one from your Chromebook, etc).
Apple already supports Keychain sync with Edge on Windows and I believe that already supports Passkey access.
Also, I believe I heard rumor that "Sign in with Apple" (their existing OpenID Connect account system) will also eventually support helping you enroll non-Apple devices to Passkeys in apps that support both Passkeys and "Sign in with Apple", though I don't know if there is yet a timeframe on that sort of support.
> Sites should likely let you enroll multiple such passkeys from different vendors (add a Microsoft Account passkey from your PC, a Google one from your Chromebook, etc).
This sounds good, except how would it actually work?
I register in on my iPhone, it uses a key kept on that phone/iCloud. I log in via Safari on MacOS and it works because of iCloud sync.
Now I go to login using Edge on Windows. How can the website find out that I'm the same user as the iPhone/Safari user since I can't sync my key, and I can't enroll my MS Hello ID (or whatever Windows uses) on my Mac or iPhone?
There is a cross-device system to sign in, using QR and proximity checks.
Once the user has signed in, a modality check shows that they logged in with another device, while a capability check shows that they _could_ have authenticated with the local device if it had been registered. This may trigger the site to prompt them to register the local device as a second mechanism (or they may just go to the self-service account management tab to do it themselves).
A new private-public key pair is generated, the public key is your user identifier (sort of), and the private key is stored on your device (browser or phone). You're logging in by proving you have the private key for the associated public key. I think the device may also be storing a mapping from key to service or something? Not sure.
From my loose skim, this seems to be more for UX than anything else: no-clicks account creation and no-clicks login, but there's still account creation and login happening, presumably with a key provided by BitWarden. But websites can start removing the login prompt as an entity to be interacted with.
Honestly, cctv is a bit pointless anyway.
My bike was stolen and the police had cctv footage from a neighbour but it was no good because they had their faces covered. Unfortunately, the best way to protect a bike is a stronger shed, or a car is fit a steering wheel lock.
My wife and I dislike both our cars. They're older than average, they're externally unimpressive, and they're internally unimpressive. So long as we remember to take anything of value out of them, we don't give a fuck if they get stolen. The only hassle would be the insurance paperwork and finding a suitably average replacement vehicle.
They also don't mark us as the type of people that have anything worth the effort and risk of stealing.
I'm currently on holiday and so we're driving a much more modern car than either of ours, and what I refer to as "the distance from the road" is noticeably further than our two cars.
What I mean by "distance from the road" is the number of layers between driver action and machine reaction. Changing gears in a manual is a direct, instantaneous (pending a crunch and grind) process from driver action to machine reaction. Pressing the accelerator in an automatic has always had a noticeable lag, for me at least, being raised in a manual. In this much more modern car, there's not just the auto-lag, there also seems to be a choice the car itself makes in what it feels like it's an attempt to be maximally efficient on fuel, in restricting acceleration. It really feels there's more a layer of software in addition to the acceleration auto-lag.
The end result feels like an unpredictable rate of acceleration as I increasingly convince the car to "fucking move you piece of shit" whilst attempting to enter traffic at a decent clip. The car ends up massively over-revving in a low gear/band and then almost skipping the next two gears to settle into the normal 60 - 80 kmph zone.
This "distance from the road" is bothersome to me, but may be (much?) safer for the majority of drivers who aren't used to being so "near the road".
Vale the manual car!
(If you can't drive a manual, you lack the concentration and skill required to safely drive any car on a public road. It was an appropriate and effective barrier to entry whose absence is a threat to every road user)
Another boon of the manual car is that you are probably not likely to have a lead foot and find yourself speeding. In an automatic with a lead foot, the gears shift for you and suddenly you are going 50mph, but you cant tell because the car put you in the overdrive gear already and you are only revving 1800 rpm. Electric cars are probably even worse in this regard because theres no sense of connection to the powerplant that you can interpret from the cockpit.
In a manual on the other hand, if you start having a lead foot, the engine lets you know. Once you are familiar with a given manual car's gear ratios, you don't need the dashboard anymore. You know what 3000 rpm feels like because (in a good drivers car at least) you can feel the engine vibrating through the pedals, through the steering wheel, and through the gear shift, in additon to hearing the exhaust note. You also quickly figure out what speed a given rpm gets you in each gear. Maybe 4th gear at about 2000rpm is your 35mph cruising gear. To go over the speed limit in this case you would have to rev the engine up which would be noticeable, or shift into 5th.
That still doesnt make you immune to theft. I had an older crappy car and it was constantly targeted for things like a $5 charging cable I got from the gas station with frayed wiring. It didn't matter if I had nothing in it, people will still rip through the glove box on the off chance I do. I feel like an older car is a mark in that sense because of how easy it is to get inside. With mine, you can activate the unlock button with a shoelace because it protrudes slightly, but the tools most thieves use these days (the air bag jack and the long rod inserted into the gap the air bag jack for hitting the unlock button) will work on all cars, and these tools are sold at hardware stores. After that 10 seconds it takes to get the car unlocked, there's no more risk of being caught stealing because then you look like you could very well be the car owner rooting around.
In the U.S. it doesn't matter what it is, people want it for the scrap value at the very least. Its like a bed frame's worth of scrap but you can ride it off versus having to schlep it.
There are a number of seriously conflicting stories. Some of them say "the cards were individually tuned and undervolted to run at maximum efficiency to make the most money so they'll be fine" and some say "these cards were overclocked and left to run in a boiling hot shipping container then they washed them off with a water hose".
But I used my Radeon RX 6700 XT for mining nearly 24/7 for about 10 months (between purchase and when it paid itself off), while using it for gaming in between (I'd obviously stop mining). It ran around 65°C during that time. Very low core clocks, but memory was run at close to the maximum recommended speed by AMD's Adrenalin software. At least so far no signs of any problems.
tldr: Cards (like any piece of other electronics) do have a lifespan, but mining doesn't affect that. Cards that are kept clean and in better working conditions will run faster.
My bike was stolen on Monday and I'm honestly still grieving the loss. A bike is more than just transport, it's freedom, exercise and a million times better at clearing your head than any other mode of commute.
I'm no fan of ads either but one undeniable benefit is equality; if Google cost money to use then those in poverty would simply be unable to use it. The service is paid for by the wealthiest (as these are the most valuable eyeballs to advertisers anyway).
