Eh, for a toaster it at least makes some sense to have ‘smarts’ for ensuring various types of bread reach the desired crispynes (Lathough none of that needs internet or anything close to 400$ in value). So it would still be more usefull then smart fridge.
Yes, especially once you have been informed about it
Then that is not what the article is about…
How do you currently store your passwords? I would also consider that a third party with an adittional atack surface if you are considering the passkey location one.
Also your argument
(if you ignore the operating system, web browser, network protocols, etc., but that’s part of using the tech).
is faulty. That is because passkeys exist in part to mitigate those atack vectors. Mitm, a compromised browser or client, etc. is less of an issue with passkeys. The information transmitted during an authentication can not be reused on another authentication attempt.
I don’t agree on passkeys complicating things either. For me the authentication-flow is not more complicated then KeePasses autofill.
Assuming one can be ‘tech savy’ enough to not fall for fishing is bad. There are quite advanced attacks or you might even just be tired one day and do something stupid by accident.
What’s that now? The weak point is the user’s ability to implement MFA and biometrics? The same users who couldn’t be bothered to create different passwords for different sites?
You don’t expext the user to ‘implement’ mfa or biometrics. You expect them to use it. And most places where a novice would store passkeys don’t just expect but enforce it. It is also way simpler to set up biometrics on one device compared to keeping with a good password strategy.
You can set a pin on most passkey devices so that it doesn’t serve the authentication without it.
However, it is that seemingly arbitrary three-times-a-year limit applied to the People section that is most concerning. Why not four? Why not as many times as a user wants?
If it works the same way Immich does, probably because they have to retrain the models every time you turn it back on and want to aboid poeple turning it on when they need it, then off again, then back on, … Although a less shitty mehod would be to limit the amount of times you can turn it on but I guess that iterferes with their goal of harvesting your data.
Furi Labs runs a fully optimized system called Furi OS
If I were to switch to a Linux phone I’d want it to be made for an open and trusted OS, not the (unknown to me at least) manufacturers own.
You cannot just say that and then not link it!
I would think most standard consumers are not using HDDs at all.
The pool is about 20 usable TB.
Thats exactly what I wanted to say, yes :D.
I have 2*12TB whitelabel WD drives (harvested from external drives but Datacenter drives accourding to the SN) and one 16 TB Toshiba white-label (purchased directly also meant for datacenters) in a raidz1.
How full is your pool? I have about 2/3rds full which impacts scrubbing I think. I also frequently access the pool which delays scrubbing.
I am not questioning the need for more storage but the need dor more storage without increased speeds.
And I’ve only ever had WD hard drives and sandisk flash drives die on me
Maybe it’s confirmation bias but almost all memory that failed on me has been sandisk-flash storage. Zhe only exception being a corsair ssd which failed after 3 yrs as the main laptop drive + another 3 as a server boot and log-drive.
What is the usecase for drives that large?
I ‘only’ have 12Tb drives and yet my zfs-pool already needs ~two weeks to scrub it all. With something like this it would literally not be done before the next scheduled scrub.
Btw, you can embedd the image like that:
It will look like that:
Another issue also is differentiating between hidden requirenments, undocumented features (positive), bugs (negative) and developer choice made during coding of the legacy software which does not affect the program positively or negatively.
A LLM (or alternatively a tem of developers) could do a reasonable well job of repuducing all these features and behaviors in a new program. But at that point you have effectively just cross-compiled your code (and maybe refactored it a bit).
The real issue is in identifying essential and wanted features from all the features, even if you could magically extract all of them. And in most cases that requires a deep understanding of the target audience/customer, which I would argue no person (including the customer), let alone a LLM, just has without years of requirenment refinement.