It's a pervasive myth that you cannot get around Haskell's purity when you want to. But although I prefer Haskell, I miss Common Lisp's ability to inspect the running data when an exception is raised.

Wait until you realize that the "classic" Haskell quicksort is nothing of the sort (pun intended).

Also I ran into more errors after this lol. I will work those out myself when I get the chance.

To follow up this, GHCi inferred "(RealFrac a, Integral a) => a -> a -> Bool" . I would like to come to my own definition by just thinking about it ....... but for now maybe I'll just do this lol

I didn't realize the usefulness of GHCi until you pointed this out, and someone else talked about the ":info Type" outputs. I didn't know I could use GHCi in that way. I've never used a programming language with a tool that does that.

And since you are putting each vector on its own page, a vector with one element takes 4KB of memory, and that’s real memory, not just virtual address space. You committed 4KB of memory but are using only 8 bytes of it, wasting 99.8%. Your program allocated 500x more memory than it needs. (A doubling reallocator allocates only 2x as much memory that it needs, worst case.)

I want to make sure I understand You're saying rather than a function taking a function pointer and an opaque user pointer, it takes a pointer to a structure that contains as its first member a function pointer (to ease casting? or just arbitrary?), along with whatever else the user needs (since it's their pointer to struct they control what the struct is?)

May the 4th be with your data... until it's sold anyway.

So basically, if your neighbor does something stupid, you're getting a knock on the door. Great.

For real. Age verification is a minefield of privacy issues and technical challenges. Hacking that app was just a matter of time. It's almost like they didn't consult actual security experts.

That's a bit reductive. It's about the tech and algorithms, not "selling our faces" directly. Still, consolidation of biometric services is unsettling regardless.

Honestly, this is less surprising than it is frustrating. Internal auditing is probably the only reason anyone gets caught. How many DON'T get caught?

Right to know? Lol. We lost that battle years ago. Just assume you're on camera everywhere.

Shocking, I tell you. Absolutely shocking.

Another day, another breach. Guess my identity is already out there a dozen times over. What's one more?

Your phone IS haunted. By your past self's bad choices in photos. 👻

It's probably more for things like crash reconstruction data or perhaps even insurance liability. Not necessarily active 'stopping you from going places' monitoring, at least not initially. Though the data collected could easily be repurposed.

Honestly, it's probably not even about the tech. It's about securing the contract for 'development' and 'consulting'. The actual product is secondary.

Time to fire up the old NordVPN subscription again, I guess. My internet bills are gonna love this.

Oh good, another bill written by someone who thinks the internet is a series of tubes.

Yeah, because a cop on the street is gonna wait for your 'full-disk encryption' to kick in. They'll just take it and put it in a Faraday bag until they figure it out or get a warrant for your PIN later. 'Raising the bar' doesn't help when they just jump over it.

The interesting part is *how* they're sharing. Is it raw query data, aggregated usage stats, or something else entirely? And for what specific purpose? Training models, ad targeting, 'improving user experience'? The wording in these EULAs is always so vague, but 'personal data' is a strong claim if it's more than anonymous telemetry.