Certainly not true. UUIDs have structure to them, and variants. Trace IDs are just 128-bit numbers, with any further semantics (almost) completely non-standardized (some systems encode timestamps, etc). They slapped a “last 56 bits are random” flag (not in the ID itself but as external metadata, so not like UUID at all) later giving IDs just a bit of semantics, but it’s not a reinvention of UUIDs.
Okay, sightly more bits than UUID v4. The whole article is merely reasoning "why at least 128 bits are required", and if you smuggle some non-random data inside these bits the entropy can only drop, making it more vulnerable to collision, i.e. inferior to UUID v4.
Actually, because the birthday paradox has k^2 as a term, this is actually less true than you might think. Having a time component actually reduces the chance of collisions over the long run, albeit at a cost of reducing the number that can be safely generated in any given quantum.
If you consider a 128-bit random number, you effectively have 64 bits of allocation space before you are likely to get a collision.
If you devote 48-bits to time, which provides millisecond accuracy for 9000 years, you then have 80 bits of randomness, effectively giving 40 bits of allocation space per millisecond before you are likely to get a collision.
Instead of approx 2^64 allocations across all time before a collision, you instead have 2^40 (1 trillion per millisecond). That sounds like a poor deal, until you realise that the factor is only 2^24, or 16777216ms or under 280 minutes.
So in reality, reducing the random space and increasing bits that are guaranteed unique is actually a great trade.
And there’s a good reason for that, because UUIDs have additional properties. I don’t know if versioning, partial ordering, or stable references are useful for traces or not, but with UUIDs those could’ve been a possibility.
It sure does a little bit, but a) quality varies a lot - some courses can get you from zero to dos cervesas por favor, some are just poorly structured noise that has no chance of sticking in mind; b) doesn’t explain grammar (it’s an exception when it does), so results greatly vary on preconditions like languages you’re already familiar with and can relate - anything too foreign and you’ll have hard time trying to understand how those examples generalize.
Duolingo it got me just enough Spanish (with zero prior knowledge) to get around, communicate basic needs (like a caveman, sure) and understand simple instructions, all without putting serious effort to learn language properly (putting serious effort into it) but only casually, as a side task.
The missing bit is that compliance is for governments and business partners, not for any end-users. For the purposes of KYC/AML process, end-users are objects, not subjects.
Your coins frozen with no reason given even internally except for "machine said no" - no one gets any slap on the wrist unless you sue real hard, happen to win, and most likely that'll be just a scratch that won't be noticed enough to change any attitudes.
The Man sees that someone they don't like transferring their coins through the fintech company - that's what those companies are really concerned about, because it would be a punch in the gut the company will feel.
Thus, the incentives. Current social design doesn't punish for false positives (until they hit really high levels), only false negatives.
Maybe I won't have to be concerned about job security some years from now, when everything becomes FUBAR and companies will need a legacy systems expert/software necromancer to a) discover, spec and re-formalize what their machine-generated black boxes are doing; b) build comprehensible and maintainable systems; and c) be responsible for what happens in the process aka swear by my work. While (a) probably can be done by a machine alone, and (b) can be done by a machine-and-human tandem, (c) absolutely requires a human.
But the few years to come are going to be wild for a lot of folks out there.
I don't expect Coinbase to publish a "we're hiring everyone back" in 5 years from now, but I hope at some point media will spot those trends as they'll - I have no doubts - will happen, and propagate that tune.
The way I see it, it’s purely a terminology/nomenclature problem. Consciousness is whatever speaker decides to call so. When listener has a different notion, communication hits a barrier. Language works only because most have somewhat overall similar perceptions on semantics, and it’s easy for everyday stuff (and even then people can easily miscommunicate e.g. colors). Not that many think about nature of consciousness in any fine detail, so for most folks it’s just… a hand-wavy something humans have related to thinking and awareness.
And overhauling the language to match scientific understanding requires getting everyone onboard with that scientific understanding. Good luck with that, given that we have plenty of people who believe in weirdest nonsense.
Brain is not the only thing that makes us conscious, the whole human is a super-weird collection of highly intertwined systems that work together and produce whatever we call “human.” As I get it, it’s a huge complexity all the way down to that gut bacteria that somehow affects our thinking too. And I don’t think we have a vocabulary for all that - we mostly think of “self” as a single entity.
A way around the nomenclature problems is to try and look at experiments instead. For example Turing seeming frustrated by similar discussions on will machines be able to think that turned on how people defined 'think' and instead proposed his experiment where you can see if people can tell the computer and a human apart by asking questions by text.
I'm not quite sure how you'd apply that to consciousness vs thinking say.
Thinking about it consciousness no doubt got there through evolution, as did other aspects of animal life, and functionally, if you think about something like a deer, it increases its survivability to be aware of what's going on, where it is, if there's a predator near, should it run away etc. The inputs are billions of sensory and memory neurons and that info has to be filtered down to something like a situation summary so other parts of the brain can make a decision like whether to stay eating or run. I presume consciousness is basically the situation summary.
That’s not exactly clear as it may seem. At least I can trivially form counterpoints to both of those - not necessarily true but not obviously false.
LLMs “live” in token “space”, and it’s “aware” of all its surroundings in form of input. (Quoted terms for my lack of better words.) It has no other surroundings to be directly (not intellectually) aware of, just like we aren’t immediately aware about the physics around us.
As for the static nature - LLMs are trained and aren’t exactly static, they just get updates at different cadences, and we call those updates different names or, more precisely, versions. Plus LLMs can exist in multiple versions simultaneously - we can’t “fork” a human mind but it’s simple with LLM. Claude Opus (not sure if e.g. Haiku is related or parallel development with distinct origins) is like the proverbial Ship of Theseus in this sense. Either way it’s undeniable it learns and evolves, just very differently from biological systems, and ot all depends on how we decide to call things. Which isn’t exactly surprising, given it’s based on different principles and processes.
No, it doesn't learn or evolve on its own, but rather through humans and a huge data-gathering and training loop.
It's aware of the input, but anything resulting from it will be discarded by the next query or restart. Even with all the supporting systems, aka external memory, it can only access and retain an infinitesimally small amount of the information to which it was exposed — a shadow of a shadow of the things themselves.
If you argue that the entire system, including humans, is conscious, then that's a kind of tautology because humans are already conscious. If any, an LLM is only a partial reflection and low dimensional projection of this consciousness.
All true, but simultaneously, if you look at it not in the “how did we get here” but “what’s out there today” light, it’s an option that can run on a lot of platforms. Not by some particular merit but because history happened this way - but that’s not a problem with the technology itself.
The article defines "success" in the Windows context as being "available everywhere". It does not address how it got to that point.
And sure, you might not like Microsoft, and you may not like how it became successful (using the above definition) but the fact that it is available everywhere is not in dispute.
Of course most successful things have murky pasts. We don't necessarily agree with how it got there, but there it is. That is, at least in the technical sense irrelevant. You may prefer LP's or CD's, but streaming is now the successful way to get your music.
That doesn't mean it's the only way though, and of course you are free to not use Windows programs, or play games via Steam etc. That is your choice.
Yep, it pulls stuff from at least npm, it’s not a secret - check the source code.
Actually it pulls latest versions (checking registry then installing that exact version, not sure why they sidestep normal resolution algorithms) no matter what .npmrc may say, so min-release-age breaks almost everywhere it integrates with JS/TS ecosystem (most visibly, Copilot). I probably should’ve filed an issue.
It also installs Go packages but I haven’t looked into that.
reply