At some level, it does. When these laws were passed, I think the main goal was to stop companies from charging exorbitant fees to other companies' cardholders when using their ATMs to withdraw money and similar. You wouldn't want to do that to your own customers, because you'd lose them as customers, so why would the government regulate it?

But now, years later, in the context of retail sales, card companies want the opportunity to charge higher fees to retailers so that they can give rewards, confident that the retailers will charge customers the average fee rather than passing on the exact fee that they get charged on a per-customer basis, and as a result, that jacking up the prices for your own cards won't cause you to lose customers.

The truly perverse (incentive) part is that customers are forced to chase the higher fees or else they're paying more as the stores up their prices to the newer, higher average.

The only sane way to solve it is for cash-back rewards to be banned outright, with the only allowed exception being retailer-issued cards that grant cash back exclusively at their store by using their own low-fee network (e.g. airline cards, Amazon cards, Apple cards). Because those retailers have every incentive to keep prices low, this very narrow case doesn't have that problem. But even those shouldn't be allowed to give cash back everywhere.

Learning multiple languages in most cases has nothing to do with parenting or nutrition or leaning habits. Most people who learn multiple languages learn them because they live in communities where multiple languages are spoken.

Depends on how crowded the market is. If there's healthy competition, they might, or one of their competitors might, forcing their hand.

Multilingualism is weakly correlated with intelligence. It's strongly correlated with growing up in an area where people speak many languages.

Well, this data says that your best bet is to have learned a lot of languages as a child. The study found that earlier acquisition of multiple languages was more effective than learning additional languages later in life.

Yes, it is "all that different". You're comparing conceptual similarity, not data similarity. Data (and usually) appear similar without being conceptually similar, or appear different while actually being conceptually similar. A dolphin, a lungfish, a coelacanth, a tuna, a shark, and a hagfish may all appear visually to be the same sort of "thing" (fish), but they're all actually radically different things. What you actually want to be comparing is the core conceptual aspects of things, not their outward (raw data) appearances.

Oh, and I forgot to include the "why". TL/DR: it just works better.

To form a latent, the network fundamentally has to learn correlations, in order to be able to accurately compress and then decompress the information in its input space. Due to the limited size of the pinch, you simply can't store every possible data-space representation of a given input. The encoding down to these vectors must persist despite the sort of "real-world" variations that occur in the input space.

Within the latent space (latents are vectors), you can think of individual subconcepts as being directions with in the latent space. In the "king" / "queen" case, for example, adding "man" times some scalar shifts you forward and backwards along degrees of masculinity. In the case of a spectrum, these vectors represent various fundamental molecular properties of the material being scanned. By computing the cosine distance vs. known entries in the vector db and picking the most similar entries, you are in effect minimizing the distance between the molecular properties of the samples.

By contrast, you certainly CAN minimize distances in data space! The problem is that data space lacks correlations. Two things being physically similar in data space doesn't actually mean that much if the things that are different are extremely definitionally-critical aspects. By contrast, definitionally-critical aspects inherently become a key part of vector encoding. Trying to work in text space, for example, a distance algorithm might declare that "king" and "kingdom" are textually similar, but "queen" does not appear at all similar to it. But it's fundamentally similar in vector space.

For YHVH's sake, does nobody in this thread know what a vector DB is?

Translating that to signatures for chemicals is basic spectroscopy which has worked for decades.

The emerging standard for spectroscopy is to translate the spectrum into a latent vector and match against vector keys. For example, Spec2Vec.

This is not what you may at first be thinking of when you see vector, e.g. "the spectrum itself is represented as a vector". These vectors are latents, not spectra.

The most common way to create a latent is via an autoencoder. You task a neural network with recreating its input and put a "pinch" in the middle of the network. This forces the network, to maximize its predictive accuracy, to create a dense conceptual representation of the concepts at hand at the pinch: a latent. The left hand side of the network becomes a latent encoder (detecting concepts in the input space and encoding them into the latent space), while the right hand side becomes a latent decoder (translating the latent space back to data space).

(Note that Spec2Vec is a bit different - it's based on Word2Vec and treats the peaks and neutral losses as "words" and the entire spectrum as a "document", seeking to learn the correlations between peaks - but this is getting out into the weeds)

One of the properties of latents (in most latent spaces) is that they let you interpolate concepts (for the classic example, the latent for "king", minus the latent for "man", plus the latent for "woman", resembles the latent for "queen"), and they let you measure the conceptual distance between concepts (cosine similarity)

Once you have a trained encoder and decoder, a vector database can handle vector queries. The ideal situation would be that all rows would be keyed with vectors and the cosine similiarity to your query key is calculated vs. the table keys, but since that's not practical to query every row, Approximate Nearest Neighbor algorithms are used to approximate that. Hits matching a desired minimum level of similarity are returned. Note that the decoder is no longer needed, you only care about encoding the latents.

Search engines widely use vector databases, BTW, though they usually also combine them with text matching ("hybrid search") - the latter is fast and cheap, so it doesn't hurt to add it.

It's not "fuzzy matching", it's a vector database. You're looking up with conceptual keys into conceptual-indexed tables. The keys are latents, not fuzzy string-matching rules.

Um, no. AI was founded as an academic discipline in 1956, at the Dartsmouth Workshop. The term has been in widespread use for seven decades.

Also, AI != ML. ML is an AI technique. Most AI today is ML, but it wasn't always that way. Symbolic AI, Rules-Based AI, etc used to dominate before NNs advanced enough to push them aside. Expert Systems are a classic example of non-ML AI.

Digging more into RxScanner:

As far as I can tell, 100% of the information out there about its claimed efficacy comes from the manufacturer itself, with zero independent studies. Innovations for Poverty Action (IPA) Nigeria+ USAID + Bloom Public Health initiated a RCT that was supposed to be evaluating it against lab tests with results to be published in April 2025, but as far as I can tell, the results were never published.

From a technological perspective, if it is what it says at all, one would expect that it could be pretty good at confirming something as genuine and in pristine condition but pretty bad at telling if something is counterfeit. E.g. a generic of the same drug, or a slight formulation change (such as the binder or coatings), or some meaningless degradation during storage, will give a different result. Sounds like they're trying to bypass this by using AI to tell what differences are meaningful or not. But without third party validation.... *shrug*

Also, the manufacturer initially marketed itself as a hardware and SaaS company, but has now been broadening out, making e.g. a B2B marketplace where the sell drugs to pharmacies (RxDelivered) and a financing and point-of-sale software service for those pharmacies (RxPay). So as far as supply chain auditing goes, i's a red flag when the "independent" quality tester also operates the marketplace selling the goods (something that they try to spin as a "trusted network"). Because they obviously benefit from any biases to detect their goods as legit but flag competitors as fake.

As it stands, I don't think one can say it's an outright grift, but on the other hand, I wouldn't put too much trust in this company as it stands.

This is not fuzzy string matching. This is a vector database. Please learn the difference.

Vector databases return conceptual matches, not fuzzy data matches (such as fuzzy string matches).

Also: "The results of drug sample scans are recorded and stored on the blockchain", which isn't something that affects the choice of AI models, but does make me immediately suspicious that this whole thing is a grift / scam.

What are you talking about? You generally run inference on the GPU. Sometimes in RAM. Almost never on swap. It is certainly possible to run a model without being able to load the whole thing into memory, but it's comically slow.

I agree with the first sentence, but not the second. I think this is an honest attempt to protect kids from something that is clearly harmful to them. I just don't think it will work. I think it's a situation where people see a real problem and feel like they must do something, but don't really know what can work. This is something, and there's a non-zero (if small) probability that it will do more good than harm.

It was never anything more than an invitation. A degree is a prerequisite for many jobs, but it has never been a guarantee.

That's part of it, but the smaller part. The more important parts are that a college degree demonstrates that you can learn, that you can take on a large, somewhat challenging, multi-year task and complete it, and that you succeeded at acquiring some level of broad-based education. Engineers and other specialists tend to scoff somewhat at "liberal education" because it doesn't seem like it's useful... but there have been endless attempts to substitute narrow vocational education in technical fields and they don't stick.

In the late 90s I worked with people who'd graduated from BM's attempt to provide narrowly-focused education. IBM had scoured the factories for the brightest then sent them to an intensive two-year course in software engineering, paying them to learn. The result was competent software engineers who were difficult to work with because they knew absolutely nothing but software. Their thinking was full of the basic misunderstandings of politics, economics, science, literature, etc. that you find in typical people without any post-high school education -- and who didn't pay much attention in high school either.

They knew information theory and could write good code, but their lack of general education negatively impacted their ability to build software systems in many ways. They didn't communicate well in writing (though technical writing courses had been part of their IBM education), but more fundamentally they just weren't very good at understanding the complex problems of the business. It's hard to pin down precisely what the issue was, but it was real. They were as smart or smarter than many of the college grads... but they were just less effective as employees.

IBM ultimately abandoned the approach and started sending bright young factory workers to regular universities. Even that was less effective than hiring people who had gotten to and through college on their own, though.

Indeed... though I also think that the trap is less crushing than many like to describe. I think the biggest issue isn't that the loan repayment is crushing, but that people don't like paying for something they got years ago. I don't mind paying my mortgage because I'm paying for a house I'm living in now. I would definitely resent making payments on a house I already sold and moved out of.

Personally, I didn't get any student loans. It would have been financially smart for me to have done so, actually, but I didn't.

Or a cheap four-year school, which was my strategy. Even better if there's such a school close to where your parents live, so you can live at home. A lot of the cost of education isn't the education, it's room and board, and if you can get that from your mom & dad for free, do it. This was my plan, though I ended up not following it because I got married -- but I married a woman who is a couple of years older than me and was close to graduation herself. She graduated a few months after we got married and started work that fall as a school teacher; not a lot of money but enough. Financially this strategy worked well for her; she quit teaching after a few years and has since lived on my income, which is an order of magnitude larger than she'd ever have made.

Another alternative is to join the National Guard or a reserve branch of the military. I joined the Air Force reserve. It qualifies you for most of the GI Bill benefits, but only requires a few months up front of full-time service for basic training and specialty training. After that, one weekend per month plus two weeks per year (which your employer is legally obligated to allow you to do). If you pick a military job that is related to your career plans, the specialty training could be extensive, as much as three years in some cases. Or you can pick something with less training requirements. I became a Security Policeman because the training was short... though what I learned about physical security has actually been useful in my software career.

I mentioned above that I should have gotten some student loans... I didn't realize until too late that part of the GI Bill benefits was that the government would have paid off my loans for me. I met another kid who was going to school on scholarships + GI Bill money who took advantage of this: He borrowed $20k (in ~1995) for "school", but used it to buy a brand new Camaro, then let the US Army pay it off.

Start to end, it took me 8 years, though I took a two-year hiatus to be a missionary. The last four of those, I was working full time, writing software. The last year of that time I was actually teaching a C++ programming course at night at the university I was attending, getting paid a small amount as adjunct faculty and getting 50% off of tuition for my own final coursework. That last part was not a common situation by any means, not something you can plan on, but it worked well for me.

I think it would have been marvelous to have done a "traditional" college education, living away from home, immersed in the college culture with lots of other young people. But I graduated with zero debt, and having already started my career, and my family, so it was a great outcome.

CaptQuark's main point is absolutely right: You don't need large student loans to get an education.