hissing meerkat

about:config is a weird thing to lock behind ditching Firefox and downloading a different browser

I’ve eaten on about that before, but decades ago when food was cheaper. Nothing is satisfying, you are hungry all the time, constantly craving some nutrition you no longer even know how to acquire or what it is, but it’s absent from everything you eat.

Peanut butter and bread was too expensive. Peanut butter was a treat. Bread from bakery surplus cost two to three times as much as rice. For your example, at $400 a year you’re looking at $8 a week. If a jar of peanut butter is $3 and has 4800 kCal in it and bread is $1 a loaf and has 24 60kCal slices in it, then a jar of peanut butter and 5 loaves of bread a week only gets you 12000 kCal a week, which isn’t enough for a moderately active adult. And you’re going to be missing out on all sorts of nutrition.

At the time the best things to buy were eggs, beans, rice, and processed dry foods. Then you buy things that make eating them bearable and are also cheap in combination: whole or powdered milk to eat cereals, raw sugar, fat to cook into things, very cheap meats, cheese when it was cheap, and processed frozen foods that are similar in price to their constituents, which at the time were common because they are a way of storing food from a production season to sell in an off season. Then you get a few things to try to stave off cravings, like some long-term storage plastic-packed cuts of meat, or canned vegetables, or concentrated frozen fruit. At a low budget a can of food represents everything you get to eat for a day, or more. Fresh vegetables or fruit were completely unobtainable unless there’s a local surplus.

Now the structure of food markets is different and everything is priced based only on demand and not on supply, so frozen processed foods that were available then due to the product being made to take surplus or trimmings and then store them are now priced based on demand for the product. The only things that have stayed similar are the prices for eggs (usually), the cheapest meats (sometimes), staples (usually), and canned foods which are priced based on the cost of transportation and are still routinely too high for such a low budget.

Random word generator

Whether or not you use downvotes doesn’t really matter.

If what you like is well represented by the Boba drinkers and the Boba drinkers disproportionally don’t like Cofee then Cofee will be disproportionally excluded from the top of your results. Unless you explore deeper the Cofee results will be pushed to the bottom of your results. And any that happen to come to the top will have arrived there from broad appeal and will have very little contribution to thinking you like Cofee.

If you don’t let the math effectively push things away that are disliked by the people who like similar things as you then everything will saturate at maximum appeal and the whole system does nothing.

There’s two problems. The first is that those other things you might like will be rated lower than things you appear to certainly like. That’s the “easy” problem and has solutions where a learning agent is forced to prefer exploring new options over sticking to preferences to some degree, but becomes difficult when you no longer know what is explored or unexplored due to some abstraction like dimension reduction or some practical limitation like a human can’t explore all of Lemmy like a robot in a maze.

The second is that you might have preferences that other people who like the same things you’ve already indicated a taste for tend to dislike. For example there may be other people who like both Boba and Cofee but people who like one or the other tend to dislike the other. If you happen to encounter Boba first then Cofee will be predicted to be disliked based on the overall preferences of people who agree with your Boba preference.

No, not as simply as that. That’s the basic idea of recommendation systems that were common in the 1990s. The algorithm requires a tremendous amount of dimensionality reduction to work at scale. In that simple description it would need a trillion weights to compare the preferences of a million users to a million other users. If you reduce it to some standard 100-1000ish dimensions of preference it becomes feasible, but at the low end only contains about as much information as your own choices about subscribed to or blocked communities (obviously it has a much lower barrier of entry).

There’s another important aspect of learning that the simple description leaves out, which is exploration. It will quickly start showing you things you reliably like, but won’t experiment with things it doesn’t know you’d like or not to find out.