Second Foundation’s plot and characters were great, I can see we’ve deviated well past that book. I don’t hate the series, but I knew keeping close to the books would be impossible. And it’s a shame, because the book Mule is a fascinating character, and so is Arcady.

“You…cannot pass. Sorry. I don’t make the rules.”

LLMs can be good at openings. Not because it is thinking through the rules or planning strategies, but because opening moves are likely in most general training data from various sources. It’s copying the most probable reaction to your move, based on lots of documentation. This can of course break down when you stray from a typical play style, as it has less to choose from in the options of probability, and only a few moves in there won’t be any more since there’s a huge number of possible moves.

I.e., there’s no calculations involved. When you play a LLM at chess, you’re playing a list of common moves in history.

An even simpler example would be to tell the LLM that its last move was illegal. Even knowing the rules you just told it, it will agree and take it back. This comes from being trained to give satisfying replies to a human prompt.

I understand the cross-posting issue, it’s something that comes with a federated discussion format and I don’t think anyone has come up with a great way to solve it without such a distributed effort. It’s ironic that before when there were so few instances (before and during the first Reddit migration) there was a concern that without cross-posting a lot there wouldn’t be enough growth and some communities might die out if they happened to be on a single failing instance. I’d rather have too much activity than none at all, at least you can filter or block the worse ones.

I’ve heard the only way to win is to lock down your shelter and strike first.

It can be bad at the very thing it’s designed to do. It can repeat phrases often, something that isn’t great for writing. But why wouldn’t it, it’s all about probability so common things said will pop up more unless you adjust the variables that determine the randomness.

There’s some very odd pieces on high dollar physical chess sets too.

And unmonitored? Don’t trust anything from Google anymore.

What makes this better than Ollama?

That’s even more why it feels like someone new in the company stepping in and questioning why there isn’t something in play officially if there’s interest in freeware/open source. Someone who talked to the lawyers first to make sure no right were signed away yet. That may be very pessimistic and conspiratorial, but if there isn’t any reason to stop someone else’s work on something, why would they send one? I don’t know a lot about copyrights and trademarks, but I do think there is a point where if you aren’t using an asset and others are interested, you shouldn’t be able to just hold it under lock and key and do nothing with it. I think patents are like that, you have like 20 years or something protected to do something, and then it’s open(?) Again, I’m not sure.

Any reason given? Not that they have to give one, it’s still their property to do what they want with it. I would keep an eye on them and if they somehow in the future come out with something very similar, I hope there are good records of the past years of work and discussion. Since it was going to be free and not for profit, not really a case for lost income, but there must be some laws to protect people working in good faith with a trademark knowingly who get their ideas stolen FOR profit. If that happens.

That seems to suggest that the American style is a preservation of the older English format, much like we kept spelling of some words like the original English at colonization while the UK gradually changed with other influences around them.

“If there’s anything I can’t stand, it’s a smart owl.”

Not designed, but trained. Training involves rewarding finding answers, so they WILL give you something. “I don’t know” is not going to fare well in the training development, so it naturally gets filtered out, while very creative (but wrong) LLMs do well.

Free will is something where people talk about it as a binary thing, but it can be both the ability to make choices, yet very deterministic at the core. If someone asks you to think of your favorite color, in your mind you visualize what that is, and it’s your preference and choice for whatever reason you like it best. But the deterministic part begins when you wonder when you made that decision. Can you even narrow down the instant when it popped into your mind as the preferred choice, or what occurred before it was made? At some point there was a triggering of thought and memories from the question asked that resulted in you thinking of your color, but when did it go from predictable neuron firings to a choice? There is a gray area there.

For what it’s worth, while I enjoyed some of the later Terminator movies for themselves, the saga ended with T2 in my mind. Where that future led could be just as dark, as someone else could come up with their version of Skynet eventually, like any other technology, but we are left to ponder that on our own. The actual previous future is gone thanks to the efforts made, and we’re allowed to try again.

This is exactly what a President, an elected service worker sworn to protect the rights of the public, should be doing.

Not.

Crusade (Babylon 5 spinoff)

looks at Ohio

Okay.

Many people here have posted the link to Climate Town’s video on expiration dates, but your comment also brings into focus a video of theirs about consumer waste. Actually he’s probably made a few on that subject, but the one that came to mind was about the circle of buying and returning products (eg. Amazon returns), and what really happens. Good lord, the waste.

Might be something else too. Many people are so sick of hearing about certain people that they’ve set up filters to avoid any news about them. So what better way to get past that than to just keep the references generic enough to get the views through such blocks but still be accurate.

Safe drivers are just that, no matter how they park. If your visibility is bad when you park front in, then either you’re hauling something that’s blocking your line of sight (in which case it would make sense), or your vehicle has terrible visibility, which is far too many of them.

Park how you want to park. I’ve never had a problem backing out of a space.