> To the average person, LLMs are too useful to reject them
The way LLMss are now, outside of the tech bubble the average person has no use for them.
> on some ultimately muddied arguments along the lines of "it's OK for humans to train on books, but it's not OK for robots"
This is a bizarre argument. Humans don't "train" on books, they read them. This could be for many reasons, like to learn something new or to feel an emotion. The LLM trains on the book to be able to imitate it without attribution. These activities are not comparable.
I find it funny when people say it's not on Carlsen when it was entirely his decision to not compete. We already have rapid and blitz world championships that are separate. This is the classical world championship and I think the format is both exciting to watch and decently fair.
That's a very privileged attitude. Certainly someone already wealthy will not get much more happiness from gaining more wealth, but for many Americans a few thousand dollars could be a life-changing amount of money.
It's not the proportion of the population that is wealthy that matters, it's what proportion of the total wealth that they have. From the article, "In 2022, the wealthiest 1 percent held a quarter of the net personal wealth in the EU." So if you are taxing the 'poor' 99% of people, you are only accessing 75% of the taxable wealth.
Perhaps I'm oversimplifying things, but I have personally always assumed that using "Medieval" to describe D&D was almost entirely a reflection of the (non-magical) level of technology, and not anything societal or cultural.
Interesting. The article does mention the aspect of tech/weapons to mention it's not particularly "medieval" either:
> The D&D weapon list has a medieval feel to it, but partly that’s just because that’s what we’re expecting to find. In fact, it’s a sort of survey of (mostly) pre-gunpowder weapons. Most of the weapons and armor appear in ancient Europe and in Asia as well as in medieval Europe. Partial exceptions: Composite bows are mostly non-European, while longbows are associated with Europe. The halberd is basically a Renaissance weapon, and the two-handed sword appears in medieval Europe, India, and Japan, but not the ancient world. No one knows what “plate mail” is supposed to be.
Early D&D didn't really make any sort of assumptions about technology level, either. Arneson's campaign was heavily implied to take place in a post-apocalyptic future - Blackmoor was full of advanced technology, including nuclear powered flying cards, lasers, androids, etc.
Gygax himself didn't lean into it as hard, but there were plenty of fairly gonzo modules that did include aspects of future and/or alien technology.
Wouldn't that lend credence to the author's claims? That it only appears to be a medieval society, but is instead something very different (a liberal capitalist world after grand social dissolution). After all, the two adventure/roleplaying game series that Bethesda is/was known for are The Elder Scrolls and Fallout, the latter, at least, explicitly taking place in a post-apocalyptic universe.
I think the discussion itself is kind of a weird one to have. Most of the author's point is about OD&D, but OD&D itself was a very short lived phenomenon. People pretty rapidly moved to either the AD&D or B/X line of products, and there are some departures in the rules, and both lines added official rulebooks that do explicitly contain things that go against a lot of the author's points. And basically every table from the start has had a pile of house rules (or at least generalized processes) to handle situations not explicitly called out in the rulebooks. Much of the question of feudalism is more of a campaign setting question than a rules question - plenty of people played campaigns back in the day in settings where there was royalty, where you would have to purchase or otherwise acquire land rather than just plopping down your claim somewhere.
The argument the author is making is messy because it mashes together the dichotomy of old school D&D versions vs. the modern equivalents. Originally, D&D was a framework that you built on top of. You might adhere fairly closely to the rules as written, but they were intentionally quite basic in nature, and additional structure and systems almost always came as part of your campaign milieu and table-specific needs. Modern D&D has significantly more rules - it's a more complete game, with all the good and bad that entails. Neither have a mechanism to force you to use any specific portion of them, and many people don't.
I agree with the sentiment that D&D is not pro-medieval, though depending on which rulebooks you use it might incorporate aspects of medieval society and structure. I also don't think it's anti-medieval, because that implies structure that isn't there. Any given campaign or table might be more or less medieval than any other and still be totally authentic D&D.
There's a company that is famous for two series, one is a fantasy roleplaying game, one is a post-apocalyptic role-playing game, but the story and the structure of the games are nearly identical and the in the latter the mechanics and gameplay make a lot more sense in the context of the story than in the former.
Could you explain more about how first responders are not empowered at their jobs? I'm mostly curious if you see the cause as something local, or more like national policy that changes your ability to do your job.
I'm curious about your experience with older men not playing video games? I'm in my mid-thirties and I would say that the overwhelming majority of my male friends and co-workers play video games recreationally. Video games are more popular and accessible than ever, so I guess I'm confused about why someone would age-out of the hobby?
Perhaps it’s just been my personal experience or maybe a Gen X thing? I tried getting back into playing video games a few years ago. I gave up after a weekend. It was dead boring. I’d much rather read the paper.
The game industry is notorious for expecting way, way above 40hrs/wk of work, especially as deadlines to release approach. Stories of game devs sleeping under their desks at night are shockingly common.
I find it interesting that the study is only about print and TV news, since I get about 90% of my news from my state's public radio channel. I'd say that I have an extremely high confidence in both NPR and my state's local stuff.
What's funny here are that the example issues you listed are exactly the issues that the vast majority of Americans actually do agree on. Those are wedge issues that were carefully crafted by political parties to try and create a division in popular opinion when there isn't one. It's a fairly common political strategy nowadays.
You think the vast majority of Americans agree on abortion? I'm pretty sure that, no, they do not. The division is real, not just a political strategy.
Abortion isn't an either or position. There are quite a few position and this stat only shows a single one: 15-20% believe it should be illegal in all cases. What about the subset who believe it should only be legal for rape victims. Or only plan B style drugs (I'm not sure how the linked data classifies the responses, but some people do consider such drugs on the same level as abortion). Or only by first trimester. Or only by second trimester. Or until birth? Or some other position that I am not able to remember off the top of my head?
The way LLMss are now, outside of the tech bubble the average person has no use for them.
> on some ultimately muddied arguments along the lines of "it's OK for humans to train on books, but it's not OK for robots"
This is a bizarre argument. Humans don't "train" on books, they read them. This could be for many reasons, like to learn something new or to feel an emotion. The LLM trains on the book to be able to imitate it without attribution. These activities are not comparable.