If the AT protocol allows public access to content, they can’t create a proprietary training set. But the content is available for anyone who want to add it to a public training set.

• Support for slavery before the Civil War

• Carter’s airline deregulation

• Clinton’s welfare “reform” and NAFTA

• Obama’s finance sector bailout

• Biden blocking a national rail strike

Are you talking about someone who’s deliberately claiming to have experienced something they only read about, or someone who’s genuinely uncertain of their own memories?

Legally, yes. (But of course, the Supreme Court has turned interpreting the Constitution into a game of Calvinball.)

If nothing else, it’s diverting views and revenue from whatever genuine right-wing media they’d be watching otherwise.

I don’t know, but there are some common names that are actually obscure forms of classic theonyms, and the people using them may not even be aware of the connection—for instance, “Dennis” is a form of “Dionysus”. Would you count that or not?

deleted by creator

If n is the day the item is introduced, the total quantity is 42-(n-6.5)².

Lapsang Souchong (smoked black tea).

Logically, yeah—it went from “all X are Y” to “no non-X are Y” (or equivalently, “all Y are X”).

When he won before, he was outside the Republican party establishment and just put his own unprepared cronies in charge.

Interesting approach—to detect fake news by simulating humans’ reaction to it rather than judging the content itself.

As many others are pointing out, cultural hegemony plays a major role—but I think there’s another factor at play as well:

Greek, Roman, and Norse mythology have been dead and fossilized for a thousand years and more, and in the meantime a long tradition grew up of mining them for allegory, with their prior religious significance stripped away. Most other world mythologies, on the other hand, still form part of active belief systems, or recently died out under colonial occupation and so carry postcolonial political overtones. So borrowing from them could be more problematical, whereas classical mythology has basically been left up for grabs by its former adherents.

The classical Romanization was more accurate in its time—the issue is that the common pronunciation of classical Latin changed after the classical era (for instance, the “c” became soft in many contexts, instead of always being pronounced as “k” as it was in classical Latin).

If you use the original classical pronunciation for Latin, you’ll also pronounce the classical romanized Greek names correctly—and if you spell them the classical way you’ll recognize them more easily in Latin sources. The modernized romanization is most useful if you’re only interested in Greece and not in the classical world as an integrated whole.

Utnapishtim.

Ukraine should offer the North Koreans political asylum if they defect.

Not with a typewriter, though.

Yeah, that’s why we need at least… two of them.

I know—I’m just saying that any other theoretical solution would always be worse than cloning, because you’d lose genetic information.

Edit: See Isodisomy for details.

The obvious solution would be to just clone the parent.

Any hypothetical attempt to simulate recombination would produce a genetic clone at best—but more likely (even if you overcame the practical issues), you’d end up replacing some of the unique material on one half of each chromosome with a copy of the genes on the other half (with a corresponding increase in the risk of genetic defects).