My sense in reading the article was not that the author thinks artificial general intelligence is impossible, but that we're a lot farther away from it than recent events might lead you to believe. The whole article is about the human tendency to conflate language ability and intelligence, and the author is making the argument both that natural language does not imply understanding of meaning and that those financially invested in current "AI" benefit from the popular assumption that it does. The appearance or perception of intelligence increases the market value of AIs, even if what they're doing is more analogous to the actions of a very sophisticated parrot.
Edit all of which is to say, I don't think the article is asserting that true AI is impossible, just that there's a lot more to it than smooth language usage. I don't think she'd say never, but probably that there's a lot more to figure out—a good deal more than some seem to think—before we get Skynet.
And I only subscribed because the one-time purchase was not available at that moment, so I assumed no such option existed. But honestly, I'll continue to pay the subscription. It's still less than a year's worth of my monthly donations to the developer of Tasker.
Ain't that the truth. People act like charging for software is evil no matter what. There's a huge difference between a lone dev trying to earn a living and a huge corporation trying to wring every last ounce of profit out of their users. And there's probably degrees of nuance between those.
Especially if they seem like a reasonable person, wanting reasonable amounts for good work.
And that's the important context in this discussion. You've got a dev who's active in the community and who builds an app not only with great features and UI but with stability too. And he has a not insignificant user base that is familiar not just with his work but essentially with this exact app... It's reasonable for him to assume we'll see the value and be willing to pay. And he is correct.
I'm personally averse to subscription models, but again context matters. Reasonable rate and you know what you're getting. And I say this as a huge fan of both FOSS and socialism. I could have easily just let my DNS continue to filter out the ads, but I appreciate quality and believe it should be appropriately compensated.
14 years myself. More than a third of my life. Your bad relationship analogy is a pretty good one (the analogy, that is). I was in a real life one of those about the same time I joined reddit. Glad that one didn't last nearly as long!
I was raised Catholic, but I've been an atheist for—oh fuck I'm old—more than half my life. But... Monastic life seems pretty dope. Why can't there be a secular order that's just devoted to knowledge/contemplation for its own sake (or the betterment of humanity). I know it kind of sounds like I'm describing a university, but I mean with the personal discipline, strong communal bond, and simple lifestyle.
And whenever you have a chart of historical data like this, you have to at least consider that an increase could be reflective of either improved diagnostic or record-keeping abilities.
It's probably also related to when a person first encountered JS. If you learned it pre-2015—even if you're aware of the changes made in ES6—I can see how it would be hard not to view JS as cumbersome. I personally love to use it, but I can't imagine that would be true without let, const, classes, etc.
Yes. Or even composition of words. I remember during a class discussion translating "Thanksgiving" as "Danksgebung" on the fly. At least I greatly entertained my professor—and I'll never forget "Erntedankfest".
Ditto Sync for Lemmy. Once the API shit started going down and I first started using Lemmy, I would not have dared to hope that any of the third party apps would migrate here too, let alone multiple.
I reference this particular moment in our national culture—I won't say often, but with some regularity—and very few people ever indicate recognition. Either they don't know/remember it, or a response is beneath them (which is probably true).
And even then, I'm fine to browse a menu on a mobile-friendly site (as long as the restaurant is diligent about providing reliable wifi for anyone who might not have great signal). But when the code has me download a PDF, they can fuck right off. First of all, I don't need the menu sitting in my cache or download folder. And when the PDF was clearly formatted for physical printing... Good lord. I'm not pinching and zooming this shit.
My sense in reading the article was not that the author thinks artificial general intelligence is impossible, but that we're a lot farther away from it than recent events might lead you to believe. The whole article is about the human tendency to conflate language ability and intelligence, and the author is making the argument both that natural language does not imply understanding of meaning and that those financially invested in current "AI" benefit from the popular assumption that it does. The appearance or perception of intelligence increases the market value of AIs, even if what they're doing is more analogous to the actions of a very sophisticated parrot.
Edit all of which is to say, I don't think the article is asserting that true AI is impossible, just that there's a lot more to it than smooth language usage. I don't think she'd say never, but probably that there's a lot more to figure out—a good deal more than some seem to think—before we get Skynet.