Dan Milway

@julieofthespirits@kolektiva.social AlphaFold had at least 2 strikes against it: (1) It does something of value and (2) It's not an LLM. ((1) and (2) are probably related)

@musevg I know someone who wrote a linguistics thesis based on the Enron corpus. They used it yes because it was large and public, but also because, unlike most corpora, the people in it didn't know they were making a corpus—minimal observer effect. @Nfoonf

@ngaylinn@tech.lgbt Speaking as a nativist myself, I don't think anyone serious has argued that minds have an innate set of "facts"—usually it's something like "concepts," and that those concepts are radically basic.

@inthehands@hachyderm.io Gah! Wrong damn biases tho. The gd stupidity is just too.much sometimes

@ngaylinn@tech.lgbt I'm a generative linguist by training, and my main focus/expertise is syntax. Based on the body of evidence I'm familiar with, I tend to think that the mind is symbolic—i.e. we have innate procedures that manipulate symbols. I'm somewhat agnostic as to the extent to which those symbols are innate, but I think they all must have some innate basis—perhaps what you referred to as "constraints." In my younger days, I did extend this to how I viewed "AI"—I was somewhat hostile to ANNs and though GOFAI didn't get a fair shake—but now I view CogSci and AI as two different sorts of fields, and the success or failure of a particular type of "AI" says almost nothing about whether the human mind is structured anything like that particular type of "AI."

@mausmalone If you're tired of it you should probably not engage with it. @cR0w

@cR0w I was About to reply that "This would be funnier if there weren't AI-bootlickers regularly making this analogy unironically in defense of Mozilla" and wouldn't you know it, you got one in your replies already. @mausmalone

@timnitGebru@dair-community.social Damn, so disappointing 😞

@noleli@mastodon.social @tante@tldr.nettime.org It was ever thus with big tech's claim of "democratizing" anything