gjm11

67 ms·

721.
by gjm11 5y ago
If you have an image consisting of a thin white line on a black background, then indeed as you scale it down the pixels covering that white line will become darker because each pixel corresponds to a small amount of white in a mostly-black
722.
by gjm11 5y ago
"Supposed to" in what context, by whom?
723.
by gjm11 5y ago
Surely the "red revil" would be these guys https://www.freebsd.org/ who, while certainly not having the exact same philosophy as RMS's, share a fair amount with it...
724.
by gjm11 5y ago
The Curry-Howard isomorphism doesn't say that an interesting proof corresponds to a program anyone would care about (other than as an equivalent to the proof), nor that an interesting program corresponds to a proof anyone would care ab
725.
by gjm11 5y ago
So, I tried this out with GPT-3 in the OpenAI Playground. (The model called text-davinci-002.) My prompt looked like this: Definition: A blorple is anything that is red and more or less round. Question: Is a tennis ball a blorple?
726.
by gjm11 5y ago
"Stochastic parrots" -- have you seen, e.g., the examples in the PaLM paper of how it does on "chained inference" tasks? I don't see how you can classify that as mere parroting.
727.
by gjm11 5y ago
I read it as unsarcastic but self-aware. "I am unashamedly an intellectual and unashamedly consider that X is better than Y and yes, I know that some people regard that sort of thinking as necessarily pretentious and stupid, and I want
728.
by gjm11 5y ago
The section of the TeXbook (the user manual for Knuth's TeX typesetting system) that talks about hyphenation has the lovely/horrible example of "the-rapists pre-aching on wee-knights".
729.
by gjm11 5y ago
Is anything in your second paragraph actually true? That is: does Pinker actually say anything like that, or did you just say he did because it makes him sound stupid and evil?
730.
by gjm11 5y ago
The author gives an example where running a benchmark tool and identifying the output for the "fizzbuzz" benchmark is ./runbenchmarks | grep '^fizzbuzz' with his preferred style of output and ./r
731.
by gjm11 5y ago
It might be worth distinguishing between the author and one of his characters .
732.
by gjm11 5y ago
No one is calling it a "peer-reviewed scientific paper published in Nature" in this thread. The only person who mentioned Nature in this thread did so exactly to point out that this paper isn't published in Nature but in an
733.
by gjm11 5y ago
It seems like you're suggesting that the authors of this paper are trying to imply that those professors were dismissed because they might give unwanted advice about Covid-19, or to hide the fact that they were re-hired at the Karolins
734.
by gjm11 5y ago
A person's opinion can be predictable in advance (1) because they are closed-minded and not responsive to evidence or (2) because they are responsive to evidence and the evidence is clear. Presumably you're saying #1 rather than
735.
by gjm11 5y ago
What do you think Pinker should have done when Dershowitz wrote to him asking "what exactly do you think this statute means"?
736.
by gjm11 5y ago
Maybe you know something about them that I don't, but to me it seems alarming that someone needs to be "rehabilitated" for those things. "Appeared in Epstein's flight logs" = "went on his plane to a TED me
737.
by gjm11 5y ago
Alan Dershowitz wrote to Pinker asking "do you reckon this statute means X or Y?" and Pinker said "It obviously means X". So far as I can tell, (1) Pinker didn't know that this was specifically about the Epstein cas
738.
by gjm11 5y ago
My favourite example of rebracketing is the origin of the word "orange". European languages got the word from Persian or Arabic. But in Persian it's "narang" and in Arabic it's "naranj". (And in the l
739.
by gjm11 5y ago
1. Everything that humanity discovers helps humanity, because knowledge is something we value. 2. Things it helps us discover about the universe may lead to new discoveries in physics (consider how astronomical observations helped in discov
740.
by gjm11 5y ago
That article doesn't give one single example of an orchestra that isn't using blind auditions any more. It does give one single example of something , namely an orchestra looking to appoint something like an assistant conductor
741.
by gjm11 5y ago
Yudkowsky says he and his wife did it in 2014, so it sounds as if it was actually pretty much at the same time as said prior art. (Also, they were specifically doing it as a treatment for SAD ; I don't think Yudkowsky claims novelty f
742.
by gjm11 5y ago
That article is behind the NYT paywall but the bit of it I could see appears to be saying not that > blind auditions are now a thing of the past but that they should become a thing of the past, so that orchestras can ensure that thei
743.
by gjm11 5y ago
France is in NATO.
744.
by gjm11 5y ago
Adding "Yes!" changes the *average length" of the sentences, which is one of the inputs to the Dale-Chall (un)readability score.
745.
by gjm11 5y ago
I don't know about you, but although I too am old and stable I have plenty of data-loss bugs, and I'm running on hardware that needs regular downtime and constant maintenance in order not to fail catastrophically. I therefore don&
746.
by gjm11 5y ago
https://i.imgur.com/gKDvqRh.jpeg is one wall of the room I'm in now. Leftmost column is mathematics. Next column is mathematics shading into computing (the top shelf is numerical analysis and the like). Next column is
747.
by gjm11 5y ago
Fair warning: I don't think either of the philosophy books I mentioned is likely to scratch the same sort of itch as Carse's book.
748.
by gjm11 5y ago
Hmm, maybe. The way I'd say it is that dealing with whatever happens to you is generally your responsibility, not that the things that happen themselves are. ("Generally", not literally always. Extreme example: If you get
749.
by gjm11 5y ago
The room I'm in right now, which is also where I work, has two walls completely covered with books. (This is where all the sciencey nonfiction lives, and some other things like history and philosophy and economics.) It is, indeed, nice
750.
by gjm11 5y ago
My brain is absolutely hopeless at top-10-like things. But here are a few gems from some of those categories. Mathematics: John Conway's "On Numbers and Games". (If you've run across the term "surreal numbers",

More ›