ianhorn

10 ms·

61.
by ianhorn 6y ago
That first point could be a case for a code of ethics, not against one. Right now decision makers tell you to build the thing, and you speak up saying that seems sketchy, but you both don’t know what the people who don’t speak up think. You
62.
by ianhorn 6y ago
“Prediction market” is such a nice euphemism for gambling.
63.
by ianhorn 6y ago
Stringly typed functions frustrate me to no end. Tooling for programming gets better every year, yet stringly typed code is so resistant to any of those improvements. Pycharm at some point had to add SQL tools in python strings for gods sak
64.
by ianhorn 6y ago
To add to what everyone else is saying, higher kinded types (HKTs) are what make monadic code really reusable. It lets you write code that is polymorphic over a variable kind of monad, rather than code that works for a specific monad like a
65.
by ianhorn 6y ago
Maybe $10/mo towards games from studios they own is more profitable than $10/mo towards games from elsewhere?
66.
by ianhorn 6y ago
I don't read that as saying Bethesda Games aren't going to be exclusives. Or maybe it'll be xbox plus windows. Meanwhile, there's this quote from Satya Nadella from press release at the bottom of OP, speaking about this
67.
by ianhorn 6y ago
Thanks!
68.
by ianhorn 6y ago
Everyone in this thread seems to be saying Pratt and PEG matter most, which I don't see in that book. Any recs on that front? (Or do you think they're not important?)
69.
by ianhorn 6y ago
I'd like to learn more about formal grammars and parsing in depth. Can anyone recommend a good textbook?
70.
by ianhorn 6y ago
The fact that this includes stock makes it that much more surprising. Microsoft’s stock has grown like 5x in the last four years, (2x over the last two). An offer that was worth $150k when you were hired is probably making you lots more now
71.
by ianhorn 6y ago
Sorry, I thought when you said missing piece in your previous message, that you meant the missing piece of what makes monads unique (arguing from a place of knowledge about “no THIS is what makes monads important”), rather than the missing
72.
by ianhorn 6y ago
I think there are two different questions: Will our protagonist achieve their goal? How will our protagonist achieve their goal? You can be unpredictable by not knowing where the story is going, or unpredictable by not knowing how the story
73.
by ianhorn 6y ago
I disagree. Map (or a functor) alone can't do everything I've said. There are a few reasons. I assume you're already familiar with the subject matter, so I'm going to cross my fingers that you know it in haskell and use
74.
by ianhorn 6y ago
A monad's special function application lets you write much simpler code in certain situations. Say you're working with some data structure that contains/emits numbers: a pointer to a resource containing a number. A list of nu
75.
by ianhorn 6y ago
> The efficiency of an estimator is a different from its bias. I think the comment is drawing a parallel to variance (better efficiency = lower variance). Still not exactly the same, I think, but pretty damn similar.
76.
by ianhorn 6y ago
> I disagree. Negative results are important. Null results are of very limited interest. The two are worlds apart. I agree they're different but, but disagree that they're worlds apart. There's a spectrum between them, cau
77.
by ianhorn 6y ago
I've been noticing a variant of this in myself lately. Maybe I think about some problem and stew on it and think to myself that it's probably not possible, or at least I can't come up with ideas. Then I hear that somebody els
78.
by ianhorn 6y ago
I've been doing sorta similar stuff with GPT-3 since I got access, and I can vouch that it's effective. Not up-to-the-hype effective, but that's more a function of the hype. It's been really great at creative stuff that
79.
by ianhorn 6y ago
This is cool! I've been playing around with GPT-3 a bunch lately with the same goal in mind. I spent too much of the afternoon feeding it a few 7-point plot examples like [0] and then brainstorming my own with it. It was incredibly fun
80.
by ianhorn 6y ago
Diagrammatic notations seem to pop up all over the place, but in my mind they all share a difficulty that interesting complex graphs tend to not be planar (can't be drawn without criss-crossing lines). That makes it hard to organize in
81.
by ianhorn 6y ago
> ...the content is merely linguistic, not mathematical. I disagree with the "merely" here. Half of mathematical practice seems to be finding the right way to think about a thing. If the author were claiming this was a new gran
82.
by ianhorn 6y ago
The number you're proposing has about 1.4 billion digits in it. It has more digits than just listing a number for everyone who votes and contains the same information.
83.
by ianhorn 6y ago
dot v :: V F -> F (dot v) is a member of the dual space. To double it up with dual dual, and to do it with the same partial application construction, we need to introduce an inner product on the dual space. ddot :: (V F -> F) -> (V
84.
by ianhorn 6y ago
Take a look at the original discussion. https://news.ycombinator.com/item?id=23893817 None of it is discussing the content, which makes sense given that it was nonsense. It just makes me further believe a very small minorit
85.
by ianhorn 6y ago
> 2. Reset the rocket equation in LEO by doing in-orbit fuel transfer. What do you mean by this? If you want to propel a mass M to delta-v V, you need a certain amount of initial mass. Dividing it into stages is already taken into accoun
86.
by ianhorn 6y ago
Unfortunately, they didn't give many details in the paper. It's frustrating, to say the least. Yay reproducibility. They say > We use the same model and architecture as GPT-2 [RWC+19], including the modified initialization, pre
87.
by ianhorn 6y ago
First off, GPT-3 is absolutely not a vanilla transformer. The moving pieces here in BigBird are a vector associated with each token in the sequence and a few more global vectors that you can think of as latent variables. Those pieces are pr
88.
by ianhorn 6y ago
Curbside has been really excellent. I buy a thing online and less than half an hour later I'm back at home with the device. Makes me realize how much of my amazon shopping was because of the fast delivery.
89.
by ianhorn 6y ago
With that analogy, the balloon would be the ruler in my analogy, which leads me back to the original questions. Using the balloon analogy, if you have a normal balloon in 3D euclidean space, you can measure/define the expansion easily.
90.
by ianhorn 6y ago
I don't understand what it means for points to move apart faster than the speed of light. I understand speed of a thing to be change with respect to space, so what does speed of space mean? Naively, I'd define it to be zero. If I

More ›