Narwhals
narwhals-dev.github.io
August 27, 2025 at 12:11 AM
It's also weird because if you switch from any other source of heat to a heat pump in order to have air conditioning, you're typically reducing your overall carbon emissions.
July 21, 2025 at 12:32 PM
It's also weird because if you switch from any other source of heat to a heat pump in order to have air conditioning, you're typically reducing your overall carbon emissions.
Basically there's a way to analyze this type of data with repeated measures of the same person, and they didn't use those methods and instead treated each issue as it's own observation. Also small sample size and almost insinuation results.
July 12, 2025 at 8:11 PM
Basically there's a way to analyze this type of data with repeated measures of the same person, and they didn't use those methods and instead treated each issue as it's own observation. Also small sample size and almost insinuation results.
It's really weird they didn't use a MLM, or at least publish a plot with the time by contributor.
July 12, 2025 at 2:08 PM
It's really weird they didn't use a MLM, or at least publish a plot with the time by contributor.
I'm finally reading this paper but my first thought is anybody on earth good at predicting how long a task will take them....?
Setting aside the small person sample, randomizing the tasks within developer is....super odd? Wouldn't we start to expect a lot of order effects?
Setting aside the small person sample, randomizing the tasks within developer is....super odd? Wouldn't we start to expect a lot of order effects?
We ran a randomized controlled trial to see how much AI coding tools speed up experienced open-source developers.
The results surprised us: Developers thought they were 20% faster with AI tools, but they were actually 19% slower when they had access to AI than when they didn't.
The results surprised us: Developers thought they were 20% faster with AI tools, but they were actually 19% slower when they had access to AI than when they didn't.
July 12, 2025 at 9:28 AM
Consensus from stats folks is that it's a bad study
July 12, 2025 at 2:07 AM
Consensus from stats folks is that it's a bad study
That's what I do! 50/50 granola to TVP ratio.
June 25, 2025 at 4:23 PM
That's what I do! 50/50 granola to TVP ratio.
You can basically eat TVP like cereal which makes for a quick, very bland, very protein-efficient breakfast.
June 25, 2025 at 1:22 PM
You can basically eat TVP like cereal which makes for a quick, very bland, very protein-efficient breakfast.
"after all why can't sufficiently complex matrix multiplication spontaneously create consciousness"
shotwell.ca/posts/2022-0...
shotwell.ca/posts/2022-0...
Large language models will never be conscious – Gordon Shotwell
shotwell.ca
June 11, 2025 at 5:58 PM
"after all why can't sufficiently complex matrix multiplication spontaneously create consciousness"
shotwell.ca/posts/2022-0...
shotwell.ca/posts/2022-0...