Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
jasonthorsness
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
8 ms
·
91.
▲
by
jasonthorsness
1y ago
“None of the safe programming languages existed for the first 10 years of SQLite's existence. SQLite could be recoded in Go or Rust, but doing so would probably introduce far more bugs than would be fixed, and it may also result in slo
92.
▲
by
jasonthorsness
1y ago
"We achieve 19.8 GB/s prefix sum throughput—1.8x faster than a naive implementation and 2.6x faster than FastPFoR" "FastPFoR is well-established in both industry and academia. However, on our target platform (Graviton4,
93.
▲
by
jasonthorsness
1y ago
The leading cause of death up to early 40s is still accidents so a fixation on heart disease and cancer might send the wrong message too https://wisqars.cdc.gov/pdfs/leading-causes-of-death-by-age-...
94.
▲
by
jasonthorsness
1y ago
The news revolves around "new" stuff, not reporting things people generally know. At young and even middle ages, people dying of anything is highly unusual and skews more towards some of the unlikelier causes compared to the break
95.
▲
by
jasonthorsness
1y ago
The change log is massive, so this is not unexpected to have a few issues. Maybe the two-year backlog of features and updates should have been batched into a few releases instead of this one big one? But there was a beta, and this is rippin
96.
▲
by
jasonthorsness
1y ago
It’s not the betting, it’s the business machine that delivers the betting. This looks the same for any business that makes most of its money only through exploiting some human vulnerability whether it’s social media; shorts, gambling, nicot
97.
▲
War on Slop
(jasonthorsness.com)
2 points
by
jasonthorsness
1y ago
|
0 comments
98.
▲
by
jasonthorsness
1y ago
"But we have exactly the same number of reviewers." LLMs can help with reviews as well. LLMs are not too bad at reviewing code; GPT 5 for example can find off-by-one, missed returns, all sorts of problems that are localized. I thi
99.
▲
by
jasonthorsness
1y ago
huh TIL "On modern Intel architectures, spatial prefetcher is pulling pairs of 64-byte cache lines at a time, so we pessimistically assume that cache lines are 128 bytes long."
100.
▲
by
jasonthorsness
1y ago
> Don’t use: > AI auto-complete in your editor > Agent mode or agentic tools for your educational projects While I am not learning "coding" as a beginner anymore, I am constantly learning new frameworks, languag
101.
▲
by
jasonthorsness
1y ago
Agree; the part of the application requiring multi-master semantics is probably a small piece and can be handled outside the database where there is enough domain-specific knowledge that it can be made simpler and more obvious how conflicts
102.
▲
by
jasonthorsness
1y ago
perfect, thanks
103.
▲
by
jasonthorsness
1y ago
Ha amazing that this is just an HTML file
104.
▲
by
jasonthorsness
1y ago
Another great use of the HN API! It would be nice to filter it somehow to threads where I have commented, sometimes I don't notice until days later that someone replied to me. To shill my own use of the API: I did an animated "rep
105.
▲
by
jasonthorsness
1y ago
If you are sweating this level of performance, are larger gains possible by switching to C, C++, Rust? How is Rust for micro-managing memory layouts?
106.
▲
by
jasonthorsness
1y ago
Complexity, state, and lack of thoughtfulness/design. When things are rushed they are lower quality and it seems nowadays almost everything is rushed. Also, time taken "pondering" outside of writing code and stuff can be robb
107.
▲
by
jasonthorsness
1y ago
I recently re-read Design of Everyday Things (I think written in the late 80s or early 90s) and noticed that despite all the criticism of interfaces, there was a default assumption that underneath the interface, if you used it correctly, ev
108.
▲
by
jasonthorsness
1y ago
For low-level small things tests can help. Go has good benchmarking built-in and you can use a tool that passes/fails based on statistically-significant regressions, benchstat ( https://pkg.go.dev/golang.org/x/
109.
▲
by
jasonthorsness
1y ago
Didn’t try Atom; just generated RSS based on the spec and examples that worked in the readers I tried
110.
▲
by
jasonthorsness
1y ago
subscribed to the str(c)spn thread for the eventual explanation of why the non-simd version seemed to give the wrong answer
111.
▲
by
jasonthorsness
1y ago
Compared to GPU programming the gains from SIMD are limited but it's a small-multiple boost and available pretty much everywhere. C# makes it easy to use through Vector classes. WASM SIMD still has a way to go but even with the current
112.
▲
by
jasonthorsness
1y ago
Uber is almost always cash in cities I’ve visited in India. Some take UPI too but it’s hard to use that as a foreigner.
113.
▲
by
jasonthorsness
1y ago
The content beyond the limit still collapses with a Show More; I think it works to still keep things short when they can be short but allow for more explanation for posts that need it. I like the change.
114.
▲
by
jasonthorsness
1y ago
At some level this just puts a huge burden on OpenAI. Because ChatGPT is so widely used, if something leaks everyone might put the blame predominantly on OpenAI rather than all the employees using it (disclaimer in case my employer is readi
115.
▲
by
jasonthorsness
1y ago
I recently enabled RSS for my own blog¹ and found it very frustrating getting the images/thumbs to display properly. The reason it was frustrating is the aggressive caching by the RSS readers. I had to debug it on a bunch of different
116.
▲
by
jasonthorsness
1y ago
It makes sense given "CTO Vladimir Fedorov notes that GitHub is constrained on capacity in its Virginia data center." and Azure has a decent setup for their AI support infrastructure and base virtual machines. But having gone thro
117.
▲
Spleen of a New Machine
(jasonthorsness.com)
1 points
by
jasonthorsness
1y ago
|
0 comments
118.
▲
by
jasonthorsness
1y ago
I keep hearing this however I was just at a career fair for University of Washington CSE new grads (representing my company) and it was packed with companies hiring. Obviously, this is just an anecdote, is there any good data showing the re
119.
▲
by
jasonthorsness
1y ago
I'm so much faster at typing than writing with pen or using a mobile keyboard - I have a hard time journaling because I get annoyed at the speed; but at the same time it seems weird sitting in front of a full computer setup to do it. S
120.
▲
by
jasonthorsness
1y ago
I don't think the GC in this version has the features required to enable a C# runtime on top of it yet: https://github.com/WebAssembly/gc/issues/77 I wonder what language this GC can actually be used for
More ›